The AM Forum

I ordered all the parts for John's (W2IMX) solid state screen modulator.
I have a bunch of rigs to try it on, but I have been thinking about the concept, plate voltage only goes as high as the power supply, there is no modulator and its power supply, and whatever the plate voltage is on the final tube is the most the rf deck will see, right?
So if it runs 2000 volts, a plate tuning cap can be 2500 volts and never arc over.
No mod iron, no modulator power supply, a VERY small circuit to drive the screens, wimpy rf deck components (voltage).

So that has me thinking about what tubes screen modulate the best.
I have 3 different ones to test the circuit on, 4d32, 4cx250, and 813's.

The 4cx250's are interesting (I have the 4x150a-7034 version, 250 watts of plate dissipation).
Small in size, cheap, easy on filament power.
Four tubes gives 1000 watts of dissipation, they are small, the rf deck parts could be reasonable at lower voltages.
I could use one to four tubes in a very compact rig.

I got the thing built and as a test I patched it into my 2x4x150a rig.
I just pulled the screen feed and inserted the unit in line, voltage in (variac) out to screen, and ground.
It takes 600 ohm line level audio in.

Picture of it almost done:
http://n2dts.smugmug.com/Ham-radio/i-DdB6qD9/A

It does not seem to work very well, at least not yet.

The design is from John, W2IMX, and he sounds very good using it, I sound like a frog.
There are plenty of adjustments on the circuit, plus all the transmitter adjustments, the downside of screen modulation I guess, everything has to be set just right...

Is there a link to the circuit diagram?  That may shed some light on the subject.

It was on here someplace, in another thread, but I have not been able to find it.
In the thread of a super clean rig or something about a year ago.

I am sure its just adjustments and voltages, but at the moment, it makes old military screen modulated stuff (carbon mic) sound hifi...

Brett, is this the thread and circuit you were trying to find?

Thread: http://amfone.net/Amforum/index.php?topic=35375.10

(Circuit is posted in reply number 11)

Circuit: http://amfone.net/Amforum/index.php?action=dlattach;topic=35375.0;attach=40659;image

Yes, that is it.

John sounds great on his.
I just patched it in and it works, did not let any smoke out, but I need to play with it.
The mod transformer is still in line, so I will need to bypass that, and my screen voltage setup may not go high enough, its variac'd up to about 650 volts, John says you need 1.5 times the voltage.
I was getting up to 250 volts on the output if I adjust it that way, 40% modulation and downward carrier shift, along with a very muffled sound.

Pair of 4x150a (same as 4cx250b's) running 1300 volts, was tuned up to put out 300 watts carrier with the plate modulation. I just shut the modulator off and put the circuit in the screen circuit, I turned the screen dropping resistor to zero and adjusted the voltage to 600 volts.

Nothing was getting overly warm, but I also seemed to have little to no screen current.

I am glad that is the one you were looking for.

I have often thought of using my stock of 4X150D / 7035 tubes, (4X250A with 26.5 volt filaments) to build a G2DAF linear, in a very small package.   But current thinking is the G2DAF design is not that clean, so maybe a small self-contained transmitter for 75 and 40 would do well with four of these tubes.  I would be interesting to hear what parameters you choose when you get it fine-tuned to your satisfaction.  Thanks in advance!

You might need a bit of mod on the control grid of the 4CX250 to reach 100% mod but be careful cause it is a bit fragile. 2 watt dissipation I believe.

Four?
I think one will do.
I run two at low voltage and low current for 300 watts carrier out and they only get warm to the touch, about 107F with little air flow.
I am sure two would do 500 watts carrier no problem with 1700 volts and 400 mills or less.
They call them 4x150's but they have 250 watts of plate dissipation each.

Last year I worked someone running a single 4cx250 with screen modulation using a transformer in the screen circuit only. Sounded very good, looked good on the scope.
It seems like everyone but me can lash up any old thing and get screen modulation to work and sound good....lucky for me I can get plate modulation to work ok...







I am glad that is the one you were looking for.

I have often thought of using my stock of 4X150D / 7035 tubes, (4X250A with 26.5 volt filaments) to build a G2DAF linear, in a very small package.   But current thinking is the G2DAF design is not that clean, so maybe a small self-contained transmitter for 75 and 40 would do well with four of these tubes.  I would be interesting to hear what parameters you choose when you get it fine-tuned to your satisfaction.  Thanks in advance!
[/quote]

It may just take some time to fine tune all the parameters to get it to modulate cleanly.  It will be interesting to see if grid modulation is needed in addition to screen modulation with the 4X150.

I suppose one or two tubes will work well, but considering the low efficiency and large dissipation for "efficiency" screen modulation, I thought it might be prudent to spread the heat across several bottles.  I want to make the cooling system as quiet and unobtrusive as possible.  The 4CX250 has ceramic seals, where the 7034/7035 version of the 4X150 is rated at 250 watts with a glass seal, not ceramic.  So while testing, and determining the limits of this tube's capability, I really would not want to melt a pair of glass seals.

I just finished rebuilding a UTC MLF (Multiple Loop Feedback) amplifier to use as a speech amplifier for the PP 250TH modded by PP 810s.  The final and modulator are fully restored, and I am just finishing the power supplies and control circuits.  I am putting in a bit more protection, fusing, etc, than was in it when my grandfather started using it in 1937.  Right now I am kicking myself whether to go with an authentic bridge of 866A rectifiers, or cheat and use silicon rectifiers.  Getting the right combination of filament transformers, chokes, and oil caps can be a challenge!  Once this is done, I will get that receiver built, then work on the screen mod rig.

Here is another design.

I haven't had a chance to prototype it yet so it might need some tweeking as well.

Oh, you are going to do screen modulation also.
Yes, four would likely do 300 watts.
Its good they do not take a lot of filament power, four is reasonable for 1000 watts of plate dissipation.

John runs the circuit into a pair of 4-400's beat broadcast pullouts he says, and does not modulate the grid.

I should research it, maybe grid leak is a no no and I want fixed bias?
Well, something to fool with this winter.
I have the remains of my 4x 813 amplifier I ran with the flex stuff (for a very short time)  that might make a good screen mod deck.

Sounds like you have plenty to keep you busy as well.

Winter time is great for building projects in the basement.  It always stays warm, and it is too cold to play outside with antennas.   

I built a 4-1000 screen modulated rig several years ago, with Frank's input.  I debated plate modulation, but did not want to rob the iron for the 304TL rig I am restoring, it is PP 304TLs modulated by a pair with a Stancor A3899 500 watt mod transformer. 

I thought for a long time about building a linear with those four 7035s, but never got started.  I am thinking that I could build a complete screen modulated transmitter, including vfo and power supply, in a new old stock HRO chassis and cabinet.  No, my dad never drilled hole one in that cabinet, it is still like new.  I was thinking of building it with the standard HRO PW dial for the VFO, and a panel layout similar to the HRO, meter location, etc.  Should be a conversation piece, and I already have the iron and caps for the power supply.  A  tube vfo and a buffer will likely be enough drive for the grids. 
 
I recall Frank stated that we should use fixed bias, no grid leak, in order to stabilize the desired bias voltage for screen grid modulation.  While he used a separate winding on the mod transformer for the control grid, it would be possible to capacitively couple the modulation to the control grid, using a mod choke to feed the bias.  A simple pot could be used to control the modulation level to the grid, and it could be adjusted separately from the DC grid bias.   Naturally I will test the solution before punching holes in the original HRO chassis and cabinet.

Let me know how you make out with your tests, and whether you find it necessary to modulate the control grid.   Hope the ideas above are helpful in your tests.

Maybe that is the problem then, the grid leak resistor.
I have a big wirewound pot plus a variac bias supply, so all I have to do is adjust things and figure out where to set things so they work. Seems like a lot of fine tuning and adjustments, so sort of a pain in the butt, but I am learning.
I have some other things going on so I may not get to it right away, I got a deal on a flex 1500 I want to play with, I had the 5000 and 3000 but not the 1500.
I am also in the thick of cleaning out the basement and garage, junk just accumulates and its time to get rid of it.

Time to break out the Bill Orr books and read up on screen modulation..

Doesn't grid leak rely on plate current to work?   And if we d op the screen voltage,  we will drop plate current,  causing our grid leak to drop as well....   Or am I off in left field?

I've got a single 250B here already set for 80-10 as an amp.   Was thinking about scream modulating it....   And if that works,  move it to the 4-1000 rig.

Maybe I should put up an antenna first.   

--Shane
KD6VXI

I think its grid drive driving the grid positive enough to draw current, you have grid current and voltage without any plate voltage or current at all.

I never looked up what you are supposed to do with the grid under screen modulation.
I know loading is important in most designs.
The way I tried it was to tune up the rig like normal, 300 watts of carrier, then insert the screen mod gizmo and adjust it for about 50 to 75 watts output.
At that point, I seem to get almost no screen current and the voltage is under 200 volts I think, grid drive was normal. I did have the mod iron in line, that has to hose things up some.

Here's the datasheet on the 4CX250 family.

Interesting read under the screen section.

Oh ya, 12W screen and 2 watt control grid dissipation.

Thin, 1/8 inch, plexiglass sheet can be found in the hardware store for replacement storm door glazing.  Smaller hardware stores like TruValue may offer custom cutting for small quantity/size.

But I want to modulate the screen, not the plexiglass!

Screen modulation can really sound amazing if done right.

First, grid leak bias is fine.  I've used it many times with screen modulation - no problems whatsoever.

The "problem" with screen modulation is that the load varies WIDELY over the audio cycle - from essentially infinity to some (not too high) finite value.

The modulator circuit shown isn't really that suitable for the job - not trying to be critical, but it's just the truth.  There's a resistive series element (the last transistor) in series with the non-linear load of the screen.  This will allow the capacitor across the transistor to charge more, with modulation, varying the DC screen voltage.  That's a problem.  Then the emitter follower transistor is there, but these types of transistors typically have a low beta, so the varying load will be reflected back to the resistance coupled stage, etc. etc. etc...

You would be MUCH better served using high voltage MOSFETs or IGBTs as the active elements.  A simple FET voltage amplifier (it can be resistance coupled) driving a FET source follower (with the source resistor going to a negative power supply to pull the screen negative for full modulation) would be just fine.

The great thing about MOSFETs is that they require NO gate current (other than that required to charge the gate capacitance) to turn them on, so looking into non-linear loads isn't a problem because the varying load is not reflected back to the previous amplifier stage.

Somewhere, I have a MOSFET screen modulator circuit.  If you're interested, I'll see if I can dig it out of the archives and post it.

I didn't look at the tube curves, but is 1200V really required for the screen?  Maybe it is - seems like a lot, but I don't know - didn't see the curves.  Looking at the 4-400A as an example, the screen voltage there is somewhere below 750 volts at full output.   So, you're going to be running a whole lot lower voltage than that to get linear modulation and leave plenty of headroom for positive peaks.  750V would be about the peak value.  If this is the case, you can use really inexpensive MOSFETs for the modulator.

Anyway, an idea!

Regards,  Steve

Steve,

I'm considering doing it with one of my amplifiers,  I'd like to see your rendition of mosfet style screen modulator.

Thanks,  in advance.

--Shane
KD6VXI

Steve
et al.

I've been following this thread with interest. I have just one comment/question:

Isn't the only purpose of the third stage of the screen modulator to provide a fixed positive voltage (left-to-right) across the large value capacitor, which subtracts from the output voltage voltage of the 2nd stage (emitter follower)?

I have been assuming that... given the large value of the capacitor... if the potentiometers are adjusted properly, and if the modulating signal is a normal voice waveform, then the capacitor's voltage will hold fairly steady (i.e. very little change in this voltage, and the changes will occur at very low audio frequencies) as the modulated screen voltage is varying above and below its average (carrier level) value... even though the screen current is a non linear function of the screen voltage.

I believe that the objective is to adjust things so that:

A) The screen voltage is correct at carrier level

B) The fixed voltage across the capacitor is high enough so that the screen voltage is driven sufficiently negative on downward swings of the emitter follower's (always positive) output voltage. This fixed voltage is critical. If this fixed voltage is too low, then one will not be able to achieve sufficiently deep negative modulation peaks at the output of the modulated RF stage. If this fixed voltage is too high, the negative modulation peaks at the output of the modulated RF stage will be distorted. I.e. negative modulation swings will be compressed/clipped.

C) The emitter follower stage is biased so that its output voltage (at carrier level) is sufficiently large to accommodate the required downward excursion of the screen voltage.

Note: the impedance of the audio frequency load on the emitter follower is less than the value of the emitter resistor. Therefore the value of the output voltage of the emitter follower, at carrier level, must be larger than the desired downward swing.

So I think the issue is whether the 2nd stage (emitter follower) has a low enough output impedance to drive the time-varying load presented by the screen of the RF stage that is being modulated.

Stu

After fooling around with my flex 1500 last nite I got the Bill Orr handbook out and read the bit on screen modulation and its circuits.

I got two things from the book, plate voltage on the tubes is better if its LOWER then normal.
And that loading is very important, the RF deck should be loaded more then when tuned up for plate modulation.

The Bill Orr handbook gave a good tube circuit and all the setup voltages and currents for an 813 tube.
One thing I am sure of is the circuit can work very well, I have heard it on the air when John uses it and it sounds great.

So I just need to set it up and then go through all the adjustments, starting off with the pots, a tone injected and an O scope on the output will show what is going on with the circuit, then adjustments of the transmitter.

According to Bill Orr, a pair of 813's will do 150 watts, one will do 75 watts, a good power level for 40 meters.
A 4x150 should be good for 100 watts, that could be a VERY small rig, 1000 volt power supply, small low voltage rf deck, and the screen modulator.

I recall talking with Frank,  KB3AHE, when I built my screen modulated 4-1000 rig.  He said this tube likes to have the control grid and screen grid modulated simultaneously in order to get the best linearity and highest modulation percentage.   I do not know if this is also true for the 813 or other tetrodes and pentodes.  It would be interesting to see what effect this method would have on the 4x150 series of tubes. 

I know you prefer to have as quiet a rig as possible, with little to no blower noise.  But a screen modulated 4x150 will be running lots more dissipation than it would in class C operation, probably more than in linear operation as well.  So the 4X150 may need a good bit more air flow to make 100 watts screen modulated. 

I look forward to hearing more about your results as you fine tune the rig.

Rick

The RF output stage will be operating in class C.

The reason the efficiency will be low (vs. plate modulation) is that (with screen modulation) the amplitude of the RF voltage from plate-to-ground will be less than half of the B+ most of the time.

For example, with screen modulation:

At carrier, the amplitude of the RF voltage from plate-to-ground must be adjusted (via the loading) to be less than (or equal to) 0.5 x the value of the B+ ... in order to provide enough headroom for 100% positive modulation peaks.

With plate modulation, the amplitude of the RF voltage from plate-to-ground, at carrier, can be adjusted (via the loading) to be 100% of the B+ ... because the B+ is being modulated to accommodate the changing (modulated) amplitude of the RF voltage from plate-to-ground.

This results in at least a factor of 2 reduction in RF output power (at carrier) with screen modulation (vs plate modulation)... given the same amount of electrical input power (at carrier) to the modulated RF stage.

Stu

You can put air on the tubes without it being a loud racket, I do it with a pair as modulators in AB1.
But I would run it at reduced power like I do the plate modulated setup.
No reason to push the tubes, I have rigs that will put out plenty of power without pushing things.

I do have 813's, 4x150's, and 4D32's.
I have an air system socket or two and blower for the 4x150 tubes and could go that way, or a single 813, or just patch the circuit into one or all of the rigs I have already built.

Made some tests, bypassed the mod transformer, adjusted the plate voltage to 1000 volts, put a tone into the screen modulator and adjusted things to get a clean sine wave out.
So far so good.
I made adjustments to the grid drive, the screen voltage, and adjusted the carrier level pot on the screen modulator, looking at the waveform on the scope.
I could get 100% negative modulation, and I could adjust things so I could get 150 watts pep out with any carrier power, 5 to 100 watts, and the waveform looked good.
I hooked the audio chain up and had a loud hum, and it sounded a bit stuffy, but not bad at all.
In keying and unkeying the rig, the screen current protection tripped, it had tripped a few times before as its set to 70 ma and the screen modulator draws about that with the voltage at 600 volts going in.
After it tripped and I reset it, something was wrong, the waveform peak no longer looked like a sine wave but a double humped camel. Only the positive part was distorted, I also had some erratic power output and cracklng in the audio, so I suppose one of the devices blew out.

I might have been able to adjust things so it sounded very good if I could get rid of the hum, but I had no chance to  try anything before the blowup.

All the adjustments seemed VERY critical, any change to anything had a lot of effect, loading, any voltage, grid drive, plate voltage, and all the adjustments on the screen modulator.
I do not really like that, but that is likely the way it is with screen modulation.

I put everything back to normal and adjusted the transmitter for plate modulation and it works fine.
 

 This is a helavu interesting thread to follow!

I was working Jerry, W1ZB today and he said to try an isolation transformer on the audio input to clean up the hum.
I will find out what blew out and fix that and try again.

R2 was open (1 meg trim pot), I replaced it with a fixed 260k resistor.

Got it working real well today.
I installed the isolation transformer, it did not do anything to correct the hum, so I put an oil filled cap on the B+ feed to the screen modulator and that fixed the hum.
I had hooked it up to the 3x 4D32 rig, and put a tone in, adjusted things so they looked right, and it sounded very good on voice!
40 watts carrier, worked N2LJO who said it sounded fine.
150 watts pep, I can not get as much peak power on the 4D32's, but it might need more adjustment.

1000 volts on the plate, about 100 ma for the 40 watt carrier.

Nothing on the screen modulator gets hot.

I need to try it on the 4x150 deck again, they seem to give a lot more modulation, I guess the screen has more control then it does in a 4D32.
100% modulation in both directions is not hard to do at all, without any modulation of the grid drive.

If you could run the RF deck at about 700 or 800 volts, you could power the screen modulator off the B+ and no other power supply would be needed.
I might try that, running the plate voltage down to 700 volts with the 4x150's and see what happens.

Four 4x150's in a very small RF deck with the screen modulator and a power supply could make for a very small 300 watt carrier rig.

Brett

Good!

A suggestion:

If there is hum on the B+ (with respect to the screen modulator's ground), then, by far, the largest effect of that hum on the output of the modulator will be from the base of the first transistor. All of the voltage gain is in the 1st stage, and the audio voltage on the base of the first stage's transistor is directly proportional to hum on the B+

To save space, and the avoid the use of the oil filled capacitor...

Add a 10k ohm resistor between R2 and the base of the first transistor (which, to the extent that it makes a difference, will be compensated by using a value of R2 that is 10k ohms less). Add a 10uF (or larger) capacitor between the junction of these two resistors and ground. The voltage from that junction to ground will be less than 100 volts (measure it before adding the capacitor), so you won't need a very high voltage capacitor.

This will significantly reduce the hum at the output of the modulator, without having to use the oil filled capacitor across the entire B+

Stu

Thanks Stu, I may try that.
I think the issue is that the rf control deck has a bunch of stuff close on the front panel, variac for the screen voltage and big wire wound pots to adjust the grid leak bias and screen dropping.
I think the variac's modulate the voltages through the wire wound pots.
A separate supply would not have that problem.
You want the screen to self modulate, so I can not put a cap at the output of the screen dropping pot if I want to use it for plate modulation.
The control deck also has a not used power supply (with variac) that was for the 3 diode neg cycle limiter, if its keyed, maybe I can use that instead.

I think I want about 600 to 700 volts to cover all the tubes I might use, at about 100 ma.
That is easy to do, and would likely fit on the screen modulator chassis itself, keyed by the plate circuit (step start).
 

Glad you qualified the above by including the "carbon mic".  The screen modulated command sets, when fed line level good quality audio can sound very good.  They do require a bit of tweaking to get them sounding right, very sensitive to final loading, like it on the heavy side.  My T22 modulated by a BC-456 stock dyno powered setup makes about 55W out on CW and about 20 on AM,feed it with line level audio through a decoupling cap into the carbon mic input, or using a military RM-12 remote control unit which allowed coupling field phone circuits into the carbon mic inuts of various sets, even has a VU meter for setting levels.

I swept the frequency response of the setup fed this way, and it looked clean from about 150 Hz to way over 10 kHz.  Feeding it into a dummy load and feeding it with music test audio is very listenable in a local good quality AM receiver.

Put the 2x 4x150 rf deck on and tuned everything up, and made all sorts of tests and adjustments.
I think I am a bit limited on my voltage feed to the modulator because it trips the screen current overload protection which trips at about 70ma.
The modulator itself takes more then that if I increase the voltage past about 550 volts.
But it was working very well at about 70 watts carrier out, 250 watts pep, about 1100 volts on the plates, about 200 ma.
Looked at the sdr and the distortion products were way down, it sounded good in the mod monitor, and I made a contact with Larry, W9SX 1000 miles away. He said it sounded good.

Due to the way things are connected, its going to be more of a pain to try it on the 813 rig.

I think I will need to do this if I want to use the power supply that is built into the rig.
If I put a 4 uf oil filled cap on the voltage input, the hum is low, its even lower with a 15 uf cap, but there is a trace.

While I am at it, I would like to put some combination of fixed resistors and a pot to fine tune things.
With the standard design, I somehow blew out the R2 trim pot.
Maybe a 10k pot plus some fixed resistors to center it around the total 260K ohms called for.

I also wonder what the effect would be of adding some resistance to the two 10K 50 watt resistors to cut
the current through the system a bit.
As it is, with about 500 volts going in its pulling 70 ma or more.

It really sounds very good to me, at 70 or 80 watts out I get plenty of modulation, to get more power, I would need to increase the voltage into the modulator.
500 is going in, and about 80 is coming out without modulation.
It wants 600 or 700 volts going in with 4x150's.
At 80 watts carrier, the tubes are running at 92F with gentle air flow, 100 watts should be easy to do.

For lower power tubes, you should be able to run the screen modulator off the plate voltage, up to about 1000 volts I suppose.

Brett

Putting the details aside for the moment:

How large does the positive-going screen voltage swing need to be... from carrier level to your desired percentage of peak positive modulation? Is 100V enough? If not, how much?

Stu

Stu,
I do not know, and it likely changes with each tube type.

A pair of 4x150's has 500 watts of plate dissipation, so that would be the best choice.
As it is, I have enough swing for about 80 watts of carrier with about 550 volts going in and 80 out.
I did not write those numbers down so they may be off a little.
1400 volts on the plate I think, and the grid bias also has a big impact, less grid drive equals less bias which equals more plate current and more power output.

I plan on doing some tests using the plate supply as a feed to the modulator as well as the plate of the tubes.
I can turn it down to 700 volts and see what happens, and work my way up to 1000 volts.
It looks like the circuit would work very well with 6146's with 800 or 900 volts on everything....

Brett

Okay.

I spent some time thinking about this circuit, and each of its stages in more detail.

I have to retract my earlier statement that the dominant contributor to the circuit's output hum is the power supply hum on the base of the 1st stage transistor. More on that below.

1st stage (common emitter)

A. Reducing the DC collector current

I believe that you could reduce the DC collector current drawn from the power supply by increasing the collector resistor from 10k ohms to 20k ohms. This will reduce the required DC collector current to half its present value. The power dissipated by the resistor, and by the transistor, will also be cut in half. Note: you will have to increase the value of R2 by a factor of about 2... because a reduction in the DC collector current by a factor of 2 must be accompanied by a reduction in the DC base current by a factor of 2. I.e. the transistor has a fixed current gain. You should adjust R2 to produce enough collector current to accommodate the desired positive voltage swing, and not much more. For example, if you need 120V of positive swing to modulate the screen voltage from carrier level to the desired positive peak modulation level, then you should adjust R2 to produce just a little more than 120V of DC voltage drop across the (new) 20k collector resistor. This corresponds to just a little more than 6mA of collector current. The reason that you can increase the value of the collector resistor, and therefore reduce the DC collector current, is that the load... looking into the base of the 2nd stage (the emitter follower)... is significantly higher than 20,000 ohms for all values of screen modulator output voltage. However, you should check to see that there is no distortion at the output of the 1st stage as a result of this change.

Note: The DC current gain of this transistor, under these (low collector current) operating conditions, is around 10.

http://product.ic114.com/PDF/T/TT2062.pdf

Therefore, with 6mA of DC collector current, the base current should be around 6ma/10 = 600uA.

With 600uA of base bias current, and a B+ of 500V, the required value of R2 is 0.883 megohms.

With the existing 10k ohm collector resistor, the collector current should be set to around 12mA, the base current should be around 1.2mA,  and the required value of R2 should be around 0.417 megohms. I suspect that the value you are now using for R2 (260k ohms) is resulting in much too much collector current in the 1st stage. I.e. 19mA instead of 12mA. Maybe even more, if the current gain of the transistor is larger than 10

B. Reducing the hum contributed by the 1st stage

If the power supply (B+) increases by 1% (i.e. the upward part of the peak-to-peak swing, at 120Hz, due to hum), then the base current flowing into the first transistor will increase by 1%. But (this is where I was mistaken in my earlier post), the voltage on the top of the 1st transistor's collector resistor will also increase by 1%. The effects on the voltage at the output of the 1st stage (i.e. at the collector of the first transistor, which is the bottom of the collector resistor), will partially cancel (because the resulting voltage changes are of opposite polarity)... but not completely cancel (because the first effect is smaller than the 2nd). To reduce the hum at the output of the 1st stage, both the collector B+, and the B+ driving R2 (which produces the DC base current) must be filtered. Just filtering the B+ that produces the DC base current will actually make the output hum a little worse!

2nd stage (emitter follower)

A. Reducing the DC collector current

In the general case of an emitter follower, the DC current flowing in the emitter must be large enough to accommodate the required downward swings of the emitter (output) voltage. The emitter current cannot be negative... so the DC emitter current has to be larger than: the required downward voltage swing / the audio frequency emitter impedance.

Note: In this application, the "downward swing" is the difference between the required screen voltage, at carrier, and the (negative) screen voltage required to produce 100% negative modulation of the RF tube. The DC voltage between the emitter and ground should be set equal to the value of this "negative swing".

For example, if the required screen voltage, at carrier, is 75V and if the screen voltage required to produce 100% negative modulation is -20V, then the "negative swing" is 95V. In this example, the base bias on the second transistor should be adjusted to produce 95V between the emitter and ground.

The audio frequency emitter impedance is the parallel combination of the emitter resistor and the impedance looking into the screen of the modulated RF tube. If this combined, parallel audio frequency impedance is at least 5000 ohms (for all values of modulated stage screen voltage below the carrier level), and if you need a 100V negative going screen voltage swing (from carrier level to 100% negative modulation level)... then you will need to have 20mA of DC emitter current (about 20mA of DC collector current) in the 2nd stage. Note, this 20mA of 2nd stage DC collector current includes the carrier level screen current drawn by the modulated RF tube.

You could probably reduce the DC collector current, somewhat, by increasing the value of the emitter resistor to 20k ohms... without any unfavorable effects on the behavior of the circuit.

I suspect that you have set the base bias current on the 2nd stage at too high a value...by using too low a value for R4... and that you, therefore, have too much DC collector current.

B. Reducing the hum contributed by the 2nd stage

The hum on the collector of the 2nd stage does not produce significant levels of hum on the output of the 2nd stage. However, you need to filter the B+ that drives the base bias resistor of the 2nd transistor.

3rd stage (DC offset voltage)

A. Reducing the DC collector current

The collector current of the 3rd stage must be equal to the screen current drawn by the modulated RF tube, at carrier. This current is set by 3rd stage adjustment pot. This current is part of the DC collector current drawn by the 2nd stage (i.e. it does not come directly from the power supply).


Bottom line:

If you change the collector resistor of the 1st stage to 20k ohms, the current drawn, at carrier, by the screen modulator should not be much more than: 5mA (1st stage) + 20mA (2nd stage). Be careful to adjust R2 properly... so that you don't have excessive DC collector current in the 1st stage. Likewise, be careful to adjust R4 properly, so that you don't have excessive second stage DC collector current.

With respect to hum: You need to filter to base bias voltage supplies of the first and second transistors. Each base requires very little bias current (a high value base bias resistor), and is easy to filter (as per my earlier post). You need to filter the collector voltage supply of the 1st stage, but not the 2nd stage.

Stu

 

Dashed home from work and tried using the same supply for the plate an screen modulator.
It works well but only puts out about 30 watts at 700 volts.
The 4x150's want a lot more plate voltage to make power...

Made more progress, with great results!
I had issues with the rig taking off if I ran things at higher voltages, plate or input to the modulator.
It worked fine at moderate levels, but when I cranked things up, the plate current would take off, and it was because the output of the modulator increased its voltage out to the screen.
As a test, I just stuck a 100k resistor across the output of the modulator to load it down some.

I can crank everything all the way up and nothing bad happens, and I can get lots of power out, 100 watts is easy, with all kinds of modulation and it sounds very good to me.

I also fixed the screen overload protection relay making the trip current adjustable.
This is turning into a very fun project!

Brett

Interesting...

I wonder if the 100k resistor made the difference, or whether the process of soldering it in fixed a cold solder joint or loose ground having to do with the screen RF bypass capacitor... or some other faulty connection.

Stu

I did nothing to the RF deck.
All I did was to add a pot across the screen current trip resistor and I added some filter caps to the voltage input on the modulator (electrolytic under the chassis).
I still had the issue with the carrier voltage level popping up an extra 100 volts or so, but only if I ran over 1200 volts on the plates of the 4x150a's or if I ran over 600 volts in to the modulator.
I then just stuck the resistor on thinking something is making the last stage increase its bias point.
I had tried various settings of the R4 and R6 pots and voltages on the plate and modulator input and it always did it.
I also tried different bias levels on the RF deck, different grid and plate tuning, but nothing stopped it.
I had thought it was something in the rf deck taking off, but the swr was low and nothing odd shows on the sdr rx. The 100,000 ohm resistor eliminated the problem.
I just picked 100K ohms as something to load/damp the modulator output..

Its only about 100 to 150 volts out to the screens without modulation.


The thing now works so well I am thinking three or four 4x150's with plenty of air would make a very nice 300 watt rig.
 

I'd be willing to bet the new resistor gave a path for neg screen current to go.

--Shane
KD6VXI

That is what I thought, as it only happened past a point, but once started stayed.
Carrier power went up 100 watts or so, and turning the voltage to the screen modulator OR the plate voltage down would stop it till next time.

I suppose a light value of resistance to work into will make the modulator cleaner since it will have a more stable load to look into.

Brett

You have mentioned that you have a screen current trip circuit.

Can you post a schematic of that circuit? Please indicate where the screen modulator feeds in, and please indicate where the 100k ohm resistor is attached.

I'm assuming that you have a resistor in series with the screen grid, and that you are measuring the voltage across that resistor (both sides of which are at the screen voltage level above ground).

I'm wondering whether, somehow, this screen current trip circuit is causing the screen voltage to lock up in the absence of the added 100k ohm load.

Stu

Stu,
The screen current overload relay is built into the rf control deck. Its a relay coil in series with the screen feed to the rf decks. Part of the screen current flows through the relay coil, part through a resistor that sets the trip point.
When excess current flows, the relay gets pulled in and connects the screen voltage to a resistor to ground which holds the relay closed. Another set of relay contacts supplies power to a light in the reset button that breaks the path to the resistor to ground.
Pushing the button resets the overload.
There is a small cap across the relay coil to stop chatter.
I had used a fixed resistor that set the overload to 70 ma, its a fast acting circuit, and to prevent the screen modulator from tripping the overload I wired a pot across the resistor so I can set the trip point from 70 ma to never trip.

The screen overload has nothing to do with the problem I was having.
The 100K resistor goes from the modulator output to ground.

In order to reduce hum, I put two 100 UF 450 volt caps in series across the input voltage with 100K resistors to balance the voltage.
I found that kills some key down power spikes out of the transmitter and increases the positive modulation a lot.

I really like how the modulator is working.

Brett

The important thing is that the modulator is now working to your satisfaction.

With the substantial series inductance of the trip circuit relay coil (through which part of the screen current passes), interacting with the large amount of capacitance from collector-to-emitter of the 3rd stage of the screen modulator (effectively in series with the relay coil)... one could get some strange effects (audio frequency oscillations).

I suspect that adding the 100k ohm resistor to ground, at the point between the screen modulator's output and the screen voltage input of the transmitter, was sufficient to damp out those oscillations.

Stu

That was what it was acting like, but it did not show.
The swr into the antenna was low, the spectrum display (100 Khz wide) showed no modulation, nor did the mod monitor.
It sure looked like some sort of DC bias point problem.

But anyway, I really like the way its working now, it sounds really good to me.

That gives another problem, what to hook it up to and where to put it.
It seems to work very well with the 4x150 deck, but that could use more plate voltage, and likely more air and real air system sockets to run at full power.
Too much stuff and not enough rack space.

I used the screen modulator on 40 today, plenty of long transmissions and it worked fine.

I did not like the series electrolytic capacitors I used to filter the B+ input.
I used two new 100 UF 450 volt 105C  caps in series with 100K resistors across each one and found one resistor was getting warm, the other not.
I measured the voltage across each cap, 8 volts across one, 540 volts across the other.
Swapped them out and had 100 volts across one and 450 across the other.
Took them all out and put the oil filled 1500 volt 15 uf cap back on.

Now to figure out what I am going to do with this thing...

Brett

I would guess that you had one of the two series capacitors in the wrong polarity. Either that, or you have a batch containing some very leaky ("new") capacitors.

Stu

Stu,
ebay specials...
On the positive side, the one that was way overvolt did not explode...

I eBay'd some real air system sockets for cheap, and have a collection of new 4x150's and an empty chassis so I might do a 3x 4x150a rig with a real blower (a nice small one on hand) and put up to 2000 volts on the plates and use the screen modulator on it.

The current RF deck does 300 watts carrier plate modulated and 150 watts screen modulated, with loctal sockets and a muffin fan with gentle air flow through the tubes, at just under 1400 volts max.
Nothing seems to get very hot as it is, but I do not know how hard I could push it.
As it is, 1400 volts, about 200ma in, 280 watts input (carrier) and 100 watts out, 180 watts plate dissipation out of 500 for the two tubes. 150 watts out would be about half power.

I suppose I could build it so it would be good for plate modulation as well, I have the parts, but it would be bigger....

 

I tried the screen modulator on the pair of 813 rig.
I tried various settings and could not get the audio to sound clean like it does on the 4x150 rig.
I noticed the 813's take a lot more voltage on the screens to get any power out.
I tried various settings of grid drive and bias, plate voltage, input voltage to the modulator and modulator settings.
I could get various power levels out, plenty of modulation, but I could not get it to sound real clean.

The 4d32 rig sounded ok, maybe not as clean as the 4x150's but close.
I wonder if there is any way to tell if one tube screen modulates better then another other then trying it...

Brett

I suggest that you repeat what you did when you were working with the 4x150 rig:

Turn everything on, apply sine wave audio, and then use a scope to observe the waveforms ... starting with the first stage of the screen modulator... and proceeding toward the screen bus of the modulated tubes... and finally to the envelope of your modulated RF signal (using an RF pickup).

With a dual trace scope, it is easy to see where distortiion and/or parasitic oscillations are being introduced.

Keep an eye out for low or high frequency audio parasitic oscillations... and also for RF on the screen or the modulator B+. Also keep an eye out for RF getting into the modulator via the microphone cable.

Stu

Stu,
It did not sound bad on the 813's, and more fine tuning would likely improve it, but its easy to get the 4x150 or 4d32 rigs to sound very good.
It was just a test, I would not run it with the screen modulator.

Playing around with fixing up the old 813 amp as a 4x 4x150 screen modulated rig.
Here is what I have so far:

http://n2dts.smugmug.com/Ham-radio/i-WDMNfJK/XL

It could be screen or plate modulated, but its mostly for the screen modulator.
The sockets will fit in the holes the 813's used to be in nicely.
Its really big, because it used to be an 813 amplifier and its being recycled.

The oscilloscope will show all !  What does the scope pattern look like with a predictable waveform such as a triangle wave.  Lacking a triangle, a sine wave will do, but it's much easier to see non-linearity with a triangle.

I'm not sure if you (or anyone) can really adjust a screen modulator (or any modulator at all, for that matter) - at least initially - and do testing, without using a scope and a predictable waveform.

If your modulator can cleanly modulate a 400hZ triangle wave, it has an excellent chance of handling just about anything else, at least up to 100% positive.  If you want to test it beyond 100% positive with a triangle, you can do that too.  If the modulator has a negative peak limiter, you're pretty much all set - just keep going..  If not, you could overmodulate in the negative direction for the test, but depending on the modulator, it may or may not be a valid test - depends if the operating point (or some other DC parameter) shifts.

No matter what, the predictable waveform test will reveal absolute evidence of what is going on, and you can trace back through the circuit and find out where the nonlinearities are.

Happy hunting!

Regards,  Steve

Yes, I have a scope, and an sdr, and a mod monitor!
I was not really interested in digging into the 813 rig since I would not be screen modulating that.

As far as the modulator goes, once its set up, and it mostly is if you set it like John says, you really have only two adjustments to make on it, the carrier level pot and the voltage input to the modulator.
You could use a fixed input voltage if its high enough, say just put in 700 or 800 volts, then you adjust the carrier level pot to give the carrier level you want.
It always seems to work best is you run the carrier level on the low side, the more carrier you dial in, the less audio you get on the carrier. That is partly offset by increasing the voltage input to the modulator, but only up to a point.

Example, on the 4x150 rig, I can max everything out, 1300 volts on the plates, 650 volts going into the modulator, and I can get 100% neg and 120% pos mod up to about 140 watts carrier.
To increase power I would need to increase the plate voltage and maybe the modulator input voltage I think.
If I turn the carrier pot up, I can get 200 watts to whatever power I want, but the modulation will be lower.

If I turn the carrier pot down, I can get 200% modulation at 50 watts carrier.

Its  subjective but its easy to get the modulator to sound very good to me  on the 4x150 rig and the 4d32 rig, while the 813 rig always had a trace of distortion.

That rig works fine being modulated by a pair of 4x150's in AB1 and can do 700 watts carrier easy, so its kind of pointless to try and screen modulate it and get 150 watts out.
I did notice that the screen modulator sounds even better then that rig to me.
It has always been my best rig, nice mod iron, nice separate power supplies for the rf and mod decks, plenty of clean audio in AB1, everything adjustable, yet the screen modulator going into the 4x150 rig sounds better to me.

I am having a LOT of fun with this thing...

Brett

What is most important is that you are having fun... and you can adjust everything so that it sounds good to you.

However, I disagree with respect to the adjustments that you need to make. This is particularly true because you are experimenting with different values of B+ on the modulator, and the current gains of the transistors you are using can vary quite a bit from lot to lot, device to device (within the same lot), and with device temperature.

You need to set R2 to obtain the correct DC operating point on the first transistor (as per my earlier post).

You need to adjust the biasing on the 2nd transistor (emitter follower) so that the DC level between the emitter and ground is higher than the screen voltage level at carrier... by an amount that is equal to minus the desired maximum negative value of screen voltage needed to produce 100% negative modulation.

Using a scope with a sine wave or a triangle wave audio input, as Steve suggested, is the best way to do this.

Not adjusting these correctly will limit the maximum percentage of modulation you can achieve, and also produce unnecessay distortion.

The scope will also reveal such things as RF getting into the modulator.

Stu

This is what I have so far:

http://n2dts.smugmug.com/Ham-radio/i-nt9cnzP/A

This circuit is amazing, I am having a lot of fun with it.
All my other rigs sound distorted after using the screen modulator, plus I was having fun waving my hand in front of the microphone and finding it modulated the transmitter over 100% both ways at about 1 Hz!
Yes, that is ONE hertz!
And no mod iron to blow up or worry about.

Got the grid circuit done on the 4x 4x150 deck and most of the filament wiring.

Gave the modulator a workout yesterday, on 40 meters all day with plenty of my old buzzard transmissions during which nothing much is said with plenty of talking.
I do not know how people put up with it, but anyway,

The circuit is working fantastic, no issues, so today I took the time to take care of a few things.
The meters on the front read output voltage and current.
Current is slight, the voltage is low also, but was up against the edge at 150 volts resting carrier, so I put in a meter to read 0 to 500 volts, not 0 to 150.
My pair of 4x150's runs right at 150 volts for 100 watts carrier.
I took care of that, then experimented with the 100k 2 watt resistor I have across the modulator output.
Without that resistor, the modulator seems unstable as far as the resting carrier (screen voltage) goes.
If the screen voltage is increased past some point, it increases about 50 volts and stays there.
I suspect electrons build up on the screen and do something to the last pass transistor (q3) and its bias point.
I tried a 220K and it still took off. I tried a 120k and it still took off.
I replaced the cap across the last transistor and it did the same thing.
The base of the last transistor is very sensitive, just putting a dmm on the wiper of the pot changes the output voltage quite a bit.

It works great with the 100K resistor, but I would like to know why it happens.
Maybe that pot (r6) is damaged, I ended up with a cheap brand, John says Bourns are the brand to get, made for the voltage. I ordered some of those.

I did not notice the carrier set point changing when I tried the modulator on the other RF decks, the 4d32 and the 813's.

I added an isolation transformer on the input of the screen modulator, 600 ohm to 600 ohm, it did not impact the low frequency at all.
I also added a phase reversal switch, easy since its a balanced input to the transformer.
Very interesting, I have a phase switch on the mic input to the audio stuff, and now on the input to the modulator.
The correct selection of both gives loads of positive peaks, 500 watts pep out of a 100 watt carrier.

What is the best way to balance the load of four tubes in a screen modulated setup?
In other rigs with more then one tube, I had resistors in the screen feed to each tube.

If I have four tubes in parallel, do I put resistors in the screens and the cathodes?
The most I have done is three 4d32's in one deck, they seem to work fine with some resistors to each screen feed, but I wonder how that would effect screen modulation.
I do not think I have any resistors in the current pair of 4x150 rf deck, and one tube runs a little hotter.

Brett

I would do the following:

With RF drive and B+ applied to all four tubes, with the pi network resonated*, and with the screen voltage set at carrier level: measure the average cathode current in each tube.

*Loading is critical for this purpose.

You can do this by placing a 1 ohm resistor between the cathode and ground of each tube, bypassed by a .1uF capacitor. It doesn't matter if all four resistors are 1 ohm... but they should all be the same value +/- 1%.

Pick the tubes to be reasonably matched, in terms of the cathode current, with the identical carrier level screen voltage.

Add resistance in series wth the screens of the three tubes with higher cathode currents... to obtain equal cathode currents.

Since all four tubes are biased into class C operation (even though the tubes may have somewhat different grid-to-cathode cutoff voltages, and somewhat different transconductances) equalizing the cathode currents, at carrier, should equalize the power output contributed by each tube, at carrier, and also the dissipation in each tube, at carrier.

Stu

Thanks Stu.
No way to have them auto balance?

Guess I should build in some test points.

Brett

In theory you could design an auto balancing circuit using op-amps to compare the cathode currents, and (for example) feedback controlled DC grid-to-cathode bias adjustments to each of three of the tubes.

In practice, making the feedback control system stable (vs oscillations at subsonic frequencies) would be a challenge.

"The difference between theory and practice is greater in practice than it is in theory" (Anon)

Stu

Got the grid circuit and filaments done:

http://n2dts.smugmug.com/Ham-radio/i-kCkgrd7/X2

Next is looking up what to do with the cathodes, they are separate from the filaments and I do not remember what I did on the other deck, ground the filiment on one side, center tap, or what.
I also have to do the rf choke to the plate supply, dc blocking caps to unload the voltage off the pie net, and go to all 4 tubes in some neat looking way.

Stu,
I was thinking more along the lines of resistors in the plate, and/or screen and/or cathodes of a value that would equalize the current through all four tubes somewhat.
Looking at old circuits, it looked like they never seemed to do much about it...
100 ohm resistors in the feed to each screen was about it, and I do not think a lot of resistance in the screen circuit is the best for a screen modulated rig...

In between playing with the SLRM receiver, I removed the current 4x150 deck and put the 3x4D32 deck in and tested it with the screen modulator.
That deck does about 50 to 60 watts carrier nicely, it can do more but the carrier starts down shifting under modulation.
Below about 60 watts, the carrier meter on the mod monitor shifts upwards under modulation, above that it starts shifting down. Any shift has to get excessive before you hear anything in the mod monitor.
The 2x 4x150 deck starts the downward carrier shift at about 150 watts at the voltage I can run it at.
Most of my plate modulated rigs downward shift the carrier a bit, the 32v does, the 4x150 modulated by 811's does quite a bit because of the shared power supply between the mod and rf deck.
The big rig does it little, not at all at below maximum ratings.


I have to be careful, as its easy to over modulate in both directions with the screen modulator, its not like a modulator that might sag on peaks some.

I took the bottom off the 2x 4x150 deck to study how I built that one, as it works very well.
It has a 100 watt wire wound resistor in each screen circuit to balance the load, so I will do that on the big rig, and its got .001 disks to ground at the socket.
Is four .001 disks going to be a problem? I will have four 4x150's so it all adds up I suppose.

Slow progress, I have a bunch of flat 20 watt wire wound resistors that mount at their ends, but could not come up with anything that would isolate them from ground, the voltage could get quite high, but the mounts take small screws that wont work with ceramic standoffs.
I tried drilling the holes larger but that is no good.

Winter fun with radio!

 

Brett

Each 0.001uF capacitor has an impedance of -j31.8k ohms at 5000Hz.

Four(4) capacitors in parallel will have an impedance of -j7.96k ohms at 5000Hz

But:

The input impedance... looking into the screens of four(4) tubes in parallel... will be 1/4 the impedance looking into the screen of a single tube.

So, if .001uF per tube doesn't roll off the highest audio frequencies that you care about, it shouldn't matter how many tubes (each with a .001uF screen bypass capacitor) you use in parallel.

Separately:

Since the impedance (at DC and audio frequencies) looking into four(4) tubes' screens (in parallel) is half the impedance looking into two(2) tubes' screens in parallel, you need to make sure that the 2nd screen modulator transistor stage (the emitter follower) is biased with enough current (at carrier) to deliver the larger DC screen current, and the larger peak-to-peak audio frequency screen current swing. The 10k ohm emitter resistor will be larger than the impedance looking into the screens of the four(4) parallel tubes... so the emitter follower's bias pot and the carrier-level screen voltage pot have to be set with the four tubes' screens attached to the modulator.

Stu

Thanks Stu.
The screens present a load on peaks, otherwise they are almost open circuit I think, so who knows what happens.
At some point I will have to set things up and measure response and distortion.
I was looking at it and thinking if it was a problem, I could use some 500 pf 5000 volt TX caps, or even 100 pf ones.
I do not like or want it to go out to 20 KHz, 8 KHz is more then enough.


I plan on running the screen modulator on the new RF deck, so I am not spending a lot of time adjusting things on other decks.
Hopefully the new deck will not be a flop like the push pull triode deck was last year.

I have the screen circuit to do, the meter/switch/lights, and the plate choke/blocking stuff, new front panel to paint, and new pots to put in the modulator.

Just finished up testing the 4x4x150 RF deck.
I had to work out a few bugs and mistakes, the 40 meter grid coil had too many turns on it and a power wiring mistake.
It did 500 watts carrier out, 1300 volts (most that rig has) at 500 ma cathode current, and with the 811 modulator giving all it could, nothing zorched.
After running it a bit, the tube temps were warm to the touch, easy to hold my finger on them.

More work to do, I need to set the neutralization, remove or reduce the resistors I have in the plates, they are running too warm, speed the fan up and label things.
Then I will put it in place of the 2x813 rf deck, run it at 1500 or more volts and screen modulate it.

Hooked the screen modulator up, likes to run about 175 to 200 watts carrier output at the voltage levels I have (plate and screen).
Sped up the fan some.
I put the audio generator into the rig and it looks clean down to about 40 Hz, then the input transformer looks like it hoses the waveform up.
It seems to modulate well beyond 20,000 Hz.

Tested ok on 40 and 80 meters, likes to run at 200 watts carrier, will do more but that generates a lot of heat.
1500 volts, 400 ma cathode current, 300 watts carrier 1500 watts pep.
1500 volts, 300 ma cathode current, 200 watts out 900 watts pep.

Sounds good to me and looks good on the scope.
40 meters seemed dead today.

Brett

The efficiency, at carrier, seems to be amazingly high... given the headroom required for (more than) 100% positive modulation.

1500Vdc and 300ma average cathode current implies less than 450 watts of electrical plate input power at carrier.

200W of RF output power, at carrier, implies that the efficiency (including the losses in the output tank circuit), at carrier, is greater than (200W/450W) x 100% = 44%

That's amazingly high for a real screen modulated stage... capable of more than 100% positive peak modulation... although it is theoretically possible with the tubes biased deep into class C operation.

Stu

Sru,
Not sure things are so efficient, 450 watts in for 200 out...
My meters have home made shunts so may not be exact.

I just looked at my notes, I have at one point 2000 volts and 500 ma for 1000 watts in an 350 watts carrier out, 1400 watts pep. Tube temps got up to 170F.

1500 volts and 300 ma for 200 watts out seems a lot better, lots of power wasted to get another 150 watts.
It seems to run more efficient at lower power levels.

The rig is flat from 20 Hz to 22 khz, and sounds very clean.
 

Brett

In a screen modulated class C RF amplifier:

If there is sufficient headroom* to modulate from carrier level up to 100% positive modulation, then the efficiency at 100% modulation will be twice the efficiency at carrier.

*Since the plate B+ voltage is not modulated, the loading must be adjusted so that, at carrier, the amplitude of the fundamental RF frequency component of the plate voltage swing is no more than half of the fixed plate B+. Otherwise, one will not be able to produce 100% positive modulation peaks. This implies that the plate power input, at carrier is, at least 2x the RF power output at carrier... and more like 2.5 to 3x in a real amplifier.

An efficiency of 45% at carrier implies an efficiency of 90% at 100% modulation peaks... and proportionately higher efficiency at modulation peaks that are higher than 100%

I would expect the efficiency at carrier to be lower than 45% in a real class C screen modulated transmitter... particularly if you can obtain somewhat higher than 100% positive modulation peaks.

Something between 35% and 40% would be less surprising.

Stu

http://n2dts.smugmug.com/Ham-radio/i-JZGmdCv/X2

I am working in integrating the new RF deck and modulator into the station, I put the modulator above the 813 rf deck with the 4x4x150 rf deck above that.
I need to run line level audio to it, make up some interconnect cables in such a way as I can run one or the other RF decks, plate or screen modulation.
Once I get it done, I can measure the power input to power output ratios and write things down.
I did test the distortion, but did not write things down, but I did notice it could be adjusted to give very low distortion levels per the sdr-iq, at 150 watts carrier and 90% modulation, the harmonics were way down.

I also noticed some difference between 80 and 40 meters, 80 seems to work better, with a positive carrier shift under heavy modulation at higher power levels, and on 40 it tends to shift the carrier downward past about 250 watts carrier and lots of audio.
At 100% modulation I get no carrier shift on any band.
It also tunes somewhat different on 80, loading up to max current works well, on 40 loading should go past peak and on its way down (overloaded) for the best peak power.
Tank circuit Q might have something to do with it, I have not tried different coils yet.

No matter how I set things (within reason) it always seems to sound very good.

Brett

The definition of "loading" has always struck me as being more likely to confuse than to be helpful... but we are stuck with it.

Even though I understand the details (i.e. the equations), I had to edit the paragraphs below several times... to get them right.

Increasing the "loading" (as loading is defined) implies a decrease in the RF load impedance looking into the tank circuit... i.e. from plate to ground. Increasing the "loading" (as loading is defined) also corresponds to a decrease in the capacitance of the "loading" capacitor in a pi network*.

In your experiments with your new rig... the reason you need to increase the "loading" (i.e. less "loading" capacitance, and less RF load impedance looking into the pi network, at resonance)... when you are tuning up the rig at carrier level... is because you need to ensure that the amplitude of the RF signal from plate-to-ground (at resonance) is less than half of the value of the plate B+ supply voltage.

If you reduce the "loading" from the above setting... producing more RF plate voltage swing at resonance (dip), and, therefore, more RF output at carrier... you won't have enough headroom for 100% positive modulation peaks. I.e. as the screen voltage is increased to try to obtain 100% positive modulation, both the average and RF components of the plate current will increase to twice their carrier-level values. As a result, with a fixed RF load impedance from plate-to-cathode, the plate-to-cathode voltage will swing down to zero for a portion of each RF cycle. This can result in "downward" shift of the average RF output power when you apply sine wave modulation, as well as distortion on positive modulation peaks.

This is the same reason why the "loading" of a linear amplifier should not be adjusted for maximum output at carrier level.

*Note: In a pi network, increasing the value of the "loading" capacitor... which is across the output of the pi network (i.e. less loading capacitive reactance)... results in a higher value of pi network input impedance (at resonance). This is because the impedance, looking into the input of a pi network, at resonance, is (to a good approximation): Z_in = Q² x R_load = [(2pi x f x L)/R_load]² x R_load, which is proportional to L² / R_load ... where R_load is the real part of the total impedance across the output of the pi network, and L is the inductance of the pi network inductor. More "loading" capacitance results in a lower value of R_load. This adds (in my opinion) to the confusion associated with the definition of loading.

Stu

Less is more! (loading)

The old 2x4x150 deck on 40 meters acted normal, and so does the new one, but the new one on 80 seems to work fine when peaked to maximum output power with the loading control for some reason.
That surprised me, but if I peak it, it modulates over 100%, has positive carrier shift, and sounds good!
That is at the 200 watt level.
Plate tuning cap 80% meshed, so I am likely a bit short in inductance, 4 tubes are low impedance I guess.

Many tests to follow, I need to try some different plug in coils and see how that changes things.
Grid drive does not seem to matter much, tuning is not hard, I just tune to max power, then adjust the loading control past peak power a bit and away we go.
The variac input voltage to the modulator is a bit of a dial a power output control within a good range if the loading is 'heavy'.
In all cases, it seems to sound good, it always seems to sound cleaner then the plate modulated rigs.

I suspect the 4x150/4cx250 tubes screen modulate very well, the screen seems to have a lot of control.
In the AB1 modulator, you can set them for 90 ma resting per tube with 350 volts on the screens, and run the plate voltage from 1000 to 2000 volts and the resting current does not change, and there is no need to adjust the bias. Not sure you could do that with any other tubes.

Brett

With the new rig on 80 meters... and the loading control adjusted for maximum RF power output, at carrier level.... what is the percentage of mesh of the loading capacitor (roughly)?

Stu

80 or 85%.
Not sure what value the cap is, two sections, both in use, maybe its a bc610 tuning cap.
150pf? per section, total of 300, so figure about 260 or 270 pf?

1500 to 1700 volts on the plates, 300 ma, 200 watts out, about 650 or 700 volts into the modulator, I did not note the readings of the screen volts and current (output) 200 or 300 volts, 40 ma?

I started out at 1500 volts, tried 2000, and settled on about 1750 as the sweet spot I think.

Brett

In class C operation... where the plate current is a sequence of pulses... each of which is significantly narrower than half the duration of an RF cycle:

The amplitude of the component of that plate current, at the fundamental RF frequency, is (approximately) 2x the average plate current.

If the average plate current (at carrier) is 300mA, then the amplitude of the component of the plate current, at the fundamental RF frequency, is (approximately) 600mA.

If one wants the plate voltage swing, at carrier, to be half of the 1500V B+ plate supply voltage... then the required RF load impedance (looking into the pi network) at the fundamental frequency is: 0.5 x 1500V / 0.6A = 1250 ohms.

If the Q of the pi network is 10, then the required load across the output of the pi network... consisting of the parallel combination of the loading capacitor and the antenna system is:

1250 ohms/(10² +1) -jX ohms = 12.4 ohms - jX ohms... where X is not critical.

If the Q of the pi network is 5, then the required load across the output of the pi network... consisting of the parallel combination of the loading capacitor and the antenna system is:

1250 ohms/(5² +1) -jY ohms = 48 ohms - jY ohms... where Y is not critical.

But

300pF (maximum) at 3.885MHz => -j136 ohms

With 50 ohms looking into the antenna system, even at the maximum value of 300pF, this loading capacitor would have very little effect on 80 meters.

I.e. -j136 ohms in parallel with 50 ohms = 44 ohms (resistive) in series with -j16 ohms (capacitive).

The effect of the series capacitive reactance (-j16 ohms) is to reduce the effective series inductance of the tank circuit.

Therefore:

Regardless of the setting of the loading capacitor (somewhere between 0pF and 300pF), the impact of that setting on the actual "loading" is going to be small at 3.885MHz.

If the Q of the pi-network is 10 with a 12.4 ohm load at the output, then even setting the loading capacitor fully meshed (300pF) will result in an impedance looking into the pi network that produces an RF plate voltage swing, at carrier, which is only about 12.4/44 x 50% = 14% of the plate B+ supply voltage (neglecting the effect of the -j16 ohms of series reactance).

If the Q of the pi-network is 5 with a 48 ohm load at the output, even setting the loading capacitor fully meshed (300pF) will result in an impedance looking into the pi network that produces an RF plate voltage swing, at carrier, which is about 48/44 x 50% = 55% of the plate B+ supply voltage (neglecting the effect of -j16 ohms of series reactance).

I was surprised that the setting of the loading capacitor that produces maximum RF output power, at carrier, is not fully meshed.... when operating at 3.885MHz with a loading capacitor whose maximum value is 300pF. But, talking into account the effect of the series capacitive reactance at the output of the pi network... that arises from the parallel loading capacitance... the optimal loading capacitance could be less than fully meshed. I.e, with this value of pi network physical inductor, and at this operating frequency...  there may be a value of loading capacitance beyond which the total pi network series reactance (the physical inductor's reactance minus the series capacitive reactance of the output load) may be decreasing faster than the effective pi network output load resistance is decreasing. Recall that the input impedance, at resonance, of the pi netwrork is proportional to L x L / R, where L is the effective series inductance of the pi network (taking into account the series capacitive reactance of the output load), and R is the series resistance of the output load.


Stu
 

Stu, the plate tune cap is 300 pf, the loading cap is a bit small on 80 meters I think at about 1700 pf.


I did some checks after I got everything in place:

700 volts into modulator:

  plate voltage   plate current   watts out   screen voltage  screen current
     1500                400 ma           200              170                  8 ma

     1750                 420 ma         210              170                  7 ma

     2000                 420 ma         250              170                  5 ma

800 volts into modulator:

      1750                 540 ma         320             200                   8 ma

      2000                 600 ma         450             210                   9 ma


That last run had things getting quite hot, 1200 watts in, 450 out, 750 watts of heat.

I think I will run the rig at 1750 volts, and 210 watts out.
If I tune the final for maximum power at carrier level, I get crazy modulation, 210 watts of carrier, positive carrier shift on the mod monitor and 1000 watts pep on the watt meter!

In place:

Brett

Your latest measured values (below) are consistent with what I would expect from theory.

At 1500V plate B+ supply voltage and 400mA of average plate current, the plate input power is 600W at carrier. The plate efficiency is 200W (RF output power) / 600W (electrical input power) x 100% = 33%. This is what I would expect for a class C amplifier, with the RF plate voltage swing equal to a little less than half of the plate B+ supply voltage.

The amplitude of the fundamental frequency component of the plate current is approximately 2 x 400mA = 800mA

The required RF input impedance of the pi network, at resonance (dip), to produce this swing, is about 750V / 0.8A = 937 ohms

If the Q of the loaded pi network is 10, then the load impedance at the output of the pi network has to be 937 ohms / [102 + 1] -jX ohms = 9.3 ohms -jX ohms, where X is not critical.

If the loading capacitor is fully meshed, i.e. 1700pF, then it has an impedance at 3.885MHz of -j24 ohms.

The parallel combination of the loading capacitor (-j24 ohms) and the antenna system input impedance (50 ohms) results in an impedance of: 9.4 ohms -j19.6 ohms.

This implies that, on 80m, (if the loaded Q is 10) the loading capacitor should be fully meshed to produce the correct RF load between plate and ground (i.e. the amplitude of the voltage swing, at carrier, is a little less than half of the B+)

Using the free application "PI-EL", I played around with various values of Q (as Q is defined in the PI-EL application).

At 3.885MHz, with an impedance (looking into the pi-network) of 937 ohms, and with the pi-network looking into an antenna system whose impedance is 50 ohms, and with a pi-network Q of 10... PI-EL computed the required tuning capacitance (which includes the output capacitance of the four tubes) as 365pF, and a required loading capacitance as 1362pF. The corresponding physical pi-network inductor is 5.44uH

With the Q increased to 12, PI-EL computed the required tuning capacitance as 434pF, and the required loading capacitance as 1700pF. The corresponding physical pi-network inductor is 4.63uH.

PI-EL appears to consistently calculate Q to be 20% higher (e.g. 12 v. 10) than I calculate Q to be (in the same circuit). This appears to be a result of PI-EL using a different definition of Q than I am using.

This is why you are adjusting the loading capacitor for maximum RF output, at carrier, on 80m. I.e. maximum RF output on 80m corresponds to a nearly fully meshed tuning capacitor, and a fully meshed loading capacitor... which (using the definition of "loading") means that, on 80m, the amplifier is still properly "over loaded", even with the loading capacitor fully meshed.


Stu

The loading cap is not quite all the way open, I have about 10% left, but depending on how I run the deck, I think it could be all the way open...

It sure does run very well as it is on 80, and maybe it would work as well on 40 with less coil.
I have a bunch of coils to try, and its easy to bend up more.

Here is how things are set for 80 meters:

http://n2dts.smugmug.com/Ham-radio/i-JZGmdCv/A

Last night I tried different coils and jumped between 80 and 40 meters to make up my tuning chart mounted to the front panel.
Very different results with various coils, and I think I need more coil or more capacitance on the plate tuning, its sort of at the edge.
I guess four tubes is a lot of current and not a lot of voltage so the tank circuit is a bit different from my usual stuff.

Without a tuning chart, I could see the need for a modulator bypass (tune) switch, to put full screen voltage on the tubes for tune up at maximum power, then switch in the modulator.

Another odd thing I noticed is that I can adjust things to give very different results.
Depending on how you tune things up, I can get lots of negative carrier shift, or lots of positive carrier shift, or no carrier shift at all.
Changing the phase of the audio can give lots of peak power with negative carrier shift, or less peak power with more positive carrier shift. Both sound good.
With plate modulation, carrier shift is mostly power supply drag down from a common supply, or a bias/drive problem. Not so with screen modulation.
I can adjust things so I get crazy amounts of peak power output, and normal operation has the peak power a bit over four times the carrier power, 250 watts carrier, 1100 watts PEP.
That is with no carrier shift, 1750 volts, 400 ma.

I think I will make a second modulator with improvements.
Instead of the input transformer, I will just use a 600 volt 1uf cap, that voltage rating should protect the audio equipment in the event of some failure.
I will also use a 1500 volt oil filled cap as the output cap, up from the 1000 volt mylar cap I have in there now, include a bypass/tune switch, and try and come up with a better and more robust system to replace the trim pots.
Like a 500K pot on a vernier drive with two 250K resistors on each side of it.

20 Hz, as shown through the receiver IF output:

http://n2dts.smugmug.com/Ham-radio/i-V4FTg9h/A

And 100% modulation pre fine tuning

Doing tests, at 250 watts carrier, 400 Hz, I could not get the distortion products (2nd audio harmonic) below 30 db down at 95% modulation.
The waveform looks very good, the rig sounds very good to me, so why the 2nd harmonic?

I think it can go quite low at lower power outputs, around 150 watts, but at 250 watts I tried adjustments to everything without much improvement.
Plenty of positive modulation and peak power.

Adjustments made:
Grid drive and bias levels, voltage input to the modulator, the modulator adjustment pots, plate tuning and loading, plate voltage.

Rig into dummy load, sdr-iq running sdr-radio, mod monitor and O scope on receiver working, audio generator into the modulator directly (has input transformer 600 to 600 ohms).

It turns out that human hearing cannot hear even order harmonics until reaching a large amount of them .... much more sensitive to odd order harmonics ... I don't know why

This has been a very interesting thread and has me considering the implecations ..... ??? ???

Past tests have shown that my plate modulated rigs were quite low in distortion, but they do not sound as clean as the screen modulator.

I really like the idea that I can run 20 Hz at 100% modulation into the transmitter and not worry about mod iron blowing, no odd stuff that happens in transformers.
In the future, I will document the distortion levels as shown on the sdr of various rf deck tubes, 4D32 and 813's.

In normal use, I can put plenty of audio on the carrier and the signal does not look wide at all on the sdr, it looks kind of narrow.
With the sine wave input, the carrier shifts way up and the peak power seems to go much higher then 4 times the carrier power.
 

Today, I hooked up the 4x100TH modulator deck to the 4x4x150 rf deck and did some tests.
At 700 watts carrier output, the RF deck runs cooler then at 250 watts screen modulated.

1800 volts at 500 ma input (900 watts) gives 650 watts out.
At 700 watts carrier the rf deck is running much cooler then with screen modulation.
I can fully modulate the carrier with the 100th mod deck at 2200 volts easy.
As a test, I took it up to 1000 watts carrier output, 1800 volts and 800 ma, that is about half the plate dissipation of the tubes.
My power supply is limited by the chokes, 500ma ccs ratings.

A pair of 4x150's would make a nice 500 watt (carrier) rig at low voltages, 1500 to 1800 volts.
 

  Brett,

   I wonder if you have one of those "Kill A Watt" power line meters that can measure actual power feeding a transmitter (volts X amps X power factor)?

http://www.homedepot.com/p/Unbranded-Kill-A-Watt-Electricity-Monitor-P4400/202196386

I have a 120 volt version here that I find quite handy. There are 230 volt versions, but I don't know how available or expensive they are.

Screen grid modulation is inherently inefficient, much the same way AM linear amplification is. In both cases the efficiency increases with modulation.

This thread has interested a lot of readers. Hams using the 4CX250B and the variants on AM are not many. You have been telling us for years how two of them as class Ab1 audio modulators work. Just vary the plate voltage....the idle current is constant. Neat!

So here is something to contemplate. Compare a 400 watt output plate modulated transmitter against a 400 watt output screen modulated transmitter. I am referring to carrier power with headroom for at least 100% upward modulation. For sure the screen modulated RF transmitter will run hotter at about 1/2 the efficiency. But then back-out the power draw from the modulator. I contend that from power cord AC line watts to coax RF output at 50 ohm watts, the two setups compare about the same in terms of overall efficiency. It takes 125 watts just to light up four 100TH's. So two power supplies, bleeder power, transformer losses, etc all add up. I'd be surprised if either setup resulted in much more then 20% efficiency overall.

Jim
Wd5JKO

I think, overall its a wash, but each system has its advantages.
Screen modulation, no mod iron, no mod iron limitations, no separate power supply and modulator, all that saves a lot of space, weight an parts. The screen modulated RF deck does not have to deal with high peak voltages so the parts only need to be good for the plate voltage.

The downside is you need a big tube to run any sort of power.
The four 4x150's loaf along at 250 to 300 watts of carrier in screen modulation.
They also loaf along (are hardly working at all) at 700 watts carrier in plate modulation.
Say 700 watts out, 900 to 1000 watts in, 300 or 400 dissipation in the tubes out of a total of 1000 watts.

Do you need to run that kind of power? I do not think you buy much improvement over 300 watts really.

I really like the 4x150 / 4cx250b type tubes. A pair modulated by a pair makes a nice legal limit TX that is running at about 1/2 power with plenty of audio power in AB1.
Run them easy and a fan on low speed cools them enough without any racket.

A pair can run an easy 200 watts screen modulated, plus, if you build it right, you can use both setups.

I bet for a 200 watt carrier, both setups draw the same amount of power from the AC lines, not that it matters really.
If you add the power and filaments of the modulator, it offsets the screen only setup.

I do not have 220 in the shack, but ran two 120 volt 20 amp feeds on different phazes from the breaker box, the rf can go on one, the modulator on the other.

If I limit things to 15 amps, that is 120 volts at 30 amps.
3600 watts total, I never trip breakers unless I do something really stupid.

On the plus side, the 4x150 type tubes are easy on filament power, 6 volts (not 6.3) at 2.6 amps each, I run them off 5 volt transformers as my line voltage is high.
15.6 watts per tube, 62.4 watts for 1000 watts of plate dissipation.

Another thing I like about the 4x150 tubes is they work at low voltages, from 1000 to 2000 volts is good, you coul likely run them at 750 or 800 volts ok. They take very little drive, four takes less drive then a pair of 813's or three 4D32's.

Amazing little tubes.



 

May be this is an interesting Schematic,   screengrid modulation and a Dorethy pa.
Anton

Sorry forgot the picture

50KW would be ok for 80 meters at night...

As a driver...

:-)

--Shane
KD6VXI

I received a chimney and got the 4x150 mod deck working again and use it on the 4x4x150 rf deck.
That is four modulated by two, it worked very well.
I think two by two would be a great setup, 1800 volts at 400 ma in, 720 watts input, about 600 watts out, with 600 watts of audio in AB1. About 120 watts plate dissipation out of the 500, run a slow fan.

2400 watts pep so you would need to turn the mic gain down....

With the right amount of air, four would do 350 to 400 watts screen modulated and sound very good.

I am working on version two, point to point wiring, higher voltage caps, and as a test, real variable resistors (pots) for adjustments.
A 1 meg pot would be too touchy to adjust and you do not want to be able to go to zero ohms, so I am trying a 100 or 250 K pot with fixed resistors to get in the ballpark and reduce the voltage and power on the pots.
The starting point for the first two pots are 260 k and 280 k, so a fixed resistor of 220K  with a 100k pot should put me in the ballpark as a start.
If the pots can not take the voltage, it will be back to trim pots mounted on a circuit board.

I am going to drop the input isolation transformer and just try a 100uf 450 volt cap.
I am also going to add a bypass switch so full screen voltage can go to the screens for tune up.
I will retain the phase reversal switch on the line level input.

It seems immune but I think I will add some RF bypassing.

Most of the mechanical assembly is done, just the meters, shunts and the  bypass and phase switches need to be done.

Second modulator so far:
http://n2dts.smugmug.com/Ham-radio/i-GCDfN4g/A

Brett

Wow!

Very nice work!

Happy New Year!!!

Stu

Update:

Finished and tested, it works very well.
I had to change one resistor in series with the carrier level set pot to give a bit more output, but that was it.
I just defaulted the other pots to Johns starting point (260 and 280 K ) and did not adjust them, the modulator works great with them set that way.
I ran 600 to 800 volts into the modulator and nothing bad happened, nothing got hot.
I tested the modulator and rig up to 300 watts carrier and 1500 watts pep and it works great.

This deck has much higher voltage rated parts and all the adjustments on the front panel and no input isolation transformer.
I think it goes down to DC.

 

Modulator number one redone.
It also works well.

Found out something interesting about the screen modulator today.
I was running things up in voltage and had the effect of the screen voltage going high and staying high that I had partly solved by putting a 100K resistor across the output.
Well, I tried various things and found the problem went away if I increased the grid bias voltage.
I just increased the grid leak resistance.
The 4x150 type tubes must generate a lot of reverse screen current or something.

Those tubes and the '250 type will make negative screen current and must have a well regulated supply. In linear amps, VR tubes are sometimes used as screen regulators on those, instead of just a dropping resistor. Much depends on loading, so it could be slightly different every time tuning is done unless the screen current meter is used for a loading adjustment.

I maybe missed the schematic of the modulator, is it the W2IMX one? - but a heavy swamping load on whatever you have, equal in current to the full screen current, might be useful on a single ended output stage.

Otherwise a totem pole type output stage as for an audio amplifier, like the attachment in some what, just as an example, ought able to sink or source plenty of current while holding a voltage, might be something to use to correct that and really regulate the screen voltage to the audio waveform.

Just going to throw some ideas out on the table as a thought experiment.

An advantage of that kind of totem pole amplifier stage is that the positive and negative sides should be able to have individually defined current limiting depending on resistance values, so the current can limited asymmetrically depending on if it is going into or out from the screen.

In audio amps it is used for the short circuit protection (or when dummys put a 2 Ohm load on the amp) and is always symmetrical. It works by placing a base emitter junction of a small transistor across an output stage emitter resistor, one on the positive side and one on the negative side. When the current through the resistor makes 1.2V the small transistor (Q3 in the circuit) turns on and shunts some of the drive current, limiting the stage to a certain current beyond which it can't go or it will clip.

In the schematic of the audio amp, Q3 and Q6 are the current limiting transistors They sense the voltage across R1 and R2. There are DC coupled circuits like this, and it need not be as elaborate as a hi-fi amp. This is one of those SWTPC "Tiger" class amps from the olden days. Nothing's in stone that a similar could not be run with a +/- supply, but with a sink for current any need for a (-) supply voltage should be smaller than the (+) side, right? I believe R15 and R22 set the output voltage operating point.

Others know more about audio amps than I, maybe someone else can comment on this wild idea.

This current limiting is never supposed to be reached in normal operation but it is a protection for the screen grids in case of incorrect loading which will burn up the screens of those tubes quickly.

What is meant by this asymmetric current limiting is that if you want to limit positive screen current to 50mA and so make R1 a 24 Ohm emitter resistor on the positive output transistor is appropriate, but you wish to limit negative screen current to 15mA, then (depending on the amplifier stage) then making R2 an 80 Ohm emitter resistance would be appropriate on the lower output transistor. - IF I understand this right.

I'm not saying redesign the thing or even do it, just thinking about the screens of 4X150/250/350 tubes and their unusual negative current issues and delicate nature.

Very interesting.
But part of the reason I like the circuit John designed is its simple with very few parts.
I am sure there are many circuits that could be made that work even better.

As it is, all I really needed to do was to increase the bias voltage if I was to run things at high voltages, and the 100K resistor on the output is likely a good thing as it equalizes the no load/load aspects of driving a screen which is not a fixed load.

And I know the 4x150 type tubes are not supposed to have rugged screens and grids, but I have never had one fail.
With the screen modulator in use, screen current is VERY low, and the voltage is about 200 volts.
I think the resting current is 10 or 20 ma for the 4 tubes.

I started off with a 0-500 ma meter which did not move, and I changed it to a 150ma meter and it still barely moves at 300 watts carrier output.
The grid is a different story, 15ma or even less on 4 tubes works ok if the grid leak is high enough.
I noticed that after I increased the bias, the rig sounded even cleaner and I have no problem running the plate voltage up to 2000 volts and the modulator input voltage to 900 volts.
That was about 450 watts carrier output.

Doing this with tubes at high power has the issue of running the cathode/filament voltage very high to get the negative swing needed on the screens.

What really amazes me about screen modulation is the amount of peak modulation you can get out of a rig running it.
No phase shift I guess, and no sort of power limitation, its easy to get 5x the carrier power out on peaks, say 1000 watts with a 200 watt carrier at 90% negative modulation.
On some rigs I get 2x the carrier power (ten tec Argonaut 5), 3 times (3x4d32 rig with shared supply), and 4x on the big rigs with separate modulator and RF supplies.

How much this matters on the far end I am not sure, but there is no lack of peak power output if you want it.

I think it breaks down like this on 40 meters:

50% run under 75% modulation.
25% run over 100% in both directions.
25% get it about right.

The modulator at 30 Hz, quad of 4x150's, 300 watts carrier, homebrew rx IF output:

Spectrum at 300 watts carrier, 90% modulation, 400 Hz

Looks to be about -35 dBc. That's pretty good!

I think its 25, 5db per division.
Looks good on the scope and sounds very clean tho.

Hi Brett -

Congrats on getting the screen modulator working! And a nice looking build.

Do you have a triangle you can put through?  See how low you can get it and still look good. Anything below 50 Hz is great.

-25dB 3rd is OK.  It's a good start, OM.  I'll bet you can do better by trying various loading and making sure all the voltages are stiff. Play around with drive levels too.  Look at grid drive and current, screen current and audio form.  Look closely at the audio power coming into the screen. Be sure it is staying true.

Determine what is causing the slight left bending on the peaks of the sinewave. (pictured)  Probably some kind of saturation somewhere. Push it to 120%+ positive modulation and it should be easier to see and troubleshoot.

As Jeff / NBC told me...  refine, refine, refine.   ;D

T

I think I got rid of the bending later on, voltage input to the modulator changed that I think, or maybe it was plate voltage.
I have a waveform program in my laptop I can use for a triangle.
I did adjust everything and about 28db was the best I could do at full power, at much lower power levels it gets to 42 db, (at about 100 watts). Might be the way the screen modulates the carrier for this type of tube.

The waveform gets funky looking at 20 Hz, nothing but caps in the modulator, so I wonder what is hosing that up.
Not that I talk at 20 Hz....

I'm seeing the carrier at -75 and the 2nd harmonic of the modulation at -107. That's -32 dBc. -25 dBc would not be very good.

I thought it was the separation between the tone (sideband) and the harmonic, not the carrier.

It can be whatever you want it to be. I stated -35 (actually -32) dBc, so it's relative to the carrier. Yours would be relative to the sideband or dBs or something like that. As long as you do comparisons with like values, it's all good.

Triangle wave solid state modulator, into quad of 4x150's 80 meters full power output:

Saw tooth, then square wave:

The qix/dx60 modulator into a pair of 4x150's, 70 watts carrier out on 40 meters:
Triangle and saw tooth.

I rebuilt the quad of 4x150's into a quad of 4-125's and gave it a try.
It seems to work very well, those tubes seem to screen modulate very cleanly with the W2IMX modulator.
2500 volts on the plates, about 200 ma plate current total, and about 150 watts out, 600+ watts pep, no carrier shift, tubes showing a dull orange color.
I also ran it at 200 watts carrier out and it worked well, orange plates.

I'm screen modulating at the exciter level (ARC-5's) and using my quad 4-125A amp in AB1 as you recall from the other thread... but of course it could be done this way too.

Do you know how much screen voltage swing you needed to modulate the '125's at 200W carrier? How far negative is the negative peak?

ok Brett ... tnx for reporting yer results ... 100 ohms in each screen lead is more parasite stopper than anything else ... do you have individual grid leak resistors  as well as overall variable res and protective bias ?  did you match your tubes ?

the individual variable bias in the linear amp (JOK) is a good idea but require more rf chokes and coupling caps, I think

Well yeah, four of each instead of one of each... the chokes don't have to handle any current, so they are very small and cheap (about the size of 1-watt resistors) and not a budget buster  ;D
Actually the four trimpots were the most expensive part, but well worth it for the ability to "match" any four tubes.

One single grid leak bias resistor (big wire wound pot in the control deck), just 4 tubes (the only 4) I had on hand, they all show even plate color.
I do not know if that is just lucky or what.

I spent a while using the deck today at 250 to 300 watts carrier output and it worked very well.
Orange plates at that level, long transmissions.
Put tone into it and it looked very good, a slight kink (at 40 Hz anyway) at about 95% negative modulation, the 40 Hz waveform looked very good otherwise.
800 volts input to the modulator, higher screen voltage then the 4x150 tubes, I am unsure of the swing on the grids, the resting screen voltage at 250 watts carrier is about 200 volts I think, and much more current then the 4x150's at about 40 ma for four tubes.
I removed the 100K resistor on the output to ground of the modulator and everything was still good.

Tried the rf deck on 80 meters as well, and it worked the same there, it sounds very clean and it seems very stable.
Nothing got hot, burned up, or arced over.



 

At 200 watts out:

I also notice that "kink" as the envelope approaches zero output (at any modulating frequency). Not sure where it's coming from but as distortion goes it's fairly minor...

Nice looking RF deck with the glowing 4-125A's  8)

I'd thought of making my tank coil from copper tubing but decided to try toroids instead. It's three T240-2 cores stacked and covered with Kapton tape, then wound with about 16 turns of #12 copper wire. It does not seem to get hot after ragchewing on 40m at 200 watts carrier (also a nice orange plate color).

The coils plug in with banana jacks for changing bands.
I guess you could do that with toroids as well somehow.

I have been searching for the source of very low hum when I turn the grid drive on.

If I just key the exciter, there is a hum in the receiver, and garbage around the carrier on the sdr display.
On transmitting, its very low on the carrier but its there. Not sure you would hear anything at the far end and it does not show on the mod monitor.
I tried all sorts of things, adding filiment bypass caps, grounding various points, adding a cap to the bias, grounding the center tap and running the filiments in series, nothing made any difference at all.
I also tried driving the grids with my little ten tec argonaut 5 and it did the same thing.

Where is this hum coming from?

Blew it up yesterday running plate modulation tests at 800 watts carrier output.
Fried something or the plate choke did not like something as the first one melted down to slag and the 2nd one cooked the first pie section real good.
Time for a redesign...

magnificent rf p0rn ... just an observation .... mounting the tubes so close together might have the plates running hotter than they would somewhat further apart due to intercepted infrared energy .... its a tradeoff between heating and vhf stability

It also precludes putting four 4-400's in as well.
That was not the plan, that chassis started out as a quad of 813's in grounded grid, kind of tall narrow tubes, then went to four little 4x150/4xc250b tubes, then went to four 4-125's.
It has turned into a real hack job, but a good experimental deck for little outlay in cash.
I had the tubes and sockets except for a few 4x150's and their sockets, everything else was on hand.
There is a lack of space in the center for all the stuff needed as well...

I just might build a pair or three 4-400 rf deck.
A pair would do 300 watts out with screen modulation and could be quite small, three would do 400 watts at lower voltages.

Something was not up to the job of 700 or 800 watts carrier and an atomic yea low and let loose.
Or the old chokes I am using are no good at that power level.
I have a B+W plate choke but its kind of tall for the space...I will see if it can be made to fit.


It may have been a bad winter, but I have had loads of fun playing with radio.

I built mine on a 17x17x3 aluminum chassis, the most expensive individual part of the whole amp project, and always intended it to use four 4-125A's due to their low cost, ready availability, and inexpensive sockets. It turns out that there is room to plug in a pair of 4-400's instead (although they'd be getting close to the plate choke), in either pair of sockets on the diagonal, and the filament transformers are also just right. But those bottles aren't cheap and then you need a bigger plate transformer to take advantage of the 800W dissipation vs. 500W... maybe the plate tuning cap spacing isn't enough now, etc. For some reason I'm thinking of Tim Allen grunting, "More Power"  :D

What kind of plate choke did you use? I bought mine, don't remember where or what brand, but it may have been from RFParts.com... anyway it's not showing any signs of "zorching" at 200W carrier.

I got it working again at 250/300 watts carrier out (screen modulation).
I put an old B+W plate choke in, its a bit tall, but it fit and it works.
I did not test it with plate modulation, and I might not...

The tubes are cooled, there is a fan under the chassis directly under the tubes blowing up into the chassis and out through the bases.

The choke I had in at first was out of an FT102 I think, maybe not up to the current of plate modulation, but it had held up with the four 4x150's at 1600 volts and 600 ma.

It went when I ran it at 2000 volts and 600 or 700 ma.

The 2nd choke was a 4 pie wound on a big ceramic base, but it was somewhat damaged.
The B+W choke seems fine so far.


Are you screen modulating your quad?
Plate spacing only comes into play with the voltage you run, not the power output.
If you put in a pair of 4-400's in place of the 4-125's and ran the same plate voltage, you would just be able to run more current.
Four 4-125's at 2500 volts can do 600 ma, and under screen mod run at 300 ma for 250 to 300 watts out.
750 watts in, 250 out, 500 plate dis. (really hifi sounding audio though)

Two 4-400's can do 2500 volts and 600 ma, unless they can do more current there is no gain, but they will not be glowing their plates any.
The 4-250's and 400,s like more plate voltage, 3 or 4 KV.

Just went back and read that you are using it as an amplifier.
What does that 5th tube do, screen voltage regulation or some sort of drive amp?

I like the way the thing works, and like the lower voltage the 4-125's run, 2500 volts in screen modulation and 2000 volts in plate modulation.

My pair of 813's can run 2000 volts and 400 ma. the 4-125's can do 2000 volts (or more) and 600 ma for 1200 watts input..

I may get a new chassis and rebuild it to actually be what it is, with a bigger space for more coil (its a bit low on 80 meters), more space between tubes and a lot less extra holes and so on.

I ordered a new chassis, bottom plate and two filament transformers (Hammond) to clean things up.
I also got a nice 1.5 amp plate choke off ebay.

I might do as you did and raise the tube sockets off the chassis a little but keep the big hole below each tube and have air blow up through them.

FWIW, any thought of using Coleman lantern globes??

Better than nothin?

klc

We've seen this before.  :D

http://amfone.net/Amforum/index.php?topic=18616.msg130398#msg130398


(http://amfone.net/Amforum/index.php?action=dlattach;topic=18616.0;attach=10351;image)

4-125's do not need chimneys, and I do not plan to run them so hard as to need them.
If in CCS service at rated output, it would likely be a good thing to have them.

And, yes, 4-125's are wimpy tubes, but I do not see the need for super high power on the ham bands.
300 to 400 watts carrier without going to the ICAS ratings is a good place to be.

It is a  4-65A, used as a shunt regulator for the screen supply. Cheaper than many other tubes that can handle high plate voltages and currents! You can't see it in the previous photo but there's a large 100 watt wirewound resistor mounted on the chassis, which takes the "economy" supply (plate trans CT) voltage and applies it to the plate of the 4-65A, which in turn feeds the screens of the quad 4-125A's through a metering resistor. (The grid of the 4-65A is biased via zener diodes to give the desired voltage output).

I don't think I've ever seen negative screen current, and usually about 4-6 ma positive at 420 volts screen, 3 kv plate.

That is a cool way to do it, with the 4-65.

I never had reverse screen current on anything before, but the solid state modulator looks mostly like an open circuit I guess, and when used on 4x150's can result in the screen voltage going up on its own and the screen current meter going backwards against its pin.
The only way to stop it was to remove the plate voltage or grid drive.
With brand new 4x150's you could do 2000 plate volts and enough screen voltage to give a 300 watt carrier out of 4 tubes, with a 100K resistor from the screens to ground.

I see no sign of it at any voltage/power output with the 4-125's.
I also see no sign of it with the dx60/qix tube modulator, that transmitter runs 1300 volts and 100 watts carrier out and shows no sign of problems. Maybe it would at higher plate voltages...
I would need to cap input the power supply to get more voltage.
I might try that as a test just to see what happens....

I suppose you mean it has limited plate dissipation .... well maybe shifting from class C screen mod to class F might help ,,, I've just been looking at Georgios work on class F in the transmitters section .... a boost of plate efficiency would definitely help reduce heat production and could only require one more variable cap as a minimum implementation

Yes, limited plate dissipation.
The deck runs well at 250 watts carrier which is plenty on 40 meters.
I should try running it at 2750 or 3000 volts and see how it works.

I also have other rigs that can run more then twice that power output, so I am good with it the way it is.
Has anyone tried the class F stuff?
How do you tune it, how do you qsy?

Tried 3000 volts, worked fine, more peak power I think.

Brett you are doing a fb job in refining this transmitter .... I am surprised that you don't have instability, esp at 3kV ... did you neutralize ? iffn not, what did you wind up with for grid circuit ?  a terminating resistance ?

Its semi neutralized in the regular way.
I made sure to ground the metal rings around the 4-125 bases with spring clips.
I should add some resistance across the grid tuned circuit to damp out any nasty stuff (no sign of that yet).

The thing is really working well as a screen modulated RF deck, and I am sure it would work fine with the plate modulator at least up to 2000 volts.
2000 volts and 600 ma would be CCS ratings, 2500 and 600 ma would be ICAS ratings I think.
2500 volts might be pushing the plate tuning cap a bit.

With screen modulation, it runs fine at 2500 or 3000 volts and about 300 ma cathode current and 250 watts output, 1KW pep.
Dull orange plate color.

I have a shorter 1.5 amp plate choke and a new chassis coming.

I do not have the tubes or sockets, but six tubes in a circle would be cool....

I have been switching things around a lot, screen modulating things, plate modulating things, using different modulators both screen and plate modulators.

The screen modulated stuff sounds much better to me and has no phase shift, and I notice the EQ needs to be changed between plate and screen modulation because the screen modulators are flat and the mod transformer changes the response some (I guess).
It likely does not mean anything, but I also like the peak power that is easy to get out of the screen modulated rigs.

So, I think a very nice rig would be a pair of 4-400's screen modulated.
Should do a nice 300 watts carrier and 1500 watts pep without working very hard.
I just need to get a pair of 4-400's.

Maybe three 4-400's would be good, or four 4-250's which might be cheaper.
As a test, I may be able to pull the 4-125's out of the 4x rig and get two 4-400's to fit (as a test).

One compact good sounding high power rig.

Got two 4-400's in the mail today and put them in the 4-4-125 deck.
Got 300 watts carrier out no problem at 3000 volts, not showing much color.
More 4-400's on the way.
Not sure if I should use two or three of them, two seemed to work well..

Put my new plate choke in to get more room, plugged the 4-400's in and tested it.
Peak power was down because I think the 4-400's want more plate voltage, at 3000 volts, they made about the same power as the four 4-125's but with little color showing.

Something also started smelling like toast (bread).
The tube sockets are too close together and to other things to run 4-400's for anything other then a test.
I also need to see if my power supply will be safe to run 3500 to 4000 volts, it was intended to be for the 813 rig with 2000 volts and 400 ma with headroom, not 4000 volts...

Looked at my power supplies and they have 3000 volt caps.
Guess I should not run them at 3500 volts any more!

I increased the air flow to the 4x150 pair and tested it at 200 watts out ok.

I ordered another filament transformer and I am going to scrap the quad of 4-125's and use the parts for a trio of 4-400's.
1200 watts of plate dis, can run 350 watts carrier easy at 3000 volts.

I suppose it would also make a nice amp for a flex or something.

I could also plate modulate it with the quad of 100th's at 3000 volts.

Its nice to have choices...

Variac on the primary gives lots of choices!

I have variacs on everything, bias, screen and plate.
I tore apart the quad of 4-125's which was a hack job and found a feed through insulator had cracked and arced without me knowing about it.
Likely when I plate modulated it.

I ordered a 4X17X17 chassis so I could reuse the front panel, and to get three 4-400's to fit will move the loading cap over and use those CV joint type shaft couplings.
4-400's are quite wide!

I am going to make it with enough space to cool well and have a decent layout.
It should do 300 watts carrier under screen modulation without pushing things at all, which is more then enough for me.
Who knows what it would do with plate modulation...


 

Started on the deck, progress is slow in summer, too many other things to do...

http://n2dts.smugmug.com/Ham-radio/i-bZMpBhD/A

Nice cabinets/chassis where do you buy these? :)

Hammond makes them, and maybe Bud, and Mouser or Allied sell tjhem, www.tubes and more carries some of the smaller ones I think.

Got the new setup done, three 4-400's with the screen modulator.
Does 300 watts carrier without pushing the tubes.
I could go up to 400 watts but that seems to exceed the power limits feeding the rack (120 volts 15 amps).
That would be 3000 volts at 400 ma on the tubes, 1200 watts just to the tubes, not counting bleeder power, filament power, screen power.

2500 volts and 400 ma seems to be the happy point, 300 watts carrier, dull color on the plates, 1200 watts pep.

At 300 watts out:

That's very nice to see!

Nice. Interesting watching you experiment and work your way through. Impressive to say the least in the day of "look what I bought" look what you made! congrats. Where do you hang out? I would like to hear it even if just reading the mail.
don

Very nice work.  One question though.   What are those glass things with light bulbs in them??  ;D
73  Nigel

I went through the rig and tuned things up on 80 and 40 meters and noted the settings on the chart on the front panel.
I noticed the rig seemed to work much better on 80 meters, more power out with less color, and WAY more peak power output.
80 was 300 watts carrier and almost 1500 watts pep, and 40 was 300 watts and 1200 watts pep.

So I bypassed the wire wound 10 ohm resistors in the cathode circuits (filament trans center tap) and the 100 ohm wire wound resistors in the screen circuits, plus lowered the inductance of the parasitic chokes in the screen circuit (at the tubes).
I also took 2 turns off the 40 meter grid coil to center the tuning cap in its range, and spaced the windings out.

A retest had 40 meters do 300 watts carrier easy and 1500 watts pep.
Looks like it will run ok at 400 watts carrier and something like 1800 watts pep with the tubes running dull red on the plates.

Very happy with the way this deck turned out.
Distortion is low, and the only thing that limits or changes the frequency response is the audio chain equipment.
I have that set on the narrow side, not much past 5000 Hz I think.


I operate mostly 40 meters on weekends, mostly in the winter.
Summer I am outside on the motorcycle, or working in the yard, or doing some other chore, but operate from time to time on rainy days.

In my twim 811A amp HB deal which I realize is a pipsqueak compared to yours I had trouble with the wirewound ten ohms in the cathode circuit. Sometimes depending on what I was driving it with I would find at the next use the amp would not work and would very quickly find out the 10 ohm wirewounds despite being 10W were popped. They never popped when in use . So after replacing them a few times I sat down and made a check list of what I was doing and why they had popped. I concluded there was no real good reason EXCEPT for the far off possibilty that the wirewound was introducing some HF level activity of an undesired nature into the circuit. I knew it was perhaps a long shot but could see any other reason for these resistor to pop in such a strange manner and they were never burnt, just open. Now perhaps they were made in the land of rice and the quality was questionable. i will give you that. However I bought several 2 watt carbons of the appropriate value and paralleled them to get at least the 10 watt rating and a similar ohm value. I have not had  problem of any kind since. It has now been awhile and I have done everything and more I did before with it so I am convinced my hunch was correct. As I suspect is yours.
don

Wire wounds have their issues, and likely are a bad idea in anything around RF circuits.
I just eliminated them and everything seems fine, all the tubes show even color...

What bypass did you use?
Isn't a cathode always bypassed with say 10nf for a short HF path to ground?

I think we can omit the 10ohm/10 watt "center tap" resistors and replace them with two diodes from cathode to ground. In that case there is no concern about wirewound issues.


And I love the combination of hamradio and motocycles....  :)

Bypassed as in shorted them out with wire.
Next time I have the thing out of the rack I will remove the resistors and put something like a 1 ohm 2 watt carbon in the cathodes and a 10 ohm 2 watt carbon in the screens.

Now thats STRAPPING

Not really.
It runs at about 300 watts out (dull red plates), could do more if the voltage was 3500 to 4000 volts maybe.
Interesting because it takes line level audio in and has no transformers in line, no mod transformer.



[/quote]

Now thats STRAPPING
[/quote]

Its coming up on radio season again and I was testing the equipment since its not been on for a while.
I hooked the QIX dx60 screen modulator circuit to the pair of 4x150's and gave it a tune up on 40 meters.
I can not find the diagram here, but its around someplace.
That design really works well, line input or D104, but I am pushing the tube (6de7) with screen current into a pair of 4x150 tubes at 200 to 250 watts out.
So I was thinking of changing the last Triode out with a 6080 triode section.
Or is there a better triode in a smaller package?
 

hi Brett .... had enough of hot wx ? .... the audio people have snatched up the 15W triodes with big $ .... maybe a triode connected tetrode ... in cathode follower service most anything will work fb

I like it hot, I have 3 motorcycles to ride.
A single section of a 6080 looks close, worth a try when radio season starts full time.

You might try the 6EW7.

Phil - AC0OB

That is the same as the 6DE7.

I tried making an adaptor to sub the 6DE7 with a 6080.
I got about .001% modulation out of that setup, the 1st stage of the 6DE7 has much more amplification then the 6080.

One nice thing about the 6080 is the 300 volt filament to cathode rating, along with the double plate dissipation rating.

I was running the modulator on 40 today, amazing that the little 6DE7 modulates a 200 watt carrier 100%.
800 watts PEP or more...
 
[/quote]

You might try the 6EW7.

Phil - AC0OB
[/quote]

    I found this chart on the net. Might be useful.

Jim
Wd5JKO

Nice chart!
The modulator as it is now has a 12ax7 mic preamp (d104) and line level amp, then into the 6de7.
The best I can come up with is to sub each section with a tube that is close to the same specs.
The 6080 may be close enough to work ok for the 2nd section, only a test will tell.
I like the 300 volt filament to cathode voltage rating and the plate wattage.

I found something close to the 1st section but forget what it was.

 

hi Brett ... the 6080 has a u of 2 which means you will need a LARGE V swing to drive it as a cathode follower .... seems like a good job for the 6de7

   Remember that cathode followers have a gain of just under 1. A good cathode follower for this application has high transconductance, and high emission capability. Having decent plate dissipation is useful also. The Mu, whether it is 2, 5, or 20 makes little difference except for biasing, the voltage gain is still just under 1. The Mu in the first stage is likely where we need some voltage gain, so low Mu there is not desirable.

   Brett, you might expand your list to individual tubes. Perhaps a triode connected sweep tube such as a 6DQ6 would do for the cathode follower, and something like a 12AT7 for the 1st tube (Mu = 70'ish). Using both 12AT7 sections together will double the transconductance. For Lower Mu on the first tube, use the 12AU7 (20'ish).

Jim
Wd5JKO

Sounds like fun, but how clean is a 6dq5/6dq6/etc as a cathode follower?

What would happen if I used two 6DE7's in parallel (last section)?

Yes, I could triode connect all sorts of tubes I suppose, but how clean would they be?
What about a 6B4, I have some of those....

I really like how the modulator works, I just want to increase the screen current capacity, the 6DE7 runs warm at 200 watts out.

gain can be approximated as:    gain = mu / (mu + 1) for a triode cathode follower so for a 6080 with a mu of 2 this results in an approximate gain of 0.67 which possibly is a considerable change from 0.9 to 1 .... this of course means that to get a 200 V swing out of the 6080 cathode follower you would have to drive it with a 300 V signal

The 6B4 looks like a good idea.
Separate filament transformer with a 1000 volt rating is better then a 200 or 300 volt filament to cathode rating.

The 6B4 has an amplifcation factor of 4.2, mid way between the 2 of the 6080 and the 6 of the 6DE7 2nd section.
Good for plenty of current.

12ax7 as mic and line amps, a 12au7 as the 3rd stage (just one triode) and a 6B4 as the cathode follower.

I think I will also build another quad of 4-125's to screen modulate.
They seem to screen modulate real well.
I will make it a no fan rf deck.

While the 6EW7 has the same pinout as the the 6DE7, the 6EW7 has a higher power rating, higher transconductance, and lower plate resistance because the plate areas are larger.

For either tube, you must not exceed the 200 volt Fil to Cathode differential unless you use a floating filament transformer.

When driving a single 6146 or 6DQ5 for example, the screen voltage needs to be set to about 60-70 volts quiescent for a 150 volt P-P signal on the screen.

When driving the Knight T-150 with parallel 6146's, you only need about 35 volts quiescent on the screens to get the same power out.

For the 4-125A, I would set the initial screen voltage to about 150V.

With those screen voltages, you might have to use a tetrode or pentode shunt modulator configuration if can't find a suitable triode for a cathode follower circuit.

What does your current schematic look like?  

Phil - AC0OB

I mis read, the 1st section is the same, the 2nd section has 3 more watts of dissipation?

I will have to try the 6EW7.

The design would be the basic DX60, WA1QIX design.

I like the idea of a big triode with a separate filament transformer.
 

I ordered a bunch, cheap on ebay...

Another idea: How about a 4-65 driving your 4-125's, as in a "stripped-down" Continental 317?   :)

Phil - AC0OB

Very cool, but I do not have a 4-65.
A bit overkill for 4-125's or even the 3X 4-400 rig.
Would be really cool though, a quad of 4-125's modulated by a 4-65.
I think a 6b4 would be better though, cleaner?

Fun stuff.

Phil, 

That Continental schematic has flown around in my head for a couple years.   I've been collecting parts to replicate a miniature version.

Single 4cx5ka.  Instead of tubes series mod on the screen,  pulse width modulated using hv mosfets.

I have tube,  socket,  3A CCS Dahl and the fil xformer and a choke,  for good measure.

Ought to be good for legal limit am.   

--Shane
KD6VXI

Got a 6EW7 in the mail today.
Its the biggest 9 pin tube I have ever seen!
Will not fit in a socket with a shield so I had to replace the socket.

Works great, I did not see any plates glowing at 200 watts carrier.

200 watts seems to be about the maximum power out of a pair of 4X150's at about 1800 volts.
I can increase the carrier power but the peak power does not seem to increase past about 800 watts.
No problem with the modulator though.

I am accumulating parts for the bigger modulator and the new quad of 4-125's.

I take a 4" aviation sheet metal shears and slice the base shield into 4 petals and then cut the rest of the metal away.

Some of the later RCA 6EW7's had the smaller 9-T6-1/2 Bulbs, while the earlier 6EW7's had the larger 9-T9 bulbs.

Glad to see the 6EW7's are working out for you.

Phil - AC0OB

You could probably use a 3-500Z "switcher" with the proper filtering for PWM.

Owing to the relative simplicity of the 807/4-65 (or equivalent) driver configuration for modulating the screen of the 4cx5ka, that's the way I would go.

Phil - AC0OB

Thanks for the input,  Phil.

I was thinking of mosfet switches because I have them from another scrapped project....   A series mod low power rig from here.

Not against using a tube,  just never done it.   Pdm is new to me.

I have a 4-125 here,  for a single 5k, that might be just the ticket.   The   use a single fet for the switch?

--Shane
KD6VXI

That is not going to have much in common with the Continental transmitter...

What is the screen voltage and current of the 4cx5000?
Maybe you could modulate it with the DX60 modulator (with a 6EW7 in it).
Be good for about 2kw carrier.

I myself wonder how far you can take the DX60 screen modulator.
It works fine at 200 watts out.
Not bad for a pair of 9 pin tubes as a speech amp and modulator, and it likely could do 300 watts, say a three 4x150's..

Looked it up, the 4cx5000 is good for 3500 watts carrier with screen modulation, with only 350 volts on the screen (7500 on the plate).

Might need to go bigger then a 6EW7 though...

The switch fets I have on hand are 1.5 kv.   Same ones as described in a homebrew series modulator elsewhere on the site.

Since I have them on hand,  that was my reasoning for use.   I could just use the series mod,  I'm sure.   But,  pdm / pwm has my attention at the moment.  With judicious use of clip leads.

My screen supply is a seperate 1.4 kv DC supply.

I'll have to check into the pdm switch tube designs.   I know the t be has to be floated......   Guess it's time to start a new thread.

--Shane
KD6VXI

I also have a 4PR1000 on hand.    Not

Started making holes for the BIG modulator:
(https://n2dts.smugmug.com/Ham-radio/i-n6PwTmt/0/L/PA020293-L.jpg)

https://n2dts.smugmug.com/Ham-radio/i-n6PwTmt/A

12AX7 mic and line preamps, 12AU7 (one triode), 6B4.

Separate filament transformer for the 6B4.

Waiting for a front panel, I seem to have used up most of my good parts building all sorts of stuff...
 

Had fun on 40 today, running the 2x 4x150 transmitter with the QIX/dx60 modulator at 200 watts carrier and the 3x 4-400 rig with the solid state modulator designed by W2IMX at 300 watts carrier, working W8GPA (Rupe) who was running an AF67 into a big amp, 2400 volts and 400 ma, 960 watts in giving 300 watts carrier out.
I was running 3000 volts at 300 ma, 900 watts in for 300 watts out, about the same.

The 6EW7 did fine in the DX60 modulator.
Also worked WB3FAU (Russ) and WA1HLR.

Some progress.
(https://n2dts.smugmug.com/Ham-radio/i-vPBDssS/0/L/PA060294-L.jpg)

Got it done today and tested.
(https://n2dts.smugmug.com/Ham-radio/i-g7GZ63k/0/L/PA100296-L.jpg)

(https://n2dts.smugmug.com/Ham-radio/i-fcKj6gQ/0/L/PA100298-L.jpg)

It seems to work fine, looks good on the scope with a sine wave input.
I had to add another zener diode to the string of three in the WA1QIX design as the negative modulation was only going to 70 or 80 percent.

Some fine tuning might be needed, it does not sound quite as clean as the normal QIX design.
Likely the operating point of the 6B4 is slightly off.

My wife knocked the modulator over while it was out of the rack and broke a 6B4 tube.
After I replaced it the modulator was all distorted, so I replaced all the tubes with tested good ones and the distortion went away.
I hooked it up to the 3x 4-400 rig and it works fine, and nothing gets hot, the 6B4 takes the current just fine, but the voltage swing out is not enough for full power.
I can only get 1000 watts peak out no matter how much voltage or audio I put into it.
Modulates the pair of 4x150's fine.

It runs best with 700 to 800 volts in, but it also seems to limit the negative modulation to 60%.
The negative peak limiter needs work or removal.
Removal would just be removing the three 12 volt zeners I suppose.

What does your "Present" screen modulated circuit look like and what are the voltages at each stage?

Does your System look something like this?

Here is the circuit:
http://amfone.net/Amforum/index.php?topic=26603.0

I just put one stage of a 12au7 in as the 1st stage of the 6de7 and a 6B4 in as the last stage (2nd stage of the 6de7).

That modulator modulates the screens of three 4-400 tubes. No tuner or amplifier.

What runs best on 700 or 800V, the modulator or the 4-400's?

You need to be clearer in your responses and we can't follow or help you until you clarify, so help us out here.

Since you don't seem to have circuit drawing or scanning capabilities, take a look at the following two pages of the pdf file.

Does this look like the present circuit you now have???

If not, tell me what is different, especially for page 2.

What are the voltages on the grids, cathodes, etc.

And you are running 100 mA of current (required for 3, 4-400 grids) through a 6B4 tube that is only capable of 60 mA?  

The 4-400's are running with 3000 volts on the plates.
The circuit is as you posted, 800 volts into the modulator.
I tried some experiments, added 10K in series with the 39k ohm to ground, no real change, added another two 12 volt zeners to the single one that is supposed to be the negative peak limiter, no change, only 60% negative, 100% positive, about 1000 watts pep with a 300 watt carrier.

I can not really measure voltages with it in use.

The 6B4 does not need to handle 60 ma, since the voltage is only about 150 volts with screen modulation, the current is very light. Might go higher on peaks, but average current is about 10ma I think.

The solid state modulator will modulate the 4-400's very well, well past 100% in both directions.
I just can not get the swing out of the 6B4.

if I understand you have 1/2 12au7 driving a 6B4 as a cathode follower which drives the screens of the pa

also if I understand correctly you have provided 800 volts to the 6B4 plate and the 12au7 thru a 100k res

check to see if 1/2 of the 800V is on the plate of the 12au7 when operating and no audio ...this allows max audio swing to be developed.... the 12au7 is a fairly low u tube .... if that voltage is low try a 12at7 .... these voltages are definitely large for all these tubes ... I would consider using something else

DTS,

It sounds as if the circuit is "flattopping" somewhere, or you may be running out of emission current with the 6B4.

In screen grid modulation, the quiescent (non-modulated DC) voltage on the screen grids of the 4-400's needs to be about 0.4 times 500 volts = 200 volts. The minimum supply current for the 3 screen grids should be 75 to 100 mA.

Also keep in mind that the P-P AC audio voltage on the Cathode of the 6B4 will only be about 0.65 X the  P-P voltage of the 12AU7's plate, since it is a cathode follower.

To find the problem, you need to take voltage measurements and scope those areas mentioned.

I will see if I can find a way to measure the voltages.
Its really only the negative modulation that is a problem, the 60% max value.
I was thinking it has something to do with the negative peak limiter built into the modulator but maybe not?

The modulator works ok on the 4x150 rig at about the same voltages, so maybe its a current problem like you guys said.

If I am thinking right, when the 6B4 draws a lot of current, the cathode goes more negative, which is the negative modulation voltage to the screen. When it draws little current, the cathode goes more positive.

I can always sub the 6B4 with a 6dq5 or a 6dq6, good for more current...

Like BFS said, you need to monitor the voltage at the plate of the 12AU7.

But how did it work right on the 4X150 rig if you didn't know what the operating  voltages were?

A cathode follower's cathode simply follows the grid voltage. It does not invert the waveform on the grid.

You set a quiescent DC operating voltage on the 6B4 grid pin 5, which the 250k resistor does. This causes a positive voltage to appear on the cathode, at the transformer center tap via pins 7 and 2.

You then couple a high voltage P-P audio signal to the grid through a DC blocking cap (0.1uF). The cathode voltage then bounces above and below the quiescent DC operating voltage, providing an audio signal to the screen grid.

The P-P audio on the grid is 1.35 X the P-P audio seen on the cathode. Stating it another way, the P-P audio seen on the cathode is 0.65 X the P-P audio seen on the grid.

What WA1QIX's, K4TAX, N1UVI, WA1HLR, W0CAB, KT4LJ, WC3K, and other circuit modifications try to do is make sure the modulated screen voltage goes down to zero volts or negative, so the modulation waveform goes down to M = 0% and - 99% (negative) modulation.

I would recommend you read and study this book before going any further:

https://www.ermag.com/index.cfm?v_link=product_detail&v_key=632

Phil - AC0OB

Well, its a working design that I use in another modulator (that works very well).
I replaced the standard 6de7 with a 6ew7 in that modulator and it still works great, so thinking I could just go a bit bigger to handle more voltage and current I would sub a 12au7 for the first part of the 6de7 as its very close to the same specs, and a 6B4 as the output tube.
It works fine but will not go past 60% negative with three 4-400 tubes.

As a test, I replaced the 6au6 plate resistor with a 220K to get more voltage swing with no real change.
Changing the voltage input to the modulator does not change anything but the resting carrier power (static screen voltage).  Its on a variac and can go 0 to 900 volts input.
I measured the bias supply at -130 volts and the preamp tubes run off 400 volts.

So what would be a good tube to use as a screen modulator in a cathode follower configuration?

811

you might also consider screen driving a tetrode sweep tube as a cathode follower ... tie the control grid to the cathode .. this forms a low u triode  that mayl require some drive current but not much with your high supply voltage... also consider driving both grids tied together as a hi u triode...

 great job for a cheap fet too

love it ... think outside of the box

I was going to mention a sweep tube as well.   Gobs of emission,  and lots of volts!

--Shane
KD6VXI

What about distortion though?
I have some 6dq5's and 6dq6's that I used as spares for the G76 when I had it, plus a boat load of 6080's, and all the usual hifi tubes, 6l6, 6550, kt88, etc.
Got some 811's as well.
What would be BEST though, and why?

Voltage and current ratings are light, I might have 700 or 800 volts on the plate, but I have 150 or 200 on the cathode, and while peak currents might equal what you would have in class C plate modulated values, average currents are very low.

BRETT,

Experimentation is at the heart of ham radio and I applaud your sense of wanting to experiment.

Sometimes throwing pieces of different circuits together may not always work.

It would be interesting to know what your voltages and waveforms were with the 4X150 circuit that allowed them to work.

Another approach may be to use an 813 as the CF in a Triode configuration.

Suggest you use the "characterization" circuit to determine what negative voltage on the grid that gives you about +200V or so on the cathode.

In addition, some modification of the biasing and clipping circuit is in order to attempt to make it stiffer and to work for the higher voltages.

Others may have better ideas for tubes and circuits.

Yes, this is the fun part for me.
There are some nice simple circuits in the original 'broadcast audio out of your dx60' thread.
No peak limiter, some do not even have adjustable power output (screen voltage).
I want to adjustable screen voltage setup, should be easy to do, plus the bias supply to get 100% negative modulation.
I need to make a breadboard setup to try different values and voltages, and maybe make the bias supply use a variac.

http://amfone.net/Amforum/index.php?action=dlattach;topic=30451.0;attach=30870;image

That N1UVI circuit looks very simple, just needs the adjustable screen voltage setup.
Then I need to pick a clean strapping tube to use as a cathode follower.
Fun with radio.

This one shouldn't be too difficult but the choice is yours.

Good luck.

As a test, I put a 6080 in place of the 6B4, both triodes.
About the same results as the 6B4.
More fun to follow.

Changed it to this circuit, but with a 6B4 as the output tube:

file:///home/chronos/u-ef22acca4e21d388f7e443bc0e09ba923be098dd/Downloads/DX-60%20Modulator%20Schematic%20NG%20For%20AMFONE.pdf

Works on the 4x150 rf deck, but the carrier control does not give enough positive voltage on the screens to get full power and its got some distortion.
Seems to modulate well though.

Brett, you posted a link that points to your local hard drive file structure.  Perhaps you could post the link from which you downloaded the file so we could review it.  Thanks!   73, Rick

Your link:
file:///home/chronos/u-ef22acca4e21d388f7e443bc0e09ba923be098dd/Downloads/DX-60%20Modulator%20Schematic%20NG%20For%20AMFONE.pdf

Or perhaps you could attach a copy of the actual file.

Yeah, Brett.  That link will only work on your machine.  How 'bout uploading it here?

Al

You mean something like this?

You keep changing things around and never taking measurements of any voltages or waveforms?

Have you thought about just trying to modulate QTY 1 4-400 and see what happens?   ???

Phil - AC0OB

I can not copy it here, its in the 23 post by DMOD at the bottom.


http://amfone.net/Amforum/index.php?topic=26603.25

Its a simple cathode follower, I did not include the negative cycle clipper/indicator.

Just fooling around the easy way, I need to make a test harness so I can have the modulators out of the rack for measurements.
Here is the transmitter (lower half) with an 811 modulator in it.
 
(https://n2dts.smugmug.com/Ham-radio/i-dZKZGmg/0/L/PB150127-L.jpg)

I did a lot of experimenting and could get it to sound very good at about 150 watts out, but not at 200 watts out.
It also had a lot more positive modulation then negative, 90% negative and 150 or 200% positive.

I did some tests and clean up of the QIX/dx60 modulator, tests show some problems and very strange results, with a sine wave tone into the modulator, it looks good till about 2000 Hz and above that the carrier level goes up, plate current goes up a lot, very odd.
The solid state modulator does NOT do that.

I like the modulator though, it sounds good, gives a good peak power output, and is a bit more stable in carrier power then the solid state modulator, and it handles reverse screen current better.
That is an issue with 4x150 tubes at high power.
I improved the negative feedback some, and added a bit of a load on the output and it sounds/works very well with voice.

I put aside the higher power 6B4 version and may fool around with series modulation using a mess of 6080's or some such.
Make up some sort of class A series test modulator.
Looking at it, the downside is the need for 2X the plate voltage on the RF tube, not a big deal at lower voltages.
Maybe the 3x 4D32 rf deck would work well at 1700 volts.
300 watts carrier at 850 volts and about 400ma should not be a problem.
How many 6080's would it take?
How much drive power?
Let the fun begin!

Show us your tube circuit and where the voltages are at each point when it took off? We could care less about the SS circuit since you're using tubes.


How can it sound good if it takes off and the audio is limited?


I thought you were using the 4-400 in some such configuration?


What about the 813 circuit? You said you had some 813s.


Well, we have attempted to help you, but all we have seen is 10 pages of vacillation, and no circuit diagrams, descriptions, or voltage measurements.

Phil - AC0OB

I am not really asking for help, just talking radio.
All the stuff is in use except the 6B4 modulator which does not work well enough yet.

The tube modulator I was talking about is the qix/dx60 modulator design.
It works fine and sounds good, does 100% modulation both ways.
I did not say anything 'takes off', if you put a sine wave into it, it looks good on the scope and the transmitter acts normal, but put a 3000 Hz sine wave (or higher) into it and it does odd things, it increases the resting screen voltage, the rf tube plate current goes up from 300 ma to 400 and 500 ma.

Design here:
http://amfone.net/Amforum/index.php?topic=26603.0

I am just posting results of various ways of screen modulating various RF decks.
I find it interesting I can modulate a pair of 4X150 tubes with a slightly modified DX60 modulator.
Two 9 pin tubes modulate a 200 watt carrier just fine.

Converted the experimental big tube modulator to the standard dx60/wa1qix design and ran it all day Sunday, worked fine at 200 watts output with the pair of 4X150's.

Thinking of building an RF deck with parts I have around that runs three 4x150's to be screen modulated.
Should run 300 watts carrier and modulate well.
I have all the parts except the vernier drive knobs I use.
They have gotten very expensive.

So I have 4 screen modulators ,all work well, two solid state modulators that are very clean and take no bias voltage, line level input, and two tube modulators of the dx60/qix design that take D104 or line input and have a negative peak limiter built in, all have adjustable power output and meters.
The dx60 modulator does odd things if you put a sine wave into it.
Looks fine at low and mid frequencies, but when anything over about 2000 hz starts shifting the carrier up (increases the screen voltage) but seems to be fine with voice. Odd.

The solid state modulators (design by John, WA2IMX ) is slightly thermally unstable, the power output goes up a bit as it warms up. Not much, about 20 watts maybe.

hi Brett ... I may have some vernier drives .... let me know

I looked at his design .... he is using a bootstrap biasing arrangement which is thermally unstable .... addition of two more resistors ( one from the base to output and one in the emitter leg to output ) would accomplish this...

That would be interesting if it does not impact the circuit otherwise.
Its only on the 4x150 decks that its an issue because I have to swamp the output to prevent negative screen current runaway. On 4-125/400's its not an issue as they need no swamping.

Got an idea about the values of resistance to try?

I did some playing around with the 3x 4-400 rf deck and the solid state modulator at 400 watts carrier out, the 4-400's modulate different from the 4X150 (4cx250b) tubes.

At 3000 volts on the plates of the 4-400's and 700 to 800 volts into the modulator, its hard to get the peak power above about 1300 watts no matter how I adjust things.
I even tried different coils in the pie net, different drive levels, different voltage levels on the 4-400 plates and into the modulator.

The 4x150's like to modulate a lot of peak power, a pair at 1700 to 1800 volts likes to do 200 watts carrier and at least 800 or 900 watts pep.
The screen in those tubes seems to have a LOT of control over the plate current/power output.

Time to play with the audio generator into the 4-400 rig and measure things.
At 400 watts carrier the tubes are showing slight color on the plates.

ok!  ;)

Bet you are a lot of fun at parties!

I may experiment at a later date with a big tube screen modulator, I tried the 6B4 and the 6080 and both worked, but I need to eliminate the negative cycle limiter and maybe increase the output voltage swing.

As a test, I put the one I just finished in the three 4-400 rig.
It modulates the rig fine but is not clean, sounds distorted.
It likely can not handle the current that three 4-400's need on peaks.

An 813 will not fit in that modulator, but I could test KT88's, 6DQ5's, 6146's, and other hifi tubes.
A redesign and I could try 811a's or a 4x150 and other larger tubes.
A 4x150 seems very interesting, very low grid drive needed, lots of voltage and current abilities, small package.
Wonder how triode connection would work....

I doubt anyone has ever tried to screen modulate a bunch of 4x150's with one as a cathode follower...
.

Actually, I am pretty boring since I only talk tubes!  ;D

Here is another schematic using a 6EW7 (Triode Unit No. 1) fed by a conventional 12AX7A speech amplifier stage, with a 6B4 audio driver, and a 6DQ5 triode cathode follower.



See the notes on the schematic.


NOTE: Updated with a 6EW7 intermediate amplification stage.


Phil - AC0OB

That looks VERY interesting, I will have to print that out.

Ran the 4-400 deck with the solid state screen modulator today at 400 watts carrier out, that gives around 1400 watts pep which is just about perfect if you want to be legal.
Tubes showing only slight color and it sounds very clean.
3000 volts, 320/340 ma roughly.

I tried this today on one of the screen modulators and it worked fine.
I do have a bunch of capacitance on the input anyway, but it did not bother the way the modulator worked, and is likely a good idea to do anyway.

The entire transmitter has hum on the carrier, 120 cycles I think and I am trying to figure out where its coming from.
Lots of filtering on the high voltage supply, the screen supply and the bias supply.

Started work on a quad of 4-125 tubes in pie net to screen modulate.

The power supply in the old 3x4D32 transmitter gets up to about 1700 volts so I will see how a bunch of 4-125's work at lower voltages.
Should do close to 200 watts out.
No fans or blowers so it will be quiet.

Progress so far:
(https://n2dts.smugmug.com/Ham-radio/i-KMVVDjb/0/L/PB190338-L.jpg)

Grid circuit done.
(https://n2dts.smugmug.com/Ham-radio/i-c2StVBc/0/L/PB210342-L.jpg)

Got it done and it works.
But the rig I installed it in is lacking in screen voltage to get much power out.
I will have to swap it with the 3x 4-400 rf deck and see how it sings at full voltage.

Put the new quad of 4-125 deck in the other transmitter and it does 150 watts carrier and 800 watts pep with no color on the plates, and it seems to sound good.
2500 volts on the plates, about 200 ma plate current.

I also tested it at 200 watts out carrier with color on the plates, works fine there also, but without any fans/blowers I would keep it at 150 watts.

Tunes nice once there is enough voltage on things.
Really seems to make a lot of positive peaks, a KW on a 200 watt carrier.

I should also try plate modulating it with the 100TH mod deck.
I could run the whole mess off one power supply at 2500 volts.
Would do a KW in that setup.

To sum it all up:
The solid state modulators designed by John, W2IMX work very well with a very wide range of tubes and power levels.
813's: not much power output, not real clean sounding in the mod monitor.
4D32: modulate ok, not a lot of power out, under 100 watts for three of them.
4x150a/2cx250b type tubes: work well, lots of power out at lower voltages, reverse screen current can be a problem.
4-125: modulate very well, not a huge amount of power output.
4-400: modulate very well, a good amount of power output, a pair will about do the legal limit.

The WA1QIX DX60 type modulator works quite well, even into a pair of 4x150's which are easy to drive tubes.

Once I got used to how things work, I find it easy to adjust things.
I have a mod monitor (shows carrier shift and the modulation) and a peak reading power meter.
I load the transmitter up past the point of maximum power output until the dip gets broad and I get no carrier shift on the mod monitor, and the peak power is 4X carrier.
You can adjust it on both sides of that point, less carrier and more peak power (carrier shifts up on modulation) or more carrier power and less peak power (carrier shifts down somewhat), say another 100 watts carrier and 3X power on peaks. It does not SOUND much different in the mod monitor.

You can change frequency quite a bit without having to re tune anything, but SWR does change things somewhat.
You may have to change any EQ of the audio as there is nothing that colors it in the screen modulators, 20 Hz passes just as well as 20,000 Hz.

Screen modulation seems to be sensitive to hum, all my rigs have some, I even shielded all the AC wires in the quad of 4-125's and still have some hum, there is nothing to stop it from passing through.
The loudest hum I have is 40 db below carrier and does not show on the mod monitor, can be seen on the scope and the SDR display.

You can ADD a screen modulator to almost any rig, and it makes power output adjustment very easy with just a pot.

Brett, these results show how much hard work you have put into this worthy approach ... nice going !

Its been a lot of fun.
It started with me working a few guys running good sounding screen modulation, and John, W2IMX with his simple screen modulator design.

I like this 4-125 deck, I hooked it up to the PPP 100TH modulator deck for a no fan/blower (silent) transmitter and it works nicely.
The 4-125 will work well from about 1500 volts to over 2500 volts.
So will the 100TH.

WA2SQQ sent me a very nice recording of most of my transmitters and all the screen modulated rigs have some hum.
Where does it come from?
Its not the audio, I can disconnect the audio from the modulator with no change.
Its not on the screen supply, I can add a lot of capacitance with no change, same with the grid bias.

Is it coming from filament modulation?

Since we have never seen a circuit how can we tell?  :D

Even a small amount of ac on the speech amplifier or the modulator's DC supply will show up as an amplified AC waveform somewhere in the chain.

Make sure each supply for the speech amp and modulator is decoupled by an RC filter.  

In addition, bias supplies should have good regulation and should be "stiff."

Insure you have a solid "star" pattern grounding system as well. I once found that daisy chain grounding can cause problems.

Phil - AC0OB

The diagrams for the modulator circuits are in this thread, but with no connection to the modulator, the hum is the same.
Its not IN the modulators.
I think, in plate modulated transmitters the mod iron inductance filters hum out.
My 32V uses the same audio drive system with not a trace of hum.
If I take the 4-400 deck and screen modulate it, it has hum.
If I plate modulate it, it does not.
80 mf in the power supply, choke input.

Everything runs to a single point ground system.
The 4x150 seems to have more hum then the 4-125 or 4-400 rf decks.

I would try and run the filaments off dc but that is a lot of current....
 
 

That may be a clue.

Recall that when looking into a screen node you will see a higher average resistance than if you were looking into a plate load.

For a simple example, let's say we're plate modulating a pair of 6146's at 600V@300mA so the modulator sees an average resistive load of 2k.

Now lets screen modulate those same tubes at 100V@7mA. The modulator sees an average resistive load of 15k, or a factor of 7.5 times the average resistance of a plate modulated system.

A rule of thumb is the higher the impedance of a circuit the more susceptible it is to EM interference.

Let's say you are modulating the screens of tube whatever with a cathode follower. The whole circuit from the cathode follower node to the screen needs to be a low resistance circuit.

Try connecting an RG-59 cable from the modulation point to the screen connection while avoiding any high current filament circuits via proper routing. And make sure both ends of the shield grounds are at the same potential (voltage).


How are you modulating the 4-400 screens? I know you keep saying that but most of us are graphical image/schematically driven.

Phil - AC0OB

The 4-400's are being modulated by the solid state modulator:
(https://n2dts.smugmug.com/Ham-radio/i-wxT9v85/A)

http://amfone.net/Amforum/index.php?action=dlattach;topic=35375.0;attach=40659;image

The last RF deck I built (quad of 4-125's) I shielded almost everything and it still has some hum.
Maybe shielded filament wires would help.
(https://n2dts.smugmug.com/Ham-radio/i-c2StVBc/A)

shielding for rf reasons means grounding at both ends ...

shielding for audio reasons usually means grounding at one end only because if you ground at both ends any ac currents flowing thru the chassis are now flowing in the shield and couple to the inner conductor ... in audio preamplifiers where signals are in millivolts, grounding is usually at a single point only for minimum hum

the problem is more prevalent with the screen modulator because of its extended low freq response compared to typical plate modulated rig which does not...

grounding can get tricky......  javascript:void(0);

As it is now, I ground things at one end only.
Coax gets grounded at both ends.
Maybe I will try shielded filament wires.
I thought about that but did not see a nice way to do it.
Maybe just use rg8x...

I know the original author of the SS modulator wanted the lowest component count to do the job, but if I were using it I would make the modifications shown in the pdf file below and here is why:

1. the transistors used are HV bipolars with relatively high input impedances, making them susceptible to noise and hum,

2. there is no RF filtering at any gain points, so I have placed caps to kill any RF,

3. none of the base pots that set base bias are referenced to ground, making them susceptible to noise and hum, so ground referencing resistors were added,

4. the original audio input capacitor is too large, hence modifying the input network to accept frequencies from 75Hz on up, above the 60Hz hum frequency,

5. source voltage filtering to kill any hum,

The selected transistors used are operating near the ragged edge of their Vceo specs, so this circuit would be better suited for HV fets or vacuum tubes, IMHO.

Phil - AC0OB

The problem is the hum is NOT in the modulators.
I also use tube modulators and they do the same thing.

Some of it might be the fan in the 4X150 rig, if I tap on the RF deck with no audio it modulates the carrier a LOT.

I just got done redoing the filaments in that deck, ran them in coax cable, no real difference in the hum.

Maybe the fan is inducing hum into the tubes, and/or the filament transformer is.
I have a pot on the fan to adjust the speed, if I turn it up high, 60 Hz comes up on the sdr display.

I suspect it has something to do with the way the rig is run in screen modulated service, loaded heavy and low (static) screen voltage.

The 4X150's need the fan...

How about kill the fan on the deck and use a oscillating desk fan,  floor fan,  something to remove the oscillator (fan) from the equation.

Sounds like microphonics.   Maybe a DC fan if this works?

I had good luck with 43 beads at the fan,  multiple wraps,  to keep rf out of the DC fans on my 16 pill mobile legal limit amp.


--Shane
KD6VXI

As a test, I disconnected the filament transformer on the RF deck and used the one in the control deck, plus I moved the fan from inside the chassis to the outside.
These are forced air tubes and need airflow though them, so a fan/blower has to be someplace although I could route air to the deck through a hose.

No major change in the hum levels which are 120 Hz mostly.

Changing the plate voltage does not change it.
Changing the screen voltage into or out of the modulator does not change it.
Changing the drive level does not change it.
Changing the bias or grid leak does not change it.

Can hum in the exciter pass through the transmitter?


Power supply has over 100 mf and is a choke input full wave center tap design.

If I had that tube I would definitely go for it.

Phil - AC0OB

I did some experiments last week, changing power supplies around.
I discovered I had changed the high voltage power supply to cap input (small cap) to get more voltage out of it.
I switched power supplies and the hum was much lower so I moved transformers around to give a higher voltage, and changed the power supply to small cap, choke, big cap, choke, big cap, hum very low.
I can not get 3000 volts under load out with full wave center tapped choke input with the transformers I have.
I am not fond of bridge supplies so I have a cap input setup.

Unlike plate modulation, you can use high voltage as the peak voltage the rf deck see's is the plate voltage.
4-400's work well at 3000 volts, they might work even better at 3500 or even 4000 volts.
They want a choke input power supply to have very low hum levels though.

The 4x150 rig uses all the stuff from the 3x4D32 rf deck and 811a modulator, designed for 1200 volts output, but shoved up to 1800 volts for the screen modulated 4x150's.

If you build something to screen modulate, make the power supply choke input and good for the highest voltage a tube will take.

I have been down the hum road when repowering old military sets (and I have built some screen modulators as well) The hum in all likelyhood has absolutely nothing to due with screen modulation. For me, my worst source of hum came when building choke filtered HV supplies with solid state rectifiers. I read an explanation on line by a former designer of why he no longer used chokes in his Hv power supplies for ham equipment. I removed mine. I put a soft start resistor in the circuit at the beginning of the rectifiers. Just a few ohms , hi wattage. I removed the bridge and built a full wave rectifier using multi diodes properly by passed by 01s. I also used multi caps series with proper resistors and I went to a voltage divider supply. Is it perfect? NO but it is so much better I don't get the "hum reports". I used a PS design for my linear from the old SSB ARRL book as I had found it to be a very quiet supply. Using that as an example I rebuilt my mil radio PSU to match best I could and was rewarded with a workable radio. I also discovered I am on a dirty mains supply. Somedays are wonderful. Some aren't. Not all noise is generated in your own home. I have not had any issues with hum in my screen mod rigs and I have 3 of them , Two homebrew. I suspect had I used old rectifier tubes (Like a pair of 80s) and the old diagrams I would have had no or little trouble. PSUs , gotta be the number one source in most cases. It is my belief now that solid state is too abrupt and the design needs to be a bit different for that. Tubes were if you will allow me more graceful

don

this continues to be an interesting thread ....

you might evaluate for your capacitor input supplies using a series resistor, say 50 Ohms, and a second capacitor to form a pi filter .... this really reduces the power supply ripple .... in my case from 20Vpp to .5Vpp in one application...

your o-scope will let you know .... its a one-eyed snitch

   It's worth mentioning that a 3KV supply if looked upon with a scope could cause carnage to the scope, probe, and OM unless the probe is a 1000X type. Of the 1000X types, there are the ones for a DVM that are just for DC, and are not AC compensated. These will have a serious rolloff at 1 Khz. Then there is the more expensive type that are AC compensated, and good to perhaps 50 Mhz.  There is also a line of 100X scope probes that are much less expensive. These are often good to 1.5 to 2 KV DC, and have a derating curve for AC as the frequency goes up.

Jim
Wd5JKO

I tried this with the Cheyenne, I tried a 10 ohm resistor, then a 100 ohm resistor with a cap before and after it, that is, out of drake supply, into resistor, out of resistor with a 24 uf cap, then to the radio.
I needed to drop some voltage as well,  the 100 ohm got me down to 700 volts.
No real improvement with the hum.

Very interesting Don.
I will have to experiment with some of these ideas.

On my current setup with the 4-400's the hum is very low, but it took a cap/choke, cap, choke big cap setup.

Started gathering parts for another RF control deck.
I have too many RF decks (5) and modulators (7) and some power supplies (3) but only 2 control decks.
The control decks do bias (fixed variac and adjustable grid leak), screen voltage (variac 0-1000 volts plus a dropping resistor (pot)), screen overload protection, keying and metering.

I have the 2x 4x150 rf deck hooked up with my 811 modulator with the only good mod transformer I have, running it all at 1500 volts (350 watts carrier).
I would like to swap out the 4x150 deck with the 4D32 deck and screen modulate the 4x150 deck at 1800 to 2000 volts with the QIX screen modulator (200 to 250 watts out).
I like the way that setup works.
 

Hooked up the 4x 4-125 rf deck to the solid state screen modulator and gave it a workout today.
3000 volts, about 150ma I think, 200 watts out, 1000 watts pep,slight dull red color showing on the plates.

I also changed the power supply in the pair of 4cx150a transmitter to bridge choke input setup and tested the rig at 1500 volts. 811a modulators with some bias, 300 ma plate current, 350 watts carrier and 1500 watts pep with the variac at 68% of maximum.
450 watts in on the RF deck (roughly) , 350 watts out, 50 watts dissipation per 4cx150 out of 250, or 100 watts out of 500 allowed for the pair.
I am tempted to try the screen modulators on the pair at 2000 volts on the plates...  

I did try the DX60/qix modulator and the solid state modulator on the 4x150 pair with the new power supply setup at 2000 volts and get VERY good results, sounds very good to me and runs 200 watts carrier and up to about 1000 watts pep.

The solid state (W2IMX) modulator also works fine, but I find it interesting to run a 200 watt carrier modulated by two miniature 9 pin tubes.

I wonder if the dx60 modulator could drive three 4x150's at 300 watts but don't plan to build a three tube deck to test it...

So, don't have a mod transformer but want good sound?
Build the dx60/QIX modulator and modulate a 4x150/4cx250 for 100 watts out, or two tubes for 200 watts out, very low drive, line level audio or d104 in.
 

What was your quiescent (resting) DC voltage and P-P audio voltage to the screens of the 4X150's?

180 volts is the resting output voltage of the modulator at maximum power output.
That is 200 watts carrier and about 1000 watts pep, 2000 plate volts and 300 ma on the 4x150's.

Not sure what the voltage swing is at 100% modulation.
The modulator has no problem doing 150% positive and has a negative limiter built in.

Average current is light, the modulator I have in line now does not have a current meter in it but from memory the other modulator was doing something like 5 or 10 ma.
About 750 volts into the modulator.

The 4x150's are working hard at roughly 600 watts input (2000 volts, 300ma plate current) and 200 watts output.
I think the system sounds quite clean and good though.

The 6EW7 modulator tube seems quite happy.

The modulator:
(https://n2dts.smugmug.com/Ham-radio/i-38qkXDJ/0/L/P1010375-L.jpg)

The transmitter:
(https://n2dts.smugmug.com/Ham-radio/i-4NkZBcX/0/L/P1010377-L.jpg)

How do you know that your transmitter is humming?
If you listen with your receiver you are probably receiving a signal that is not coming from the antenna.
I had the same problem and the humming was due to the fact that I was monitoring a signal  radiated by the transmitter power supply cable.
RF inside the transmitter cabinet coupled to the HV power supply transformer, was modulated by the diodes of the supply circuit and radiated by the mains cable into the room..... I filtered the mains circuit and solved the problem.
So ask to a nearby ham what is really radiated by the antenna.
Obviously this is only a possibility, the one that happened to me!!!

Happy new year
Giorgio

I got reports from people while in qso with them.
Screen mod rigs seem to need very good filtering as there is no mod iron inductance to filter the B+ maybe, but I got the hum down to very low levels by really filtering the power supplies.
I did a full wave bridge with choke, cap, choke, cap, big cap (50UF) on one supply, and did a full wave center tap, cap, choke, cap, choke, cap, big cap (40 UF) on the other.
I did cap input on that supply because the chokes would buzz loudly (sound waves) without the input cap.

Maybe the solid state brick diodes are harsh like Don says.

While developing a tube hifi amp I also
observed that  when the ps diodes switch on to "refill" the first capacitor (120 Hz) strong pulses of current flowed through that circuit. Pulses of em field were also emitted by the ps transformer, these pulses were detected by the magnetic pickup at 1 m  distance!!
Anyway, to avoid having current pulses going around the chassis, it is always better to install all ps components on an insulated separate board,  then connect the ground of the last capacitor of the ps to the main chassis.
A ~10 ohm (or higher)  resistor in series with each diode can also mitigate the peak value of the current pulse.
Have  fun!

Giorgio