The AM Forum

Well, I got the quad of 4x150's built and it works well, I built two screen modulators and they both work great, everything else is working great, so its time for a new project, something I never got to last winter.
A low power home brew single band (40 meters) AM rig.

Something like a 6146 modulated by a pair of 6l6 tubes, or a super DX60, a pair of 6146's with screen modulation.

D104 input, not hifi, I want clean but punchy audio like a low power rig should have.
A super dx60 or a single band ranger, xtal control, 7280 to 7295 needing no grid tuning (preset).
Mic gain, maybe a tone control, tune and load, plate current meter, modulator plate current if its plate modulated.

Or maybe a 4x150 run at 6146 voltages with little or no cooling (I am hooked on those tubes).

I need some sort of nice looking box/cabinet to build it in, or I guess I could make it a rack mount thing.

What is the power output of a DX60?
I have heard many great sounding DX60's with the usual mods.

Also, I suppose no one makes crystals anymore, but that is the easy way to do a 40 meter AM rig.

ICM still supplies them:
http://www.icmfg.com/hamradio-manufacturers.html

Thanks for the info.
Guess I could have 2 each for the usual frequencies made up, 7280, 85, 90 and 95 should do it.

I am thinking about a single 4cx250b at 800 volts or so, screen modulated by the solid state modulator run off the same power supply.
I have a boatload of old 4cx250b tubes.
They are smaller then a 6146, take very low drive, and not much screen voltage, the screen modulator takes care of screen protection.

800 or 900 volts on the plate, I need to see what kind of power output I would get, it would make for a very small rig size wise.
I can test it by pulling one tube out of my 4x150 rf deck, hooking up the screen modulator and the rf deck to the same power supply and turning the plate voltage down to 800 volts.

Or, I could plate modulate it with a pair of 6l6's and the 25 watt mod transformer, not sure it would even need any air at the 50 watt output level.

 

Eimac had a note about the 4x150 / 4cx250 family about a min air flow past the base to keep the seals cool .... it said the flow was needed for just fil power ... of course they also had an expensive socket to sell !!!

I think you should build a 813 rig, modded by 811s or even 2e26s, using a power transformer as a mod transformer. Thats watt I think you should do.

Think of the children.

klc

Maybe.
But I ran a pair for a year in loctal sockets at 300 watts out.
That is NO air flow past the base as it was stuck in a socket like an octal tube.
The RF deck is still that way, still works.

I would use the right socket and be safe, but I do not think they need much air unless you push the tubes.
I checked the loctal sockets and they show no signs of getting hot.

Pete

I've used this guy with good results

http://www.af4k.com/crystals.htm

Thanks Al.

I have some of my old ones but they were for the CW section I think.

As far as VFO suggestion, this is not vintage, but works great.

There are DDS boards on eBay using the AD9850 chip for about $10.  see http://www.ebay.com/itm/AD9850-DDS-Signal-Generator-Module-0-40MHz-IC-Test-Equipment-Sine-Square-Wave-/390769215986?pt=LH_DefaultDomain_2&hash=item5afba909f2

Couple this with an Arduino microprocessor and make a rock solid Digital VFO - see http://www.ad7c.com/projects/ad9850-dds-vfo/

I built one and added a small buffer amplifier using a couple of 2N2222 transistors to get about 5 V P-P which is more than enough to drive my Viking II

It is enclosed in a case with the digital readout and voltage regulator to go from 12 VDC to 9 VDC to power Arduino. You tap off the Arduino's 3.3 regulator to power the DDS board.

Note: Output of these DDS boards rolls off pretty fast above 10 MHz, but for a replacement VFO for any transmitter that uses doubling or tripling in circuit like the Viking II does, it is perfect. Total cost for case, Arduino, DDS board, 1602 LCD display and misc parts was perhaps $65.00

Well, I want simple, and I have built VFO/exciters and they are easy, without the digital hash and garbage and phase noise. Its very easy to make a small vfo in a small box, 2 tubes and a handful of parts, but it does not beat plugging in a crystal.
Three crystals and a switch would take care of the job, 7285, 7290 and 7295.
Its not like this would be the only rig I have.

The 12BY7 is your friend.  :)

Yes, seen it done with a 5763 also.
The 4cx250 takes very little drive, so I think a 12BY7 into the grid would do.
I found a circuit in a handbook with a 6b?? something tube that drives a grid of a 6146 in a screen modulated setup, I could try that setup.

Did a test, took one tube out of my 2x4x150 rf deck, hooked up the screen modulator and ran both off the same supply at 600 to 800 volts.
I had to change the pie net coil (plug in) to match the new setup, but I could not get the thing to work well at a reasonable power output. The 4x150 wants more plate voltage to work well under screen modulation.

So its back to the design phase, 4D32 modulated by a pair of 6l6's?
Or a 4D32 screen modulated like a turbo dx60, 20 to 25 watts out?

As a test, I think I will pull two of the 4D32's out of the 3x4D32 rf deck and hook the screen modulator up to that and see how it works off one supply.
I think the dx60 does about 12 watts out after mods, a 4D32 should do about 25 watts carrier.
Now how to go from a D104 to a 600 ohm line level without transformers....

Would the circuit in the dx60 drive a 4D32 as well as a 6146?
I think that would be easy to make in a small space, and it supports a d104 input.
I hear great sounding dx60's often.

Crappy power supply, full wave bridge to a 15 uf oil filled cap, center tap for lower voltage, a tube for grid drive, a mic preamp and the dx60 screen mod with the modifications listed on this site into a 4D32 at 700 to 800 volts.

Brett........ I have a single 4CX250B in the Viking Bud (VK1). At stock voltages I get, on average, 75-90 watt carrier.

I have a variac on the plate transformer primary that'll lift the voltage to just over 1KV. That makes the tube really stand up at 150 watt carrier on cw. The 807s can't mod that to 100% so I run the variac at 120% for a 135w carrier and get 100% mod.

But most of the time the variac is turned WAY down to give 20w carrier to drive the amp to legal limit.

Yes, a 4x150/4cx250b would plate modulate very well up to about 250 watts carrier.
They work at 500 volts on up to 1600 or 1800 with plate modulation.
They are rated at 1600 volts and 200 ma for plate modulation and 2000 volts and 250 ma for CW.
That is 320 watts input (275 out) for ONE tube.
They screen modulate VERY well at 1500 volts on up to 2000 volts, but at 600 to 800 volts they do not put out much power at all with low distortion levels.

For a very small rig, a plate modulator is a pain, a mod transformer, two modulator tubes, phase inverter AB1 or class B driver transformer, various voltage levels needed, bias circuits sometimes, screen protection.
A ranger or a G76 is quite amazing for the amount of stuff they got into a small package.
My home brew single band ranger would be DX100 size and weight....
Anything I build just tends to get huge...

I did a test, took two of the 4D32's out of the 3x4D32 deck, hooked up the screen modulator to it, powered the rf deck and the screen modulator off the same supply at 600 to 800 volts and it worked very well, 35 or 40 watts out and 150 or more watts pep, and it sounded very good.

So a 4D32 with a screen modulator puts out enough power to be useful, sounds good, and will be simple and easy.

There are a number of circuits for the dx60, I am not sure if they will do the 4D32 as well.
Some circuits use a separate -120 volt supply, some do not, and I was surprised to find they use a 6DE7 dual triode, a miniature 9 pin tube!
I also know the Knight kit T150 transmitter used the same tube for a pair of 6146's, same as a 4D32 really.
There are a number of circuits, I do not know which one to try.
And why not use a 6080/6as7 or some big low mew dual triode that is good for some power?

The solid state modulator is great but it needs some high voltage parts and some big heat sinks.

I ordered four 6de7 tubes (they are cheap) to experiment with.
I like the circuit from Ed (KS3K), d104 in, only 2 tubes.

Those 6080 and related tubes are favorites around here because of their high current. Never tried one as a screen modulator but they should accommodate a heavy bleeder where needed to absorb negative screen current. It would be nice to find a schematic for a DC coupled totem pole arrangement instead.

I have a lot of those 6080's and even the JAN shields for them.
I use one in the screen voltage regulator in the 4x150 mod deck and that is it.
I always wanted to use them in audio service but never came up with anything.

This rig uses a 6080 as a screen modulator.

http://amwindow.org/tech/htm/ne1stx.htm

Wow, thanks for that!
Very interesting!

ask and you shall receive! 

Audiophools have been using those for a long time to drive speakers directly. Fine if you have 20 of them.. and the results are not so great without a lot of fussing and a lot of idling current whereas you probably won't ever wear them out as screen regulators or modulators.

I think I am going to prototype the KS3K circuit and try it on my 4D32 deck, and if it works well, build the rest of the transmitter and put it in a cabinet with my first homebrew receiver for an all homebrew station in a box.
6AG7 xtal osc into a 4D32 with as much voltage as I can get on it, screen modulated by a 6DE7 with a 12AX7 d104 mic preamp.

Hello N2DTS, i guess the time is now to try that      http://amfone.net/Amforum/index.php?topic=26603.0 (http://amfone.net/Amforum/index.php?topic=26603.0)

Yes, that looks to be the ticket!
I need to order some 12 volt zeners.

and don't skimp on wattage. It's only a few cents and they might survive a crapout better. just a suggestion.

I would just get the 5 watt jobs.
That is my problem when building, everything gets made crap out proof which makes everything huge.
Looks like a great circuit.

Two or three ways to go, large, small, and/or with a vfo:

Big
http://n2dts.smugmug.com/Ham-radio/i-4r753qd/A

Small with outboard supply:
http://n2dts.smugmug.com/Ham-radio/i-hN3Gj2D/A

Supply:
http://n2dts.smugmug.com/Ham-radio/i-dng5VJ2/A

vfo?
http://n2dts.smugmug.com/Ham-radio/i-Q9bMNFb/A

Small with vfo would be tough.

I'm wondering what sort of output one will get from screen modulating a Class F stage (with 3f traps on both the tube's plate and cathode).

Nice start on the new rig.  Wondering where you get the chassis?  Those HP 4 gang caps usually require some sort of reduction unit on the shaft otherwise the cap won't stay put.  I used two of them in my xmtr, one for the 6146 final and another for the 813 final.

I have no idea what class F is.

I use those 2 inch drives:
http://www.ebay.com/itm/2-Pieces-of-Calibrated-Vernier-Dial-for-Amplifier-Builder-/370770555222

Prices have really gone up.

Hammond makes chassis, and www.tubesandmore sells them.

$12.00 to $14.00 each!
I ordered three.

In a grid modulated amplifier the efficiency at carrier conditions are the half of that during maximum output level, that is during class C.
So when you use a class F amplifier the efficiency is about 88%, a class F amplifier equipped with grid modulation is about 44% under ideal circumstances.

Martin

Four 6de7's came in the mail yesterday, little tubes for modulators!
I was playing around with building another vfo as a test.
Its just an experiment to see what happens if I build it for 40 meters and key the power to it instead of running it on 80 and on all the time and doubling it to 40.

I wonder what the limit is for the 6de7 as a screen modulator.
Could I drive two 4D32's at 50 watts output?

well it'll take a pretty high peak voltage on the power section and up to 50mA. Looks like plate dissipation is the only thing to watch. What screen current do you consider for the 4D32? Do you have a schematic drawn up?

Schematic?
What's that?

The 4D32 is like two 6146's, but the screen voltage and current is very low in screen modulated rigs as the carrier level is what it runs at with only peaks of audio getting up to normal levels, with the other peaks in the negative voltage and current zone.

The actual screen voltage and current with a 4D32 will be around 200 volts and almost no current.
Peak current will be whatever the screen current is at full output (100 watts).

It seems to me like what you modulate with the tube would not have much difference between them, more so the voltage input to the modulator and the voltage out to the tube.
A 6146, or a 4D32, or a number of tubes are all going to run about 700 volts on the plates, and 700 volts input to the modulator, with about 200 volts out to the screens, with little average current, but more current on positive audio peaks the more/bigger tubes you use.
Average current is going to be very low, peak current will be the normal screen current.
A 6de7 was used in the knight kit t150 to modulate a pair of 6146's...
A pair of 4D32's would be 50 watts carrier which is a nice power level...

First rule in building a VFO if the VFO is going to be built in the xmtr.  The base frequency should be half of the lowest freq of operation.  Having the VFO oscillate on the same freq of operation may cause pulling or FMing to occur.  This problem is eliminated by doubling up to the operating freq.  OTOH, if the VFO is going to be some sort of solid state or digital VFO  you may not have this problem.

Thanks for that.
The last vfo/exciter I built runs on 80 and is on all the time, and when keyed is doubled to 40 meters.
It works very well.
I built it that way because I wanted it to run all the time to be stable, but not foul the receiver with its signal.
So I guess on frequency is out and I would need to double again.


I think I am going to build a standalone version of the 6DE7 screen modulator to try on various setups before I start building the standalone 40 meter rig.

I have tried and used the solid state modulator on various setups, and can try the 6DE7 on one, two and three 4D32's as well as one or two or four 4x150's to see what happens.

It would be VERY wild to modulate four 4x150's at 350 watts out with a little 9 pin tube!
D104 in, a 12AX7 and the 6DE7 would be quite crazy.

I got sidetracked, I received the 2nd socket for a 4x150 and changed the 2X 4X150 rf deck from the loctal sockets to the air system sockets.
The loctals worked ok for a year or more at 300 watts out but I want to push things a bit more.
It ended up being a real hack job because there was not quite enough room.
At some point I will have to re do the chassis.

Next is to build the 6DE7 standalone screen modulator and try it on various RF decks and see what happens.

New standalone tube type screen modulator so far:

i'm very interested by this project!

Me too. Brett's signal sounds FB also.

Got everything painted, parts mounted, and started the wiring.

Got the standalone screen modulator done and tested it.
Its a total failure (so far).
I hooked it up to the 2x4x150 rf deck and gave it a shot.
Lots of hum in the microphone input, plenty of distortion, the carrier level adjustment works fine, nothing burnt out/up.

With the line level input it sounds a bit better, still lots of distortion.
A pair of 4x150's may be a bit much to ask the modulator to deal with.
The mic input needs work to get the hum out.

I need to put the 4D32 rf deck in place of the 4x150 and put only one 4D32 in and see what happens.

Grounded one side of the 6.3 volt filament at the 12AX7 tube socket and got rid of 99% of the hum.
What hum is left is ok with me.

Doing tests, if I reduce the grid drive to the 4x150's to almost nothing, the audio gets very nice indeed!
Why is the grid drive so critical?
It was not with the solid state modulator, it did not seem to matter at all.
The adjustable grid leak resistor did not seem to make any difference in how it was set.

I adjusted the voltage input to the modulator, no real change in anything, the carrier level control works very well, I can get plenty of audio on the carrier, 50 watts carrier and over 200 watts pep, or run it much higher.
As the circuit is, the negative peak limiter does not work, a single 6146 in a dx60 is different then a pair of 4x150's, that may be why it does not work.

I actually like the way the D104 sounds in the modulator, but I need to reduce the low end a bit, it sounds a little bit too good for a low power rig.

So why would the grid drive be so critical and need to be so low?

My first thought is that it isn't too low.
Your whole audio and 6de7 circuit was designed to work into a 2 or 3 ma 6146 grid drive circuit. A la Dx60.

I wonder how clean was the audio back then in the Knight with a pair of 6146's?

Your single 4-150 or the 4d32 might be the edge of your circuit's capability.

Well, probably just mirrored or repeated your thoughts.

Perhaps you can parallel a couple of 6de7's with a low mu driver, say a 6c4 or half a 12au7.

But what does the grid drive got to do with it?
The modulator drives the screen grid, that sits at say 200 volts, I had 1000 volts on the plates of the 4x150's, and it sounds bad unless the grid drive is reduced to a very low value.
With my other screen modulator, the grid drive does not seem to make much difference in how things work.
Low or high grid drive, the screen is still going to be at 200 volts and modulated from there.
There was no problem getting 100% modulation positive or negative  with high grid drive, but lots of distortion.

Maybe WA1QIX,Steve, could explain that!

Some things I will try after work, adding a 100k ohm to ground from the output of the modulator to stabilize the load, and also trying the 4D32 rf deck with 1, 2 and 3 tubes in it.

If I want to change the negative peak limiter, I guess I would need to change the 12 volt zener zd4 to something less then 12 volts.

I also want to see what happens if I stiffen up the bias supply and add the feedback from the KS3K design.

It looks like a simple circuit, but it ends up being a lot of stuff.

Lots of fun playing with screen modulators!

I think going very light on the grid drive with the solid state modulator makes things worse, but above that, it does not seem to matter much over a wide range.
With the solid state modulators, it all comes down to how you load the rf deck really, over load it and you get loads of peak modulation, load it light and you get a lot more carrier power out and less or no peaks.
Its easy to tune up once you get used to it.

Well!
The first thing I tried after work was adding the KS3K feedback circuit, from the screen output to the 2nd stage cathode with a .003 and a 1 meg resistor.
Vast improvement, the grid drive now makes little difference in the audio quality and the audio sounds very good, d104 or line input.
I also tried a heavy bleed on the - bias supply, no change, more capacitance, no change, a 100K on the output (screen to ground, real no change.
I thought it sounded quite good, so I put a sine wave into the line input and it looks good down to about 80% negative modulation then it looks funky like the negative limiter is not working well/right.
I tried various experiments with the zd4 and the 10 uf cap across it with no improvement.
Otherwise the circuit seems to work quite well, it modulates a pair of 4x150's nicely, with 1300 volts on the plates its easy to get 100 watts out with boat loads of positive modulation.

The audio chain into the line input sounds smooth and has lots of lows, the D104 sounds like a D104, but I like it, just right for a low power rig.
The only flaw is the poor negative peak limiter.

 

It's a piece of art for sharing with friends!

Yep. Without a schematic, this is just a blog about the project. This is the Technical section.  :D

Current source vs voltage source?

Neg screen current making current modulator unhappy?

--Shane
KD6VXI

Its the qix screen modulator from page two here or the original thread:
http://amfone.net/Amforum/index.php?topic=26603.0

I just added a d104 input.

I did more tests, the negative peak limiter looks ok at higher frequencies, at 40 Hz it looks funky, but at 400 Hz it looks good....
At 75 watts carrier and 90% modulation, the distortion products were 42 db down with a 400 Hz tone.

I hooked the front panel meter up, and set it to read 0-500 volts on the output, and at 75 watts carrier it reads 140 volts.
75 watts is about the best the rig can do with 1300 volts on the plates of the pair of 4x150's.
I might be able to get more with a different tank coil.

I wonder if the modulator would run my quad 4x150 rf deck at 300 watts out.
Crazy to think a 9 pin tube could modulate a 300 watt rig...

Last update.
Tried the qix/modified DX60 modulator on the quad of 4x150's and it worked well up to about 100 watts carrier, and then I think the current overwhelms the modulator. Audibly (over the air) it sounded like something was flashing over or breaking down.

Hello Brett
Maybe you could put 2x6de7 in parallel!

I tried the qix/dx60 modulator on the 4D32 deck and with one 4D32 in it I could get about 5 or 10 watts out with it sounding somewhat clean.
I tried everything, grid drive, carrier level setting, loading and tuning, various plate voltages, it did not want to modulate well in the positive direction.
At one point I had the plate glowing very red, only at very low powers could i get it to sound somewhat clean.

I put the pair of 4x150 deck back on and it sounds very good at 75 to about 100 watts out.
That nixes the idea of a screen modulated 4D32 transmitter.

The 4D32 calls for a stiffer modulator I guess.
The solid state modulator works well with the 4D32 or even a mess of them.
The 4X150 seems to be very easy to screen modulate, but in order to get much power out needs a lot of voltage on the plate. 1200 to 2000 volts.

The solid state modulator with its big heat sinks is sort of out for a small standalone transmitter.
Back to square one...