The AM Forum

Hola!

Well, finally worked out the many problems this 4-1000A modulated by a pair had in store for me. I was so confident it would work right from the start, but it took several days to work them out.

Here's the original thread for reference:     http://amfone.net/Amforum/index.php?topic=21969.0

I swept it and the audio looks quite good. It may be one of the better performing plate modulated rigs I've built so far.  Originally the blower was mounted on the back of the sub-chassis. The noise was terrible!  I mounted the blower on the outside house wall. It's now quiet and the air is cool enuff so I can run the blower even slower... ;D  Good air flow through the tube seals and chimneys.

Anyway, here's the problem...  When I unkey the modulators, I get a big audio spike that makes the modulator plate meter jump almost 1A and the mod xfmr gaps spark. I see the spike come back on the audio line into the low level stages too.  If I disconnect the 200W solid state amp output, (which drives a backwards connected 8 ohm audio xfmr) the spike is gone - so it's not the modulator tubes themselves.

I tried keeping the mod tubes keyed all the time and then unkey the final and HV, but it's still there. I tried keeping the HV on the rig all the time, but same thing.

I put 800 ohm resistors as a test into the modulator grid leads, but still had a bad spike on unkey.  I have a delay in the ant relay that keeps it on for 500ms to dump the RF power on unkey, and tried changing it to instantaneous unkey - same spike problem.

When the audio input to the SS amp is disconnected, no change.  I put .001's to ground on the output leads of the driver xfmr secondary - NG.  I may try some on the 8 ohms leads - Haven't figured what values to try there yet.

It appears the spike is getting generated within the SS amp itself during the RF unkey. It's in a well shielded cabinet. I've used it many times in the past for this purpose with no problems.

The rig sounds so good in the monitor I'm dying to get it on the air, but don't dare cuz of the risk of blowing out my mod iron. (1KW RCA BC piece)

Any ideas?    If I can't get this working right I may have to build up a tube driver maybe using 813's, but I don't want all the heat just for a driver...


*** Here's some shots of Fabio. Notice the ducting to the wall. It leads to a squirrel cage outside pushing air into the rig. I figure the air is cooler than inside most of the year, so why not?  Pictured are two modulators and one RF final on the same chassis. It is neutralized, audio negative feedback around the modulators, vac variables, bandswitched, 70H in the Heising, 20H screen choke, regulated 90V fixed bias with 125V grid leak, variac on screen supply, 3500V on final, Plexiglass view and covered fully for safety.


T

More shots with Plexiglas removed.  All parts are easily accessible for servicing.

#1 shows the problem SS amp on top:

#2 below is with the full dress Plexiglass protection:

Hi Tom,

She looks like a beauty!

Have you considered the transient to be possibly orginating within the negative feedback loop between the plates of the modulator tubes and the solid-state audio driver stage? Have you disconnected the loop and observed the results? Is it possible the high-voltage DC blocking caps in the feedback loop are discharging upon key-up and creating the transient in question here?

I'm looking forward to hearing that rig, and working you, of course. I put my rig back on the air last weekend, and I was on for about an hour-and-a-half this afternoon.

73,

Bruce

Very nice Tom!

Hi Bruce,

Yes, I disconnected the NFB as a first test.  No effect. I can disconnect the input to the SS amp (with the NFB) and it still generates a huge spike on unkey.  The only way to make it stop is to either turn the power off on the SS amp or unplug its output from the driver xfmr.  I can also see the modulator tubes flash a faint blue color as the spike occurs. There was also some arcing within the modulator components that I fixed - shows me the voltages generated are way too high for longevity and this problem must be fixed.


It will be nice to hear your big 4-400 rig and join in the plate modulated scene again... ;D

Thanks for the comment - well, I KNOW it's not in a all-metal paneled rack and it could be made prettier. But it will function flawlessly when I'm done - that's for sure. I do have a few amps that are in all metal cabinets. But this rig is a showboat that I just love to look at every day and see it operate like the invisible motor or invisible man toys...  The enclosed metal rigs are FB, but I soon forget what's inside... ;)  This keeps it magical for me.

T

Hi Tom,

Yes, it does look like the Renwal of ham radio transmitters! Renwal was the maker of the Invisible Man/Woman and V-8 engine kits!

What happens when you leave the entire rig intact, but short the audio input only to the SS audio driver, and then key-up?

For what it is worth, in my rig I apply the 1250 VDC to the 845 audio driver last during the transmit start sequence. I also interlock the 1250 VDC audio driver B+ so it turns off first during the transmit-to-receive sequence. Note that I did not do this for purposes of elimination of any kind of key-up transient, but to ensure that I can't apply audio from the 833A modulator to an essentially turned-off final amplifier stage; the modulator would be unloaded in that case.

Keep feeding us ideas and I'm sure we can determine the origin of this transient.

73,

Bruce

Well, keep in mind plexiglas is no sort of shield, there will be lots of rf in the area.
That might be getting into the solid state amp.
Where is the amp located when you were doing the testing?

My 3x4D32 rig had spikes on unkey, not on the mod iron but an RF spike, and talking with Chuck one day
(wb3fjj) he said he had a rig do that and it was parasitics, i added neutralization and they went away...

Maybe some odd length of cable between the rig and the audio amp, or its unshielded...

My amp is in a cabinet on the desk, and shielded cables run to various modulators.
I used some twinax stuff, twisted shielded cable, about 14 gauge wire inside.

I heard you testing today, you were about 10 over in South Jersey.
Something did not sound quite right with the audio I think.

Brett

Looking at the pictures, the mod and rf tubes are right next to each other?
How do you keep rf out of the audio?
I try and keep the modulators in their own cabinet, and the rf in the other, even at lower power levels I run.

Brett

OK Bruce and Brett -

Bruce:  I will try your suggestion of shorting the SS input and keying up. Though I did try turning its low level pot all the way down and it helped a little, but not much.

I like your idea of sequencing the unkey audio, in my case.  Maybe I can unkey the SS amp output first, then unkey the rig. That is a pain though, vs: making it a delay.

Brett:  You heard me testing ??!!  I was testing into a Bird dummy load this evening located in the shack.  But maybe there was some leakage into the antennas, cuz it uses a common switch.  What frequency did you hear me on?  That will tell the story. But yes, the audio was horrible when I first fired up. It progressively got better until I thought it sounded as good as any plate modulated rig I've built so far. It's quiet running and has some soup, so I'm excited.


Yes, all audio leads are shielded in and out. The SS amp sits up on the rack underneath the steel cabinet rack top. (see the pic) The RF key and unkey looks clean on the scope - it's just that local mod iron spark gap arcing and current meter 1A jump that is troubling.

T

Broadcash rigs mount their mod and RF tubes together all the time. Nice for air flow simplicity. I see a few 4-400 models that have them in a quad line up.  The mod tubes are high level at that point, so a gain of 10db in triode connected 4X1 service is no problem.  I see no signs of RF in the audio during normal operation.

However, your observation may uncover a clue as to why this spike is occuring. Must think on it.

Remember that is does not happen if I discoonect the SS amp output or turn off the SS amp.  It appears to be generated within the SS amp itself. I'll bet a different amp or tube driver wud solve it, but I don't have either available right now. This amp has worked b4 with 4X1 X 833A's in Plexiglas as well as other exposed rigs like this in the past.  Maybe a focus on SS amp filtering in/out is in order.

T

It was in the morning, or around noon time I think.

If you get the spike with the amp on, but not with it off, its rf in the amp.
Cant see it being anything else since turning the amp off eliminates it...

Brett

Broadcast rigs only run at lower frequencies.
lots of RF around the modulators, and back into the driver trans into the amp?

Brett

Tom,
You can store a lot of energy in a 70H choke and mod iron primary. Maybe you need to bias everything off before you remove the load. I would kill the RF drive and remove audio drive first, Then remove the antenna. Just having DC current flowing through the modulation transformer can induce a big spike if it turns off quickly. You might want to bias the modulators into cutoff before the relays drop out. Just like in an old point ignition, you get the spark when the points open and primary current stops flowing..

Brett - You heard the testing in the afternoon and morning - good. I was testing only in the evening after dark on about 3856 into the dummy.

Yes, no doubt the RF is being generated in the SS driver amp on unkey only.  I'm just trying to come up with a way to fix it.

I'll try some more filtering ideas tmw and think more about a relay sequencing during unkey too.

T

Good thinking, Frank.

Yes, there IS huge energy in that 70H inductor during unkey. Where does it normally go?    I always thought that keeping the RF final TR relay on as the last relay was to dump that energy into the final/antenna.  I know sequencing can get complex.

I'll have to think more about first biasing off the modulators on unkey. I could switch in some diodes in the fil CT to do that. But still, that inductor energy needs to be dumped somewhere - into the final as I described above?  Right now the TR antenna relay unclicks 1/2 second after all else is unkeyed. That's the only sequencing I have right now.  The RF carrier unkeys very cleanly.

T

Tom and Frank,

As goes without saying, the start-up/shut-down sequence in any high power plate modulated transmitter is very critical.

I have 40 hys in the mod reactor, a Gates BC-1G mod xfmr, and a 2 uf DC blocking capacitor, along with 38 uf of distributed energy storage/filtering/decoupling in the HV PSU section my rig (probably roughly equivalent to Tom's rig), and I do not have any kind of turn-on/turn-off transients of any significance anywhere in the transmitter. This is because the transmitter turn-on/turn-off sequence largely prevents this from happening, by virtue of interlocked time-delay relay logic switching of the HPA RF drive, HV, HPA screen supply, audio drive, and T/R relay switching. The entire start-up time takes about one second, with the bulk of this delay (about 750 ms) in the HV soft-start. The shut-down sequence takes about 750 ms. I never claimed that my rig was ideal for fast break-in operation; it was'nt designed that way!

I'm sure that you have largely completed the electrical design and construction of the rig, and it may well be a PITA to add the appropriate time-delay relays and circuitry, but the protection and reliability improvements are certainly worth the time and cost involved.

Just my 2 cents!

73,

Bruce

Sounds like a nice sequencing setup, Bruce.

Yes, I plan to add a system here, now that I see it may solve the problem.

For the Class E rig I used a sequencing board that allowed up to six events to be programmed. Worked out FB.

I can eliminate the HV 750 ms delay, cuz it's on all the time - I just key a relay to direct the HV to the particular rig I'm using - quick key up.

So Bruce, how would you suggest I sequence the various events, A-Z?    I've done it all b4, but is hazy at this point for a plate modulated rig.  Bear in mind the modulators are triode connected with the driver xfmr CT to ground, so no grid or screen bias to deal with there.

T

Tom You could just open the CT with a relay contact with a large resistor across the contacts. Also put a resistor in series with cap across the contact so it shuts down slowly.

Hi Tom,

Glad to be of help! Here is how I do it in my 2x 4-400A by 2x 833A rig.

Here goes:

A) For transmit start-up, the following events occur serially:

1) Plate contactor closes 220 VAC circuit to HV/plate PSU

2) Soft-start limit resistor is shorted approx. 700ms. later.

3) Apply HPA screen voltage

4) T/R relay contacts transfer

5) Apply RF drive to HPA

6) 1250 VDC audio driver plate supply contactor closes, enabling audio to be applied to the modulator stage.

B) For transmit shut-down to receive mode, the following events occur serially:

1) 1250 VDC audio driver plate supply contactor opens, effectively removing audio drive to the modulator stage, precluding the possibility of the modulator looking into an unloaded condition.

2) RF drive is removed from the HPA

3) HPA screen voltage is removed

4) 220 VAC plate supply contactor opens, removing the plate voltage from both the 4-400As and the 833As.

5) T/R relay contacts transfer to "receive".

The time delay relays I use are made by a company named NCC (National Controls Corp.). They are microprocessor controlled and programmable by virtue of DIP switch selection, and plug directly into a standard octal socket. Hell, you can even DIN-rail mount them if you want for housekeeping purposes, or if chassis real estate is now at a premium. I bought mine years ago thru e-Bay, but both Allied and Newark carry them, and they are reasonably priced. I bypass the 115 VAC input to each TD relay with a couple of .001 uf caps to ground, and never have any issues with RFI affecting their operation, etc.

Forget about those adjustable RC time constant-controlled TD relays; the ones with the control knob on the top. They are fine for use in a soft-start circuit for power tube filaments, etc., but not in those applications where you want precise and repeatable time delays. This is why I used the NCC relays.

Let me know if you need more help with this!

73,

Bruce

Hi
 
Maybe a stupid idea from me,
unkeying such big transmitter can cause a spike in the power line/220 VAC line.
This spike got into the SS power supply trough the AC line,maybe small but amplified by the SS Amplifier and drove the Modulator,and since its not loaded( the RF amplifier is of) than a big surge current gets into  the modulator,makes a spark jump across the Gap.
Try to use a line filter for removing spikes in the power line/220 VAC line on your SS Amplifier.


Gito

T,
Any DC on the output of the SS amplifier?

Tom,

how about a good old fashioned speech amp just up scaled a bit? If'n it were mine, I'd be protecting that iron at all costs, S.S. amp b damned. Aint gonna fret over no 2 bit amp - just zag instead of zig.  ;)

Italics and bolded text my emphasis

I think this is worth investigating. Is there some sort of line filtering that can be inserted for the SS amp AC  connection?  Another possibility would be to use another tube amp to drive the 8 ohm input to the backwards connected AF xfmr.  I'm interested in this problem because I'm considering using the same approach for my 813 x 2 - 813's xmtr.  I'm already convinced that I'm going to have to spend a lot of time figuring out the sequencing.

I have a similar rack except the superstructure is made of extruded aluminum.  I still like your plexiglass approach but think that some sort of shielding will be ultimately needed to keep RF out of the shack.  I think I will be putting the 2 x 813's mod tubes on their own shelve.

Al

   Tom,

    I think the quick unkeying of the HV B+ is the root of the problem. You want that big 70H choke to decay at a L/R time constant. I'm having trouble visualizing your circuit, but If the B+ line floats after unkeying, when energy stored is yet to be released, then all sorts of stuff could happen.

    Some ideas come to mind such as a HV commutation diode placed where it will work to aid the L/R energy dump and yet be transparent during normal operation. I also might have another relay ground the B+ line through a 1K (large wattage) resistor whenever the upstream HV relay opens up.

   Also keying HV DC is tricky business. Most relays will fail this test except for very exotic and very expensive ones. What kind of relay do you use to key the B+?

If you can control the transient, then maybe the AF SS driver amp won't be affected.  What comes first, the Chicken or the egg?  ???

Jim
WD5JKO

   Brian,

   You bring up a good point. The issue here though is that Tom keeps the HV on all the time, and keys the HV B+ with a relay. This unquestionably creates a different monster to contend with. I'm not saying this is bad, but in doing so there might be some unforeseen consequences, along with some different ways to tame an unruly shrew.

Jim
WD5JKO

I wildly speculate that the RF is biasing some device in the SS amp enough to cause a DC offset in some stage or another but not enough to be noticed in the audio during operation. When the Rf disappears, so does the bias, and so comes the pulse from the SS amplifier. When the rig is keyed, this event inside the SS audio amp is over with before the modulator is powered up enough to be affected by it.

Tom,
I'm thinking it is a rate of change transient. Try a series RC across the modulator key contacts to slow down the turn off of the tubes. Make sure the antenna is delayed a bit longer. You want the tube to bias itself off slowly so the iron doesn't see a transient. The driver closed loop may be generating another transient when it sees a load change. IF you have extra drive power put a resistor across the grids of the modulators to hold the tube input Z more stable. A sequencer will work as long as the transient has a place to go. Remember the old point ignition system and how bad the car ran when the condenser across the points went bad. This condenser was a shaping network to extend the turn off a bit to generate a longer spark.  You need to extend it a lot longer so there is no transient. Killing primary power slows down the turn off but you have to live with all the extra hassle of step start noise and shocks to the power supply turning on and off.

Great suggestions! That's just what I needed - things to try until I find a solution. Lot's to work on now.

Yes, the HV primary AC is on ALL the time and the HV DC is keyed by a HV relay and routed to the particular rig. It’s one of those long arm swing types on insulators. I do this simply for safety reasons since the HV is on the rig ONLY when I key up and not sitting there all the time. However, for this un-key spike problem, by keeping the HV on the 4X1 rig all the time did NOT have any effect on the inductive spike. I tried that as one of my first tests... sigh.

Bruce - that's a beautifully written and detailed explanation for sequencing and shud be accessible for future use. Well, right now I am partially there with the RF drive being unkeyed first and the TR ant relay last. But the mod tube relay and RF tube final relay come on and off at the same time. I plan today to put in separate relays and unkey the modulators before the RF final as suggested.

BTW, Frank, I already tried putting a cap across the cathode keying relay contact's resistor - no help. I also left the modulators on as an experiment and other relays trials - no help.


Gito - I notice you usually come up with great solutions in other threads.  I will most certainly try one of those spike protected AC strips to see if it helps the SS amp - thanks.

One of the first things I want to try is to put a relay on the output of the SS amp that shuts off early, before all the other relays, as Bruce suggested. Since the spike is obviously coming out of the SS amp, that cud be an easy fix. (The spike problem goes away when I shut off the SS power or disconnect its output from the driver xfmr)


Derb, if I had a tube driver amp I wud use it, but don’t.   I’m starting to think that maybe it, too, wud have the same problem if it’s a sequencing problem. We’ll see. I’ll post the solution, of course. If I run outa things, I’ll build up a triode direct-connected 813 audio driver. No driver transformers.

I’m running the rig at 250w out so hope to minimize stress. (Yes, by loading it lightly, a 4X1 WILL limbo low)  But still, that inductive kick is pretty severe.   It caused me to find a few weak spots amongst the mounting of the mod parts already that were sparking. That’s a good thing, actually, but now the mod xfmr arc gaps are taking all the abuse.

I plan to take a pic of the outside mounted blower today. It is encased in a Plexiglas housing that makes it look like a tiny air conditioner. I’ll paint it white to match the house later.


Bear sent me an email about trying a slope capacitor technique to slowly unenergize the mod iron on unkey.  He said it worked for W1RKW’s rig in  some past thread here. Anyone know where that is?

Plexiglas:  I don’t worry about shielding. As others have seen here, I’ve run many rigs under Plexi over the years, including a full blown pair of 4X1’s PDMed by a pair with no RF in the audio problems I couldn’t cure. Radiation calculations work out to be minuscule at 75 and 40M, the bands of operation.  I just love watching 4X1’s operate thru the Plexi.  BTW, Fabio’s nickname is gonna be, “The Invisible Mawl”  - named after the Invisible  Man and Invisible Motor toys..  

Thanks again for all the help, OM’s.

Hey Tom it's a beaut.  You sure do good work.  Can you lift that thing yourself?  Must be pretty heavy.

This probably has nothing to do with nothing but I'm curious about the braid to the anode caps--I've never seen that done before--I've always seen copper or silver plate flat strap.  Maybe I have not been around long enough.  I wonder if the braid has any additional inductance.

73

Rob
K5UJ

Interesting. I wonder how I would test this. I could move the SS amp closer to the oprating position, away from the rig, but then the NFB audio lead coming back wud be much longer.  Right now I just use a longer 1v audio feed to the SS amp.  

Which, BTW, I have a VERY slight 60hz hum in the audio. It disappears completely when I disconnect the low level audio from the SS amp.  The RF carrier is perfect at that point.  I wonder if a 20' run of regular shielded  RS mic cable is good enuff to handle the task from the 1V low level compressor output to the SS amp?  (Next problem to solve)

Right now I'm testing into the Bird dummy load located on the floor under the audio gear. If there wud be RF problems, it wud be now... ;)


Rob - On the silver plated braided shield used for RF connections... I've used it for ground connections for years.  I use it for the plate cap connections to take the mechanical stress off the fragile plate cap. I think copper strap is a little harsh, but it wud probably be OK anyway. It shud be OK for RF despite the mesh.

Tom,

It sounds like you have a lot of good suggestions to try out for the rig and you should have a good handle on sequencing.  I don't really like the idea of keying the rectified DC since that is a recipe for arcs and transient events.  Although it is less of a problem with the fairly rugged screen structure in the 4-1000A you do want to make sure that the screen doesn't come up before HV and is dropped simultaneously or before the HV is dumped to avoid frying the screen structure.  It is also a good idea to drop the modulator drive prior to dumping HV.  

A scope is your best bet for looking for stray RF in the modulator and also capturing transient events.  Although a storage scope is handy for glitch capture you can do a pretty good job with a regular scope set for single sweep and a decent digital camera (darker room with proper mix of ISO, aperture, and long exposure shutter setting so that the shutter is open during the event).

I may have missed it in your description but do you have RF chokes in the output of your SS amp (bypassed on each end) where it feeds the modulator tubes?  SS stuff does some interesting things when RF shows up where it doesn't belong.  My first attempt at using SS relays to key the plate transformer in my Desk KW produced interesting results until I took care of RF getting into the relays via the control line; the primary voltage was being "modulated" by RF on the control lines which produced a really interesting sounding carrier  :)

Rodger WQ9E

Rob,

My TMC GPT-750 also uses regular tin-plated copper braid as the interconnect between the heat radiating plate caps on the 4-250As and the RF plate choke, so it has been done in the past in commercially built rigs.

In my homebrew 2x 4-400A rig, I use fairly flexible soft copper strap for this connection. Works FB.

73,

Bruce

Tom,

Are the leads between the output of the negative feedback ladder/DC blocking cap circuit and the SS audio driver amp shielded?

I'm not suggesting that this is the cause of the transient event that is occuring, but if those leads are unshielded, they are a potential source of noise and RF to find their way into the low level stages of the driver where the NFB is introduced.

73,

Bruce

Hi Rodger,

The switched DC is a carry-over from using it with linears here. Seemed to work OK there, but maybe not for plate mod. The RF carrier keying looks very clean with no spikes on the scope – without the modulator involved.  Bear in mind keeping the HV on the rig all the time had no effect on this problem, but you bring up a good point about the screen being naked for a few milliseconds….I’ll address that later on for sure.

Interesting on using chokes in series wid the audio leads to the modulator grids. What values shud I use there, like 2.5mh types or less?  Bypass caps on each end of .001?  I have nothing there except for a 5K load power resistor across the driver transformer for protection.

I am going to add a relay to the output of the SS amp that will disconnect it off first, after unkey.  I wonder where I shud place this relay – at the 8 ohm output of the amp, or at the driver xfmr output secondary on its way to the mod grids?      The relay idea may only be a band aid since it will probably still be there. I will try the power strip test first, then look at sequencing.

T

Hi Rodger,

The switched DC is a carry-over from using it with linears here. Seemed to work OK there, but maybe not for plate mod. The RF carrier keying looks very clean with no spikes on the scope – without the modulator involved.  Bear in mind keeping the HV on the rig all the time had no effect on this problem, but you bring up a good point about the screen being naked for a few milliseconds….I’ll address that later on for sure.

Interesting on using chokes in series wid the audio leads to the modulator grids. What values shud I use there, like 2.5mh types or less?  Bypass caps on each end of .001?  I have nothing there except for a 5K load power resistor across the driver transformer for protection.

I am going to add a relay to the output of the SS amp that will disconnect it off first, after unkey.  I wonder where I shud place this relay – at the 8 ohm output of the amp, or at the driver xfmr output secondary on its way to the mod grids?      The relay idea may only be a band aid since it will probably still be there. I will try the power strip test first, then look at sequencing.

T

[/quote]

Tom,

I would locate this "audio driver muting relay", at the 8 ohm output of the SS driver amplifier.

Being that it is a SS audio amp, and most probably does not utilize an audio output transformer for impedance matching to a loudspeaker load, you will do no damage to the amplifier if the load is lifted from the amplifier output when any kind of audio is driving the amplifier.

And you are correct; this is more of a band-aid over an open wound solution, than finding and solving the cause of the transient.

73,

Bruce

Bruce,

Yes, both the input and output of the NFB ladder uses shielded cable.  The NFB works very well and lets me put in MORE NFB than I can use before taking off. It cleans up the deep low end tremendously. I ran a sweep last night and found it clean down to about 25-30hz. Without the NFB, it was good to about 70hz at best. Love that NFB!  In fact, when using the NFB I can run the audio phase in the "shark-fin class E mode"....  which can only be done with a very clean low end.  Normally I need to run the phase the other way. The balanced modulator rigs here are the only ones I can normally do that with.

Will-do on the 8 ohm position for the relay.  - tnx.

T

Tom,

Just out of curiousity, how much NFB are you running in that rig?

I use 6 dB from the plates of the 833As to the low-level voltage amplifier stage in my push-pull class A 845 driver circuit. As you and I had discussed some time ago, 6 dB of NFB is fine with all-triodes in the audio line-up, as I am using.

Obviously, as you have verified, the NFB loop is not contributing to any kind of instability.

73,

Bruce

Tom,

Standard 2.5 mh RF chokes would be my choice.  I am not sure what your audio impedance is at the output of SS audio unit but that will determine the highest value of capacitance you can use without impacting the audio response.  For example a .001 cap will have a bit of 15K of capacitive reactance at 10 Khz. so you would want to compare that to the audio impedance of your circuit so as not to excessive roll off your high frequencies.  There are a lot of handy reactance calculators on the web such as:  http://www.sengpielaudio.com/calculator-RC.htm in case you don't have one handy.

Rodger WQ9E

Here's the pics of the outside mounted blower for the 4X1 rig.  I first had it mounted on the back of the sub-chassis in the shack. The noise was horrible!  Blower noise gets into the audio as intermod, just like hum in the audio produces artifacts.  I could not run my compression at all - and still had blower noise with just an EQ and mike.  To cool three 4X1's effectively, you need a lot of air. The squirrel cage itself is what makes most of the motor and air noise via turbulance.

Anyway, once I mounted it outside, the noise dropped to a whisper. With the covered 4X1's in the rack, there is just the sound of exiting air - white noise and almost soothing. I can run full compression now with the blower Variaced up all the way. With the cooler temps outside, I can variac the blower down to 120VAC - it's a 240VAC blower.  

As you can see from the pics, it looks almost like a small air cornditioner - hardly noticable on the house. I mounted the blower itself on 1" of styrofoam to keep vibration off the wall.   I'll tell ya, now I can cool the tubes fully and still have a quiet shack. I'm real happy about that!

T

AFTER painting - hi hi FB OM/YL/XYL/OW approved!

Cool fan system, Tom!  I have been following this thread with great interest; you have accomplished so much in a short time span!   Just wondering what sort of filter you are using to suppress the unlicensed critters from entering the squirrel cage?

heheh  - Brian, thankfully, the nearest neighbor is 1/4 mile away and the next one 1/2 mile..  BTW, how much wud you charge me send send you Fabio to "HRO Beautify?"   ;)


Rick:  I'll bet the Lady Bugs are gonna love this white-color shroud. Perfect spot.  Maybe a fine screen on the bottom is a good idea...

Rodger - I'll try your 2.5mh / .001 filter as a next test - thanks.  Another thing - I just tested it -  I have a fixed, regulated 90V on the grid all the time. When keyed with no plate voltage, there is just 120 ma screen current with RF full drive. So that is no problem for the screen. Unless you can think of another related risk?


 ** BTW, THE RFI/Noise AC LINE FILTER FOR THE SS AMP DID NOT HELP AT ALL. **

Bruce: I'm not sure how much NFB I'm using, but judging from the gain reduction, it must be at least 6-10db. I set it to the point of diminishing returns for the low end sweep.

Still testing and trying things...

T
  

OK, Brian - I'll have to live with looking at Fabio until you get caught up... ;)

  I don't have a schematic of the SS amp, but will have to open it up.  If it does have that cap, do I just take it out or reduce its size?

**  Rodger:  I tried the 2.5mh / .001 filters on the output of the driver xfmr to the grids. Made no difference.

I'm concerned that the mod meter almost slams against the pin  at over 1A on unkey. I have the mod transformer primary gaps close, but think I'll add some gaps on the mod xfmr secondary too. That's all I'd need to blow this iron in the testing phase when I KNOW there is a problem and can do something about it.  The rig runs so nicely otherwise in all respects.

Think I'll try a bandaid sequencing relay on the SS amp output next. I don't care if IT blows - I'll just build up a tube direct coupled driver... :(

T

Tom,
  If the 200 watt amp is disconnected from the rig, sitting where it would normally be but connected to a speaker or load of some sort, do you get a "thump" out of the speaker when you un-key? If not, then it would be some kind of interaction between the rig and amp when connected. If you do, then it is something getting into the amp.
  Also, I wonder if you have some other impedance taps on the modulator side of the backwards 8 ohm transformer you might try and see if that changes anything.
  Just a thought.....Bill

If you are REALLY an OLD BUZZARD, the simple solution is to NEVER UNKEY the rig.  How long can you talk?   Time to set a record!

Very good idea, Bill. The "disconnect / divide and conquer" technique works well. Will do.

Brian - The Tron always told me to use the primary for gaps- though there has been a debate about that for years. This 1KW RCA BC iron has the gaps STOCK on the primary. But Tron always told me to protect both pri and sec anyway. The 10KW iron I used to have had gaps on both. In fact, I'm making up TWO more arc gaps - one for the mod sec and another for the 70H Heising reactor. That shud cover all bases... ;D


Yep, never unkey - something like the Roach Motel... ;)
T

Tom,

Another possibility is to look for an AC line voltage transient upon unkeying that may be on the AC line that feeds the SS audio driver amplifier.

If that amplifier is DC-coupled (a real possibility), it's low frequency response can be down to DC in the event no effort was made by the designer to limit the low-end for speaker driver protection, etc. Any switching transient that appears on the 120 VAC input to the amplifier may be amplified and is being seen at the amplifier output.

Many solid state audio amplifiers can be unstable at very low frequencies when driving an extremely reactive or inductive load, such as a driver transformer. I have a Crown D-75; it is a high quality amplifier, DC-coupled from input to output. Crown does not recommend that this amplifier be used to directly drive an audio transformer at the output; if a transformer is required at the amplifier output, for driving a 70.7 volt line for example, a 600 uf non-polar capacitor shunted with a 4 ohm/20 watt resistor is required in series with one side of the audio output. Crown was emphatic about the use of a non-polar cap. As far as I know, such a capacitor is not a commonly available item, due to the value of capacitance.

By the same token, you may be switching so much energy in that transmitter between transmit and receive that the switching transient may be appearing at the amplifier audio input and is being amplified accordingly. That is why I asked earlier if you tried to short the amplfier audio input and to observe the result.

Regardless, I really believe the issue is in the solid-state audio driver amplfier you are using in this application. A scope looking at that amp will be your best friend at this point.

73,

Bruce

Yes, I protect both the primary and secondary windings of my modulation transformer with spark gaps. I also protect the swinging choke for the HV plate supply with a spark gap.

Now that you mention it, I should add a spark gap across the modulation reactor as well.

73,

Bruce

  I was just reading in my Crown M600 manual very similar language. The M600 is an old DC coupled 600 watt @ 8 ohms brute of an amp. A place I worked moons ago used these in a bridge to drive a HV transformer to make 40,000 volts at 50 ma (yes 2 KW). The HV transformer was a DC short, so we HAD TO use a series NON-POLAR capacitor between the AMP one side and the transformer. We used dual 1600Uf @ 450v electrolytic caps in series with the plus signs together, making a poor mans AC capacitor. These large caps act like a short when reverse biased. The machines ran this way for years 24/7. Adding a power diode across  each cap made NO difference.

  I used the M600 as a modulator once, and did the same thing with two caps in series. It worked fantastic. So Tom, maybe your SS driver in the presence of RF is generating an offset voltage in the output leading to a large DC current into the transformer. Then when you unkey, the current goes to zero almost instantly causing a pulse to come through. If so, a series AC cap combo between the SS AMP and the driver transformer might just be the ticket.

Jim
WD5JKO

Very good, Jim and Bruce -

Yes, I will try a non-polarized electroylic combo as you described.

I will also try the short-to gnd test too, Bruce.  Though, the SS amp's input audio pot turned all the way down had no effect. I think it probably puts the input circuit to ground - not sure yet.

Well, so far the RF is not coming in thru the input audio, NFB, AC line - or coming in thru the grid leads output as based on my tests.   I will put a scope on the ac line, Bruce, to check the AC dynamic stability. And, also poke around with the scope probe at the various in and outs to see what they look like.

If it's not fixed  the coupling cap, then it almost seems to be generated by magnetic coupling thru the steel case from the steel cabinet housing the 4X1 below it.   We'll see.


After dinner I will build a pair of spark gaps FIRST, install them and continue testing with all iron protected.

Thanks again for all the suggestions. I wuda run out of then by now... ;)

T

Tom,
Rather than guessing, shoot a tone through the system and monitor the output of the SS amp with a scope and unkey to see what happens. Many times a little RF in an amp will reduce the gain. Remove the RF and gain goes back to normal. If the input stage of the amp is a bipolar op amp or bipolar transistor this is very possible.

Tom Bruce, did not know that about the TMC braid, thanks.  I used to use braid outside--big mistake but that's a whole different deal.  Time now for a noob question--what is NFB?  Can't decode that one.  Tnx

Rob

Hi Rob,

No problem! NFB stands for negative feedback.

73,

Bruce

Yes, in this case audio negative feedback derived from a LONG string of 220K resistors across one modulator plate cap to ground - then tapped at the low voltage bottom into an IC which mixes it with the regular 1 volt audio. The plate cap is selected to provide negative feedback - the other cap wud give positive feedback and take off.

Just to show how easy it is to make a set of arc gaps for your iron, here's a set I'm adding to the mod xfmr sec and Heising reactor for protection. Once set to arc very occassionally when a big, rouge modulation peak is hit, they can remain set as-is for a long time.

Shown installed in a place you don't want your hand when keyed-up... ;D

Now to fire this baby up and start looking around wid the scope.


T

Maybe it was Derb then?

I think your outdoor blower setup is cool (pun), I could never get used to a loud blower in the shack, it kind of ruins the nice audio to have something like that in the background.
I really like rigs with no fans or blowers, the push pull 812 rig was like that, and so is the 3X4D32 rig, the 813 rig has blowers but not loud ones.
The new computer has the loudest fan in the shack, I should do something about that...

Brett

Tom,
As usual your threads are the best.......this one IMHO is great.
Just a couple of points........
When Jamie was finishing my 4x1 by 833's, for a while I drove the 833's with a backward connected transformer as u are (Hammond). I drove it with a QSC SS amp (600 watts in the bridge mode into 4 ohms). The power of the amp has nothing to do with it.....its the voltage available that counts. I had no problems with the QSC amp. This amp is very forgiving.......will work into reactive loads with no problem. Finally Jamie finished the cathode follower ckt. to drive the 833's.......a pair of 807's driven by a pair of 807's.
Using the backward connected tranny was fine untill the grids of the 833's drew grid current. At that point the Z of the grid ckt went WAY down causing some drive problems.
Anyway.........I think u should consider some sort of cathode follower to drive ur modumalotors.....
Tom......I am NOT trying to tell u how to design ur rig.....u are a LOT smarter the me!
Again.....JMHO.

Bill

The 8 ohm to grids setup works very well with all sorts of tubes.
I have done it with 811a's, 100TH's, 4cx250b's with great results.

Both the 811a's and 100th tubes are class b with grid current, and I think an 8 ohm source is quite stiff, so it works well.

Most people terminate the secondary of the driver trans with resistors that match the output impiedance so its loaded all the time.

I have never had any kind of spike on unkey, the audio stays on all the time (to the grids).

There really should be no change in the audio amp or the mod grids when the transmitter is unkeyed.
The mod tube grids should look like an open circuit, while transmitting, or not.
The audio amp should see the same conditions, key up or down.
The only time things would change (in class B) would be on mod peaks when the grids draw current.
With a pair of 4-1000 tubes, I am not sure you would need to drive them to class B to get 100% modulation!

Now if you were running AB1, key up or down, full modulation, the audio amp would always be looking into an open, no load, no change at all, ever.

Brett

Hi.
I always use AC filters/surge suppressor in my Transmitters ,I think /hope it will cleanup the AC line from RF and surges

Regards

Gito

I added the back-to-back pair of electrolytics (1600 uf @ 450V = 800 uf) to the output of the SS amp.  It has made a huge difference, almost livable, though still not there.  The spike on the modulator current meter is now only about 200ma. (It was over an amp before.)  There is still an ocassional spark on the gaps. (The new gaps are working FB)

It was DEFINATELY the SS amp not liking what it saw without the capacitor - AND now being influenced by the RF or collasping inductor fields..  Dig this: I can disconnect the SS amp's input completely, disconnect the amp's output 8 ohms completely, and see no signs of pulse from the amp on the scope when unkeying, as you'd expect. BUT by adding a small 12" wire hanging off one of the 8 ohm terminals, the SS amp spikes and I can see the overload diode flash.  It is obviously very sensitive to mag fields. Maybe I will move it to the other side of the room and deal with that.

All in all, the 4X1 rig seems very stable for both RF and audio when the SS amp behaves. But the SS amp is squirrelly and that's when the arcing starts.

So, Bruce and Rodger, your suggestions were pretty close to the problem, as well as a few others in general.

Thanks for the comments Bill. I may take you up on it and axe for a schematic of the tube cathode follower if this doesn't pan out. But when I hear the audio in the monitor of this thing with the full NFB working, I am really excited... :-)

Thanks again for all the help, OMs! - I'm gonna try some different transformer taps and if still NG, I'll move the amp. I won't stop til it is totally clean on unkey. It actually did unkey a few times without sparks or current meter spikes, so I know it's possible with this rig and general config - with some more work.

Seez ya tmw.

T

Tom,
This is a pic of the chassis that has cathode follower ckt........

This is a wider view of the sub chassis........

Oh my... what a nice looking driver!  I wonder how you drive it?  I have about 1-2 volts available for whatever audio driver I use.

I'll bet the requirements for the 833A's are very similar to triode-connected 4X1's.

Please send me (or better yet, post) the schematic and what voltages are required for it.  I don't have enuff 807's but wonder if I cud sub them for triode connected 813's?  Maybe the 807's have more gain as-is  - have to look it up.


General comment:  I just measured the grid bias on the rig when in operation and was surprised. I have 90V of fixed regulated bias, but at 40 ma of grid current generates an additional -215 volts of grid leak for a little over -300 bias. The specs call for about -200v, so the final is in DEEP class C, but seems to like it.

I'm using a 5K grid leak resistor which is a little high.

I heard it said here before that deep class C makes better modulation. (as well as the higher efficiency.) I wonder if anyone has experimented with this?

T

Hi Tom,

I'm pleased the suggestion I made was workable and things now appear to be going in the right direction for you! Glad to be of service, OM! Did you also bypass the caps with the 4 ohm/20 watt resistor as I had suggested? That may well be the icing on the cake.

Now, one more point. Personally, I hate to couple really clean audio through an electrolytic capacitor, particularly such a large value as what you are now using. They generally sound horrible, which is one reason, aside from cost, that virtually all high quality solid state audio amplifiers did away with the large electrolytic DC blocking caps at the output and went with an all-DC coupled architecture. Speaker crossover network designers did the same thing; they dumped the electrolytics once low-cost, high quality film caps became available.

That said, you may find the sonic degradation imposed by the back-to-back electrolytics (to effectively create a non-polarized cap) is not that significant, and that you are still pleased with the end result.

In my rig, all audio stages are coupled or decoupled with either polypropylene film or oil caps. The power supply filter caps in the 1250 VDC plate supply for the 845s are oil caps, and the 400 VDC supply for the input and driver tubes to the 845s is filtered and decoupled with polypropylenes. Come to think about it, the electronically regulated shunt bias supply for the 833A modulators uses all polypropylenes for power supply filtering as well. These capacitors are also much more reliable, long-term, when compared to electrolytic capacitors.

Personally, I would consider the use of a solid state (or vacuum tube amplifier) that does not require the use of the output cap/resistive network, and will tolerate the inductive/reactive load of the driver xfmr while still remaining unconditionably stable under any load condition or phase angle, etc.

Just my thoughts!

When can we hear this rig???

73,

Bruce

Tom,
The driver ckt. using the 807's was not designed by Jamie. It is a ckt used by some BC transmitter and I cant remember which one.
Perhaps someone on the board will know.
Trying to contact Jamie is almost imposable!
Bill 

Hi Bill,

It was used with some variations during it's production life in the Gates BC-1T thru BC-1H rigs, and probably the 500 watt versions of these transmitters as well.

I have the 807 audio driver board assy. from a Gates BC-1G; I may consider using it in my next 833A modulator, which will be mated with my TMC GPT-750.

73,

Bruce

Thanks Bruce,
Perhaps u can supply the sceezematic to Tom.........
It is a very good ckt.

Bill

OK, Bill. I'll bet someone will come up with a broadcash 807 cathode driver schematic.

* whoops modify - I see Bruce did already. Can you post it, Bruce?


Bruce - I decided to move the SS amp to the other side of the room.  Do you think I should feed the 8 ohm cables over to the rig, or mount the driver transformer next to the SS amp and run the high impedance modulator grid leads over to the rig?

Seeing how sensitive the amp is to the 8 ohms being exposed, maybe I shud run the grid leads instead.  I will use shielded cable for whatever I do, of course.

T

Tom,
I drive the 807 stuff with the output of my BL40 Modulimiter. It takes about 2 Volts PEP.
Bill

Hi Tom,

Please make the long cable runs only with the 8 ohm amplifier output! You want to keep the high impedance (secondary) side of the driver transformer cable run as short as possible.

Further to my previous thread on polypropylene caps for power supply filtering, etc.; I have attached a photo of the underside of my electronically shunt regulated bias supply I use with my class B 833A modulator. The polypropylene caps are the large black capacitors, and are frequently used for AC motor starting or motor running applications. These caps are rated at 40 uf each, and at 630 VDC. The three meters on the front panel are for monitoring the 845 audio driver cathode current, 845 plate voltage, and the regulated negative DC output to the 833A modulator grids.

73,

Bruce

Bill,

Two volts is very do-able.  I will have to get this SS amp working temp and see about that 807 driver.  What various voltages does it require?


Very professional looking unit, Bruce.  As I remember, your work is indistinguisable from factory work...(American, that is)  ;D

T

Hi Bill,

Yes it is. It saved Gates a lot of cost when they were able to delete that massive push-pull class A 845 audio driver circuit along with it's large 1200 VDC plate supply, and the problematic and very costly class B audio driver transformer.

Unfortunately, I only have a poor quality copy of the original schematic. Let me see if I can redraw it along with the special accompanying DC power supply, and post it here on the board.

Bill: did you get my private message to you of late yesterday? I never heard back from you!

73,

Bruce

Tom,

Just a thought... Have you thought of bypassing the amp on the output for RF?  .1, .01 and .001 uF across the terminals, or from each speaker terminal to chassis ground?

Had good luck with that combo when running AM and linears in the past, when TV's where more radio receivers.  Got rid of more than a bit.  You also might wanna try running the cables from the amp through some ferrite RIGHT at the speaker terminals.

Good RF bypassing can be a bear sometimes...  I know, been there.  Eventually you'll find it, tho.  I LOVE the design.  See Thru is SECKZI, just ask Victoria's Secret :)


--Shane

Another good place to look for high value AC caps is lighting supply stores.  Metal Halide, High Pressure Sodium and associated lights take one to three caps to run correctly.

I've found many different sizes in ballasts across the years.


--Shane

I dont think the cap should be needed.
All you need is to move the amp someplace away from the rig, and find some shielded cable heavy enough to run the 8 ohms in.
I bet that cleans things up.

More might not be better here, as far as audio power goes.
I use a symetrics dual 20 watt amp I got at a pawn shop, because it is only one rack space high, and it was dirt cheap.
It takes less than half power to drive anything I have.
Each output runs in twinax to 1/4 inch trs jacks on the modulator cabinets.

No matter what antenna experiments I do (antenns are over the house) I do not get into the audio.

If a pair of 4-1000 tubes take 40 watts drive power, I would use a 100 watt amp, not a 600 watt amp.

Brett


 

Brett,

Yep, wish I had a smaller SS amp to use, but it's all I have right now.  I'm moving the amp across the room right now. The good news is I can now unkey the rig and it is not affecting the SS amp across the room. However, it is not hooked to the rig yet, just has leads dangling from the 8 ohms output. Before, the SS amp wud act up when in this config.  We'll see.

Thanks for the redrawing effort, Bruce. I'll bet others will be interested too, considering it's a direct coupled driver and will use 2 volts drive. And, easy to get 807's.

Shane, yes I did try several bypassing schemes with the amp in the old position on top of the rig. I think it was simply magnetic coupling thru the steel cabinets into sensitive solid state circuitry.  This amp always worked FB when placed remote from the rigs. I shuda remembered that from before, but didn't.

Bruce, yes, using a low impedance 8 ohm run is better - I just was concerned about the amps 8 ohms sensitivity from the last test. But thinking more, of course the 8 ohm run is superior to the high impedance grids... ::)

But all in all, this episode has caused me to beef up a lot of things that wud have not shown themselves without the arcing. And many of us have probably learned a few things too... ;D



T

tnx Bruce; T if you run shielded audio cables (which you should do) you might try grounding the shield at only one end of the run; leave it open at the other end.  I assume the s.s. amp is not built for broadcast tx site; couple things Orban did on my 422A compressor is put small disc ceramic .01 mfd caps to ground on the audio leads inside the cabinet right at the terminal strips on the back panel and small ferrite beads on the audio leads.   I think I read somewhere above the low level audio is driving a transformer so that should give isolation to the rig; if the s.s. amp has a three wire balanced output it would good to use shielded balanced cable to the transformer--ground the shield at the s.s. amp and connect the audio + and - to the primary of the transformer.

73

Rob
K5UJ

OK, Rob, will keep that in mind when I put in the new SS amp.

New?  Yes new.  I moved this one to the new location and wired it up. It appears all the problems were gone and the rig was very stable. Then being tired, I finalized the wiring and put the wrong cables on the SS amp's output - dead short.  It didn't do anything except the fan came on after 20 seconds. Then the audio power output  slowly dropped off and never came back. Smoked after all this effort. Right at the end of the job too. Son ofa bitch.

So, tmw I'll hit a few pawn shops and find a used 100 watter to replace it. I'm pretty confident it will work FB now. The old one was just too close to the RF field.  I wonder what type I shud look for to avoid the coupling capacitor problems again?

I will use the new one and maybe later build up that 807 driver.

Happy days.

T

Opcom, you were axing about the NFB details via PM. Here's the schematic below. The NFB circuit is from the WA3KLR cathode driver.  I use the resistive ladder of one modulator plate cap tapped at the bottom as shown to derive the NFB. The audio and NFB are mixed into an IC and this output is fed to the audio amp at about 1 volt.  Works FB.

T

Tom that SUCKS reminds me of the time I worked all tksgvg weekend putting up an antenna and on Sunday evening finished, was so tired and excited that when I went to put the swr analyzer on the feed in the basement I dropped it on the concrete floor.
I bet we have all had something like this happen when tired, excited and eager to see results.

R.

I bought a samson 2x120 watt amp from some on-line music place, the same place I got the berringer box from, it was about $120.00 I think, and was also one rack space high.
It worked fine, but the symetrics had a bridge switch which dualed up both inputs allowing me to split the line level signal to both amps but with level control on each amp, so I kept that and sold the samson amp (cheap).

The combo of a berringer (one rack space, and a power amp (one rack space) makes for a small, compact setup that can sound very good, sit it on the operating desk and it takes up very little space, and you have the adjustments at your fingertips, and can watch all those purdy lights flash.
Stick it in a nice cabinet with the mod monitor and it makes things nice.

In my shack, I have the berringer and amp in a cabinet above the homebrew receiver, the mod monitor and marantz audio amp are in another cabinet.
I got the cabinets (3 matching) from mouser, they were called econo cabinets and were not expensive.

Brett



 

Tom

I have a Samson Servo 120 Amp (puts out 120 watts into 8 ohms when operated in the bridged mode) that is surplus to my needs. I use it to externally plate modulate (via a backward connected Hammond output transformer) my Johnson Ranger and other rigs as well. I will make you an offer that you can't refuse. Contact me here: sdp2@verizon.net.

Hi Stu,

I sent you an email, thanks.

Aw, that big old 300w+ amp was so darn heavy and needed retirement anyway.. :-) 

The arcing did uncover a lot of weaknesses in the rig. I blew out the Heising cap the first hour. The neutralizing cap was arcing over. The Heising cap mounting was arcing. I added needed spark gaps. And there was more. But finally, the rig handled the arcing FB. That was a big acid test. Now that it's calmed down, it will loaf along nicely.

That's the similar way I test out my gear anyway. After it's working, I usually load the rig up to max and beat on it. This shakes out the weak components. As long as the mod iron is not at risk, it is a way to eliminate years of future dicking with failed parts. I hate babying a rig until it has proven itself. Then it can have a cushy life.

I can at least use the rig on CW in the meantime... ;)

T

Tom,
I have some 4 uf mylars at 400 volts somewhere in the cave if you need any.

???  Not sure what I'd use them for, but tnx for the offer anyway, Frank... ;D


Stu/AB2EZ  just gave me the best deal I've had in years on his Samson SS amp...and is shipping it out today! I'll mention more when we get on the air. No better friend than that. Thanks again, Stu.    So I'll be all set in a few days for Round II.  Stay tuned.

Again, thanks for all the help, guys!  One way or another the problem was identified and solved. Great group and resource for help here - a homebrewers delight for sure.

T

Just looking at a few posts earlier and maybe hard cut-off of the modulator pubes would be the answer. I had that arrangement in a HB outboard 810 modulator for the T-368.

Fred

Good work Tom.
Just a quick note about my amp. Its a QSC RMX 850. I use it to modulate my Valiant and Ranger using backward connected Hammond output transformers. To drive grids its way to big but I believe in headroom! To big is much better than to small.
As I said, this amp is bullet proof.
http://www.qscaudio.com/products/amps/rmx/rmx.htm (http://www.qscaudio.com/products/amps/rmx/rmx.htm)  

Yep, nice looking 850w unit, Bill.  I can't believe they have a 2400 watt version that's packs into only 2 rack spaces. I can imagine making a mistake with that thing - driver transformer - oh the humanity! ;D

Stu's amp is on its way and I think 60w per or 120w bridged will keep me outa trouble. The specs these days are unreal and put the older tube amps to shame. Things like 3hz - 50khz with .05 % THD... unreal.

But I'm still interested in giving that 807 driver a good look at when the schematic gets posted.

Right now I'm making some more mods to the 4X1 rig to insure the safety of the new amp. Can't wait to get on wid it soon.  ("Calling the 4X1 net")


T

My RCA KW iron has factory gaps on both sides. Maybe having only one gap is an engineering philosophy or a cost saving function.

I'm not sure if you solved the original problem or not...

How much peak to peak grid voltage is required for your modulator tubes?

If it's less than about 700 volts p-p, you could easily use source-follower MOSFETs to drive the grids - eliminating all of the driver phase shift (because no transformers would be involved), and providing the absolute lowest driving impedance possible (less than an ohm).

The circuit would be fairly straight-forward.

Also didn't get whether you are switching the high voltage on and off, or are you leaving it on all the time.  I tend to kill the HV supply at the primary and allow it to fall to 0V (or a very low voltage), and use a step-start on transmit.

Anyway, just a suggestion  :D   Geesh, a pair of 4X1s modulated by a pair... guess it's time to start building that 100 pill rig  :P

I wonder what blew in the original 300 watt audio amp.
I would have to tear it apart and find out.

I also do a step start on my rigs, although I know its better to leave the voltage on all the time.
When built, most of my designs came out of the Bill Orr books...

It works, big diodes, choke input, even with a lot of energy storage...

Brett
 

I see the entire unit is to drive class B audio grids, Indeed, the 1 volt to feed the commercial SS amp could be picked off early. Thanks!

Hi Steve,

There's a lot to read here, but look at the pics and you'll see it's a SINGLE 4X1 X mod'd by a pair.  With only 3500v on the mess, the modulators can only handle one final. But that's plenty...  ;D

The HV supply idles on all the time, but I key a HV relay to route the HV to the particular amp I'm using - mainly for safety reasons to keep the HV off the rigs most of the time.

The mosfet grid driver - I've already done that years ago with the old K2IJY board - too many supply voltages needed and certainly takes some heat and power to run. The WA3KLR MosFet cathode driven circuit is a much better approach.  _+ 12V and a 5v supply to power 5 watts to drive a pair of 833A's in the cathodes.   I still have the KLR finished board here and might even try it wid the 4X1 modulators, but am leaning towards building the BC 807's direct coupled tube driver we've been discussing here.  For now, using a SS amp wid backwards 8 ohm scheme.

I posted Tom/KLR's circuit below if you want to refresh. I'm presently using the NFB circuit in it for this rig.

I figgered you'd have that 100 piller finished by now! ;)

T

Close.... the board drives the cathodes of 833A's or whatever tubes you pick - I used it with triode connected 813's and the audio was flawless. Best rig I ever saw plate modulated.  It takes about 5 watts! of total power to drive the 833A's. I'm surprised more guys have not picked up on this clever design to drive their modulators. The only drawback I can see is it takes TWO separate fil transformers for the modulators.

T

I really admire the fine job you did on the transmitter!

BTW this whole thing encourages me to do something really scary with loose parts around here.
Four 3-1000's
Gates by Dahl BC-5B hypersil HD mod xfmr
UTC CG-309 plate unit, 3KVDC/1A CCS (heh maybe not enough)
I just do not have the time, I'm afraid. I swore to fix what I have first. Parts is parts.

But thanks for the encouragement!

Hi Tom,

That old K2IJY board was/is a very bad (and unauthorized) implementation.  It came about as the result of a telephone conversation discussing the concept of a driver, and not an actual circuit.  No one would do such an implementation in the practical world.  I have a real-world MOSFET source follower driver circuit that I may post here to show a proper implementation, and something that will work correctly, and with proper arc-back protection and other niceities.

What you use will surely depend on how much grid swing you need.  What is this value?  I don't remember if you are using class AB or class B; triode or tetrode connection.  This will make a huge difference in the power output and the drive requirments.  I generally assume a high grid-to-grid swing is needed, particularly in triode connection - and also assume the grids are driven positively with respect to the cathode, necessitating a low impedance driver.

The cathode driver schematic you posted appears to be capable of 100 to 150V G-G (ultimately).  The DC voltage applied to the tube GRIDS and the MOSFET device ratings will be the major determining factors.  With 0Vdc at the grids, you will be unable to drive the grids positively - which may be OK, not knowing all of your operational parameters  8)

However, if you need a high grid swing, I do have a circuit which will deliver the goods.  There is a 450TH modulated by a pair of 833A's getting ready to come on line (no, it's not mine!!), and this transmitter will use this driver.

 ;D Pills - I've never made it too far.  The 100 pill transmitter has been in my head for years, but I've been too complacent - need a "mark" so to speak  ;D  to spur on the implementation effort.  A pair of 4x1s would definitely have done it!

Talk later and Regards,

Steve

Steve,
Pretty easy to change the cathode swing with a higher pull up voltage and positive bias on the grids. (and enough heat sink).
I know what you mean about motivation.....get the big rig back on or put down beautiful white oak flooring at GFZ South.

Exactly !!  I like the cathode drive method within certain parameters.  One has to watch out for capacitance to ground issues with the heater transformers, but it shouldn't be too bad.  Applying a positive voltage to the grid will allow the grid to be driven positively with respect to the cathode.  

The major down side is the fact that quite a bit more power is dissipated in the driver than would be otherwise be dissipated using grid drive.  This is due to the fact that the cathode current flows through the driver MOSFETs all the time, and as the max grid-to-grid voltage is increased, the driver losses increase as the square of the voltage.

Hey, it's winter - time for HAM RADIO PROJECTS !!  Get that rig on the air  :D

Regards,

Steve

Agree, but it is a nice place to pick off NFB at a low voltage using the source resistors.

I put down 3 X 41 foot length of flooring. Looks so much nicer than green plywood. Afterburner is on the bench and drilling blasting is going on to add vacuum variable caps.

Wow,
-wish I had a 3 x 41 ft. shack!  ;D

Guess all your stuff is 'hairpin' matched too.

Maybe Frank's talking about a new "alley shack" behind the house? - I dunno... ;D

Steve: Yes, I'd be interested in seeing the MOSFET schematic. I'm not sure how much grid swing is required of the 4X1's in triode-connected service.  I've never seen anything published.  But certainly it can't require more than 833A's.

For the cathode driver I put 50V positive on the triode-connected 813 grids to get all the headroom I needed. Without this bias, there was not enuff monkey swing.  I'll get by with this 8 ohm transformer scheme to get on the air, but will have to build up a transformerless driver later - and choose between the SS MOSFET stuff or the 807 direct coupled design. No rush.


Opcom: Two 3-1000Z's X two?  Those tubes are like gold - do you actually have some? I'd rather be using 3-1000Z's for modulators, if they were common, but I think the triode-connected 4X1's are pretty close in performance from tests.

T

The south wall of the house is 41 feet wide. I covered 3 feet with flooring with 27 feet to the north yet to do.

Frank, How are the knee's holding up?

Ok, triode connected tubes... how are you connecting them?  Grids together?  Control grid at AF ground and drive the screen only?  There are a number of possibilities.

Let me know  ;)

Steve,
Will u PLEASE stop trying to shove this SS stuff down the throats of the toob guys.....u are polluting the toob gene pool.... ;)

ONLY KIDDING guys.........

Bill

Yep, as I always connect 813's or 4X1's:

Screen and control grid are tied together and driven together.  The bias is supplied by putting diodes in series with the filament center tap. No bias is needed at 3500V, so the fil CT goes directly to ground.

I'm driving the tubes now with a solid state 8 ohm amp/backwards xfmr and from the amp LEDs lighting up, it looks like maybe 30 watts of audio are needed to drive the p-p pair of 4X1's. Then again, this is in the low RF power tap and need maybe 60w of audio later.  But I cud make some careful voltage and power measurements if you need.

T

Bob,
First two courses were by hand on my knees so Sunday I was pretty crapped out.
Taught my son to operate the nailer by Saturday so that made it easier. White oak is tuff stuff and wider width made it worse.
Drilling and blasting the rig would have been more fun...but just before I left I sanded it and put down a coat of varnish. Looks rally nice. Now I know why old ships were made of white oak.

Tom,
I would think a triode connected audio 4-1000A would take the same grid swing as a GG linear needs cathode swing. Say it needs 100 watts of cathode drive  then E^2/50=100 so E^2= 5000 so E = 70.7 VRMS  70 X 2.8=198 V P-P

Tom,
Let me know when your gonna be on 75 with the rig.........I would love to hear it and record it also.
Thanks,
Bill

Tom

In an application like this, the backward connected audio output transformer may be best modeled as a voltage step-up transformer:

The solid state audio amplifier behaves as a voltage source (unless you place a very low impedance across its output). I.e., in bridged mode,  its output voltage swing will be (essentially) the same whether you put 8 ohms or an open circuit across the output terminals.

As an approximation (below) I will assume that the equivalent circuit (in bridged mode) is an ideal voltage source in series with a resistor whose value is "less than 8 ohms". However, one could probably assume that the equivalent source resistance is much less than 8 ohms, provided one does not draw too much current from the amplifier.

Since the nominal load is 8 ohms, and the nominal peak output is 120 watts (in bridged mode), the nominal peak output voltage (red-to-red output terminal in bridged mode) is the square root of 120 watts x 8 ohms = 30 volts (independent of the load, as long as the load is 8 ohms or more). The associated, maximum, peak-to-peak output voltage is 60 volts.

Note that the B+ and B- supplies are ~ 18 volts... so this is consistent. Also note that when specifying the peak output power of the audio amplifier... the manufacturer is defining it as the absolute peak output power (not averaged over a cycle of a sine wave audio signal).

If (for example) the audio output transformer has a turns ratio of 16:1 (i.e., the output winding has been optimized/designed for driving an 8 ohm load, and the input winding is optimized for being driven by a source with a 2048 ohm source impedance... so the transformer is specified as: "8 ohms on the output winding" and "2048 ohms on the input winding"), then in this mode of operation (backward connected), the output of the transformer will look like the following (as an equivalent source of audio)

Peak output voltage = 30 volts x 16 = 480 volts (960 volts peak-to-peak)
Effective source impedance= less than 8 ohms x 16 x 16 = less than 2048 ohms

If you don't need such a large voltage to drive the grids of the 4-1000's, you can reduce the effective source impedance at the output of the audio output transformer (and also improve its frequency response at low frequencies) by using only one of the outputs of the solid state amplifier. Each output (i.e. stereo mode rather than bridged mode) has a source impedance that is less than 4 ohms ... so the effective source impedance at the output of the transformer would be less than 1024 ohms (in this example). This will reduce the peak output voltage to 240 volts (480 volts peak-to-peak)

Even better: If the transformer has a 16 ohm winding, use that (with the amplifier in bridged mode). You will reduce the peak output voltage to .707 x 480 volts = 340 volts (680 volts peak-to-peak);  and you will reduce the effective source impedance at the output of the transformer to less than 0.5 x 2048 ohms = less than 1024 ohms.  

If you drive the grids of 4-1000's positive, with respect to the cathodes, then the solid state amp will deliver power. Otherwise, it will not. The nice thing about the solid state amp is that it is a voltage source... and its output voltage will not be affected (significantly) by the changing load impedance presented by the grids of the 4-1000's (as seen through the backward connected audio output transformer).

Best regards
Stu

Back in my "destroy the ears" days (read, car audio installer), we also found that by bridging our amps to mono, and driving a higher impedance, we could get the same power, but THD dropped like a rock.

Not that we have to have .005 THD for an AM transmitter, but it's NICE to have clean audio chains :)

--Shane

Thanks for the info, Stu -

Here's the impedances and specs of the audio xfmr I'm using.

It's a 62 watt 20hz-20,000 xfmr.

The windings on the secondary are 7,000 ohms CT.

The primary has one large winding with 80 ohms CT and 265 ohms CT.

I am feeding amp between the 80 ohm and CT  - and across the 7,000 ohms CT to the grids.  This is the highest step up I could find on this transformer.

The 80 ohm to CT is less than 80 ohms cuz it uses 1/2 the winding.  The overall ratio seems to work well, though it's not a true 8 ohm to whatever transformer.

The transformer is about 10' away from the 8 ohm amp output. The transformer is about 4' away from the 4X1 grids. The tubes idle at about 100ma each at 3KV - I plan to raise this to 3500V later on. I think the idle current will be fine with zero bias.
Any suggestions?   I don't want to buy another transformer at this point since this one works FB.

Bill, yes it wud be nice to hear a tape of the rig. Shouldn't be too long now. I'm trying to get minor details worked out now like NFB, etc. I tried Stu's amp last night and it basically solved the problem. Sounds good and now the un-key spike is about 40ma - hardly noticable. I'm going to work on that some more too. There is absolutely no sparking in the gaps at all and the rig is very stable overall for both RF and audio.  The other amp was just too close to the RF field - now the new amp is 10' away and loves it.

T

Stu,
I'm thinking it would be a good idea to use a little resistance in series with the 8 ohm secondary since it is further away from the transformer core. When flipped around as a  primary, I bet the z is lower than 8 ohms. Triode connected 4-1000A at 3500 volts should cut off between -20 to -30 bias so the tubes should load the transformer pretty well.

Tom,
80 ohms you might take a look at the wavefrom. It may want a shunt load  depending on the SS amp. 16 ohms in bridge may be happy to drive it.

Frank, it's not 80 ohms - I'm feeding 1/2 of the 80 ohm winding single ended = 20 ohms. It's an 80 ohm, CT winding and I'm feeding 80 to CT.

What does that make the impedance of the input and the overall ratio to 7K?

T

Tom
Et al.

As I mentioned, thinking about impedances may be more confusing than helpful... in this application.

If you are using half of the 80 ohm winding, then that is equivalent to using a 80 ohm /4 = 20 ohm winding... in the sense that: if you put a 7k ohm load on the 7k ohm "output" winding, the resulting input impedance (looking into the input winding) will be 20 ohms.

But, if you just think of the turns ratio (which is equivalent to voltage the step-up ratio)...

Using half of the "80 ohm" winding, and all of the "7000 ohm" winding... the step-up ratio is 2 x the square root of [7000 / 80] = 2 x 9.35 = 18.7

This is a good number. I.e., the peak output across the output winding (not from one side of the secondary to CT) will be (for this amplifier, in bridged mode): 30 volts x 18.7 = 561 volts. The peak voltage from one side of the secondary to CT will be 561 volts / 2 = 280 volts.

The output of the transformer will have an equivalent source resistance (from one side of the secondary to CT) of less than 8 ohms x 9.35 x 9.35 = 700 ohms... and that is surely be low enough to serve this application. [The grid-to-grid equivalent source resistance of the output of the transformer will be 2800 ohms... but I think it is less confusing to think of the grid to CT equivalent source resistance]


If the transformer were used as an output transformer: driving an 80 ohm load with 62 watts of peak power, then the peak current (not the rms current) would be the square root of [62/80] amps =  0.775 amps. Saturation effects will come in when the product of the current and the number of turns exceeds some critical value. With only half of the 80 ohm winding being used, you need to put twice as much current (to produce 62 watts of output), but you have half as many turns... so the peak power you can run stays the same (from the perspective of saturation effects).

As you are using it (as an interstage coupling transformer... driving the grids of your modulator's output tubes), I suspect that you will not cause excessive heating (resistive losses) of the 80 ohm winding (or the half of it you are using) because the average power will be much lower than the peak power. However, you should be careful testing the rig at 100% modulation with a sine wave applied for a long period of time.

By the way... some time ago I posted the .wav file of a test waveform that puts very little stress on the components (low duty cycle); but still modulates to up to 200% on positive peaks. It consists of a short positive-going pulse, followed by a short negative-going pulse (having half the amplitude and twice the width of the positive-going pulse), followed by a period of 0 volts. It repeats every 200 milliseconds (if I recall correctly), and has a duty cycle of about 10%.

Stu  

  

Why wouldn't K1JJ's solid state amp be considered a current source at 8Ω, even in this application?

Bob

The way the amp is designed (using feedback), it delivers a specified voltage into any load... provided the current delivered to the load stays below a maximum value that is set by the overload control circuitry.

For example, for a given audio input signal, it will deliver the same output voltage whether the load is 8 ohms or 16 ohms or an open circuit. Therefore it is, by definition, a voltage source. Except for the maximum current (overload) constraint, its effective source impedance is essentially zero... but, for simplicity, I have approximated the source impedance as "less than 8 ohms". 

An amplifier designed to work as a current source would deliver twice is much voltage (for the same input signal) if you doubled the output load resistance.

Stu

Hi Stu,

Good on the transformer calculations.  I understand - and have the rig running very well now. It took some time to get the NFB right, but I'm real pleased with the response and power of the modulator system now. Your amp is working FB.

OK on the pulse tester. I've been looking for something like that for my linear amp ssb tuning. I don't ever use a carrier to tune up - rather a few "chhh-chhh's" into the mic to keep the audio duty cycle low. I will play your .wav and tape it into a recorder for use on both AM and ssb.  Tnx.

I'll do a search for it now.

I've also heard those "chirper" units that sound like telementry some ssbers use. If seen them sold in a little box. I'd like to build one of those too. Probably a simple 555 wud do it if set up right for narrow pulses.

Later -

T

Quick question Stu,
I thought if using a backward connected transformer, when the grids draw current on positive peaks.....the Z will drop and the transformer output will drop causing some distortion on the audio output of the modulator. Im not trying to split hairs but just wondering. When I went to the cathode follower, because of the low Z of that ckt., it could handle the varying Z changes a lot better.

And Tom......I use my WinRadio SDR receiver for recording...... no tape.......its digital to digital.....

Bill

Question:

The recommended bias in class C for a 4X1 is -200V.  I have a fixed -90V and an additional grid leak of any value I choose.

I tried a 5K leak resistor that gave me about -315 total grid voltage. This was at 40 ma of grid current.  (Deep class C)

Then I tried 10K which gave me about  -600 V grid bias.  (Into class D?)

Both seemed to modulate well.

I'm curious - Does the grid of the tube HAVE to run 40ma? What if I ran it at 20ma with the 10K resistor giving about -300V? Is there some reason the tube needs 40ma?

The efficiency seems to be very high now in class D. I did the math, but it seems too high to even say, so I won't ... :-)

BTW, I just looked up the max grid rating and it's -500 V, so guess I'll stick wid the 5K and 315V for now.


Bill: The last recordings you sent me were very good - I was pleased the way I sounded for once... ;D

T

Tom as duty cycle goes down when you bias further into class c the efficiency goes up, but you need to maintain grid drive to saturate the tube when it is kicked on.
A third harmonic tuned circuit in the screen will drive it even higher. I read somewhere 85% is possible
A backwards transformer will have a good portion of the core air since it is sitting on top of the other winding. This will reduce the inductance of the winding. Sounds like the 20 ohm winding worked well at 8 ohm source but may end up a bit low at 16 ohms bridge. You will have to look at it to be sure.

I think 99% for sure there are four. In return for fixing a computer, I was given three 13.5MHz 1500W RF generators. One was said to work and has the remote control, all had the big tube and associated parts. 208V filament transformers though. I also have a spare tube. Amperex, very badazz looking. The tubes, the two I have pulled anyway, appear to be like new, because these generators were under maintenance and the tubes were replaced well before lifetime. Cheaper to replace a 3-1000Z than to put a semiconductor line down. I parted one of the not-working generators, the other 2 are left intact.

I know they are costly little beasts. Last I saw was $650-$1500 depending. I was shocked the guy only asked me to fix his computer in exchange for them and I tell him, well, you know those tues are worth quite a bit, are you sure you want to trade all this for just fixing your computer? - I discovered the issue was that it was his XYL computer, and had somehow been corrupted by viruses and worms from "Asian dating sites" so he said, and that this had so far been successfully blamed on his teenage son, and that he needed the computer totally sanitized asap to be presented back to the XYL in proper and non-filth-laden working condition.

What I hate is when you shop for items like this and it says "call". What that tells me is that they do not have it and will act as a broker to get it for you, or will decide the price after you call. I just lose trust there and go to the next vendor.. Does anyone else feel the same way about dealers 'hiding' the prices from prospective customers, or is there a good (honest and customer-beneficial) reason for it?

On the other hand I thought the 3-1000's would be good in case I ran out of 4-1000's for the Tucker xmtr. Which is not likely to happen.

Well back to it I do have 4 of them with all the trimmings and I am not likely to use the set.

A very nice triode....

Yep, I wud pop those 3X1's in as modulators in a second.  Use a 4X1 (tetrode) as the final and you're golden.
Not a bad deal for fixing up his computer... ::)

Well, after some more extensive testing and mods tonight, I'm ready to put this baby on the air maybe Thurs or Fri.  All systems seem acceptable.

I've decided to go sailing with Yaz tomorrow. It's gonna rain in the morn, but then clear up with high winds and 60F temps. Probably the last sailing of the season. Been working too hard and Yaz deserves a good outing. I like to land a lot and he jumps off, sniffs around, gets back on, we raise the sail and off we go to the next destination.

I hear Derb on tonight at 45-50 over. Nice sig Derb!  
Update:  I just gave Derb a call and made him #1 contact on the new 4X1 rig. Tron was #2.  It's nice to hear ya back on OM!  

T

Bill
Et al.

A lot of our intuition (including mine) is based on assumptions that are outdated. In the past... when we designed an audio system to drive a given load... we designed it so that the output impedance of the audio system was matched to the load impedance. This produced the best possible efficiency.

Now, we can buy a 600 watt audio amplifier (if we wish) with essentially a 4 ohm output impedance (or less). If we hook that to a 20:1 step up transformer, we can drive a load whose impedance varies all over the place... as long as it stays above 4 x 20 x 20 ohms = 1600 ohms.

The audio amplifier doesn't care if the load is infinity or 1600 ohms.

The "price" we pay is (perhaps) relatively low efficiency. My amplifier may be consuming 50 watts of average AC power from my 120 VAC outlet... and the average output power may be 1 watt (or less).

Tom is using a step up transformer that he has on hand.

In my shack, I am using a $10.00 (~current price for a new unit) toroidal power transformer from Antek to externally plate modulate my Ranger in Heising configuration. The Antek transformer has a pair of "120 volt" primary windings and a pair of "12 volt" secondary windings. Thus the step up ratio can be set to anything from 1:20 (12 volts to 240 volts) to 1:5 (24 volts to 120 volts). I'm using a 1:20 ratio to drive the Ranger at full output power: 30 volts from the audio amplifier produces 600 volts of output from the Antek transformer. The bandwidth is flat over the relevant frequency range (20 Hz - 10kHz).

Since the output impedance of my 150 watt per channel audio amplifier is (much) less than 4 ohms, the effective output impedance of the toroidal transformer is (much) less than 4 ohms x 20 x 20 = 1600 ohms.

If I were using this transformer to drive the grids of a modulator, I would use a step up ratio of 10:1. In that case, the effective output impedance of the transformer would be (much) less than 4 ohms x 10 x 10 = 400 ohms.

A source with a (much) less than 400 ohm output impedance would not "care" if the load varied from infinity to any value greater than several hundred ohms.

Best regards
Stu

Tom

Regarding grid current and self biasing resistors:

I think it is best to think about this by starting with your principal objectives

A. You want to achieve a specified output power
B. You want to operate the tube(s) in a manner that is consistent with their rated plate dissipation
C. As a matter of pride, "greenness", and the cost of electricity...you want your high power rigs to run at reasonable efficiency when they are putting out high power.

The following things are not objectives... they are just things you need to do to achieve your objectives:

a. For the tubes you are using, you need to drive the grids positive with respect to the cathodes for a (small) portion of each cycle. If you don't, you cannot achieve the objectives above

b. Given the choice of where you want to bias the tubes (to achieve the efficiency objective and to meet the plate dissipation constraint) there is a certain grid current pulse (amplitude and shape) that will occur each cycle.

c. Each grid current pulse has a certain area (units of Amperes x seconds = Coulombs). The average  grid current is just the area of a single pulse of grid current x the number of cycles per second (units = Coulombs per second = Amperes)

Thus the average grid current you need is determined by: the tubes you are using, how you bias them (for efficiency), and how much output power you want to obtain.

Separately, there is a question of what combination of self bias and fixed bias you decide to use to bias the tubes.

If you want to depend entirely upon self biasing, then the grid resistor value you need is completely determined from a) the average grid current you need... as described above, and b) the bias voltage you need... as described above.

If you want to use some amount of fixed bias (or cathode bias), then the remaining bias voltage and the average grid current determine the needed resistor.

In my Ranger, given the choice of a 6146 as the output tube... Johnson designed it to be biased in Class C (of course), at around -70 volts grid-to-cathode.  Given this tube choice, and given this biasing choice, and given the choice of a 600 volt plate voltage: it follows (from the tube characteristics) that about 2.5 mA of average grid current is required to obtain the desired ~50 watt output power level (at carrier). Since Johnson decided to use only self-biasing, the required grid resistor is 70 volts/2.5 mA = 28,000 ohms

Stu

I didn't sift thru all the thread Tom, and maybe it's been said but;

Why not put the SS amp somewhere else? Put it near your operating table and run shielded cable from it to the bassackwards transformer. Sounds like an induced problem to me. Relocating the amp might be the simple thing. But, there again, I'm just a buddly.

OK Stu -

Good summary on class C.   I've pretty much settled on 5K grid leak which gives me the recoomended 40 ma grid current at -315 total bias. They recommend -200V, so it's well into class C.  I see no plate color at all at full power.

Yes, Buddley, moving the amp was the final solution. I blew it out right after moving it and found it worked FB -then blew it by wiring the output into a short. I was tired and screwed up. Stu sent me another amp and it works fine now. Both RF and audio keying is pretty clean. I still have a tiny pulse on unkey, but it appears related to simply picking up the audio transients on the low level audio line, since I can disconnect the input audio and it goes away.  No more arc gap sparking like before.

Should be on with it over the next few days for testing. I'll be fishing for "opinion" audio reports, mainly related to EQing, audio quality and bandwidth.  If ya hear me, please give me a call.


T

Tom

If I catch you on this weekend (including Friday), I can make a recording with my Flex 5000 and send it to you.  There are others too like Bill IFR who can do the same thing

GL, Al VTP

Tom,
I will also listen for you. I'm home today XYL under the WX but I will be in the shack later with the RX on 85. I'm doing some metal work on my rig. I used this hard 1/8 inch aluminum for the chassis rear deck and need to file .030 off each side so the vacuum variable holes end up in the right spot. Too thin a cut for a saw.

Tom,

you was strapping pretty large here, audio to me sounded super clean but it also sounded kinda flat, meaning it didnt have a lot of
color to it. Matter of eqing and getting some peaks and dips in the right places. I am using a big hi fi speaker with the yeasu, a mid 70's 3 way with a 12" woofer, 6" midrange, and 3"tweetie. It's good to go 20hz to 20K.

 

I use the same type of speaker, 12 inch 3 way with the Marantz audio amp.
My son found the pair of speakers in the trash, some new meterial on the covers and they looked like new.
Big, heavy old school construction...

The Flex 5000 into the Marantz, into the big speaker sounds fantastic, the best cleanest hi fiest receiver I ever heard.
I often leave the system running as I do things, or watch TV in the den (keep wife company) and I had rumbles from someones voice the other nite, they were using a flex and had the low end boosted I guess, it sounded good, and rumbled as well....

I cant understand how the Marantz still works, its old as the hills, and besides some dirty switches (fixed with deoxit) it has been trouble free for the last 30 years of constant use....they dont make stuff like that anymore...

Brett

Well, while we are at it.  Here's my RX / AF system  -  My Flex 5000 into a KLH R3100 100 / ch receiver and then to MTX TP112 speaker.  You HIFI AM guys really drive this RX / AF system FB

Looking forward to working you, Tom

Al

I have a Marantz 2275 receiver.  It's a 70's vintage SS receiver.  I picked it out of the trash back in the 80's.  It looks like new still. Had to do a few repairs over the years but it works OK fine.  Have the original wooden cabinet. It's a nice piece of furniture. And it's a heavy mother.  Works like a charm today.  It's part of the home audio system and  loves the old LP's.

T,
Hope to hear the 4x1 in the next few days. Maybe the 2 of us can critique and adjust the audio of our rigs under full quieting mode since we're so close.
B

OK guys, thanks for the offers.

Looks like we will have some good, diversified gear and listening ears... ;D

My objective is to take my average voice and try to get the "Big Boss" AM sound in the low end with the highs rolled off starting at about 5KC. So it will be an effort to get on the edge before the low end starts to get muffled - while the on-air signal  bandwidth stays within reason on the highs.

I have a variable cut filter that is good at rolling off the highs, but I am more concerned about generated IMD garbage. Hopefully the signal will tune and end with wisps and not harsh splatter sounds.

Getting the EQ balance right does take some time. But I know there is a certain combination with my voice where most everyone likes the sound and so do I.  Some guys have great voices and need little tailoring. But mine has mid-bass that needs to be cut on the EQ.

I'll be running the rig at low tap for now until I can make sure all the bugs are tamed.

I'm also looking for just a touch of blower noise, cuz it is there, but not much. It's a challenge to keep it low with the compression running. It is faint with the compressor off, but then there is no audio limiting for transients or mistakes.

So, that's what I'm trying to achieve so you'll all know when you hear it.

Thanks.

T

Hola!

Found out something new.

I swept the rig again tonight and found the highs were rolling off and getting distorted above 4 kc. At 6kc they looked terrible. I didn't take the time to look closely before.

The bottom line is I took a hint from the Tron's old saying: "Use the WHOLE transformer winding."  For the 8 ohm input I was using about 1/4 of the winding = 20 ohms. This time I went into the whole winding = 265 ohms. It made a huge difference. The highs are not perfect, but pretty clean up to 8kc or so now. I'm still using the whole 7K ohm secondary as before.

Logic would tell you to use the biggest step-up in this case, but was wrong. Now the amplifier uses less power to drive and the unkey pulse is even less, two good things.

Last night during operation, I couldn't understand why the rig would not do over 50% modulation on the S's.  Now I can easy over-modulate the highs. I located the culprit (driver transformer) by putting a tone through and methodically looking at the waveform in each component, from the low level stuff all the way to the mod transformer. The driver xfmr stuck out like a sore thumb.

Eventually I'd like to try an 807 tube driver or a MOSFET driver, but this SS driver and xfmr will suit me fine in the meantime.

Glad I found this problem or the reports wuda been disappointing, I'm sure.  Derb was right when he said my audio "lacked color" last night.  Poor/no highs will do it.

T

Hi.

Yes logically we needs to used the highest turn ratio for the interstage transformer.

 But I think its not always like that,when You used the 20 ohm winding and 7000 seconder ,it's impedance ratio of 350,and the turn ratio is 18.7

If the  the grid to grid  of the Modulator tubes has a 500 ohm impedance (for instance) than your SS amplifier looks into a 1.4 Ohm load,and if it needs a 200 v peak to peak grid voltage  it needs a 11 volt swing in the SS output, it seem not right to me,but maybe I'm wrong.

I think there must be a compromise between the turns ratio  to have a good drive for the Modulator tube.
and ,I think we needs the right turn ratio to have a good signal/power for driving the Modulator Tubes.
It depends on the grid impedance of the Modulator tubes and the impedance of the amplifier,of course SS amplifier can have different load and still gives a good sound.
But I think there's a limit to it.

Gito

Tom

In this new configuration... the step of ratio is lower... so I am surprised that you have enough voltage to fully drive the grids of the 4-1000's.

The high frequency rolloff you observed suggests that the transformer has too much leakage inductance, particularly when used with only a portion of the low impedance winding. The leakage inductance appears as an inductor in series with the winding.

I would suggest that you purchase one of the Antek toroidal ferrite power transformers. They work as almost ideal audio transformers (in the range 30Hz - 10kHz)...  as long as you run them in Heising mode. I'm using a small (25VA rating) Antek transformer to externally plate modulate my Ranger in Heising mode. I'm using a pair of large (800 VA) Antek transformers (in series to get the step up ratio I wanted) to plate modulate my legal limit homebrew amplifier (a pair of GS-35b's in parallel, running at 1700 volts B+)

Pick one with a pair of identical primary windings, and a pair of identical secondary windings ... so that you can place the windings in series of parallel to obtain a step up ratio of X, 2X, or 4X.... where 4X is the step up ratio when input windings are in parallel and the output windings are in series.

Pick one with a VA rating of at least 100VA

For example: the transformer has a 200 VA rating, and it will give you a step up ratio of 230/15 = 15.3, or 115/15 = 7.66, or 115/30 = 3.83

http://cgi.ebay.com/15V-15V-30VCT-200VA-POWER-AMP-AUDIO-TRANSFORMER_W0QQitemZ370192779213QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item56313587cd

Stu

Hi

One  thing not to forget ,a 100 watt ampli design for an 8 Ohm load ,if i'm not wrong needs  a 3.5 A current (I X I X R), so it is design with a power supply with maybe 4 A rating.

When it is used for an 2 Ohm load it needs a 7 A current, Can the power supply gives this current,can the final  power Transistor pass this current?

 In my opinion used the lowest turn ratio that still can supply the peak voltage needed but with enough inductance in the interstage Transformer and again maybe I can be wrong.

Gito

Tom,

I was comparing it with audio I've heard you on with before from memory, not the best test, but "the JJ sound" is pretty firmly fixed in there. There's one more observation: it lacked a quality I've heard in the past for lack of a better term; it didn't have any 'projection' quality. It was very flat sounding, just like a good BC transmitter would sound with a maladjusted audio rack that has'nt been set up yet. Smooth as vanilla puddin, but there was just no bark to it.

I hope that my overuse of highly technical terms does not offend, OM hi hi fb.  :D

Tom,

   I agree with the Tron to use the whole winding, and with Gito's concern about the driver current budget. Now I know I'm paddling upstream with my suggestion, but the series cap between the driver AMP and the interstage transformer will still be helpful. Remember what Bruce said, a series cap shunted by a resistor (3-4 ohms)? I suggested you use the dual 1600uf/450v caps in series (++ together). Here is why, your voice does not speak in sine waves, and likely has a DC component to it, or asymetry. This DC component will see the DC load (primary of driver transformer), and 1 volt into 0.1 ohm is 10 amps of real current wasted. The caps/resistor eliminate the issue.

Good luck with things!
Jim
WD5JKO

"its better to have a bottle in front of me than to have a frontal lobatomy"

I think you just need some oxygen free cables between the audio amp and the transformer.

The correct driver transformer would likely fix you right up, 8 ohm to 5k or above, 20-20khz response, whatever wattage you need.

Does Hammond make one?

Brett

  Brett,

   A direct coupled SS amp looking into a DC short is a bad thing. In order for that to work, the AMP must have zero DC offset, and the audio waveform must have ZERO asymetry. Those two things are tough to achieve. I don't put the series cap idea in the catagory of oxygen free cables. Sorry.

Jim

Inductance too high hurts highs with excessive leakage inductance.
Inductance too low hurts lows. Sometimes you can shift things with series caps on primary or even a series resistor. also a grid to grid load.

Jim,  I will try the non-polarized caps again with the 4 ohm resistor across them.  Your DC asymetry point makes sense.

Yes, the highs were getting killed with the bigger xfmr step up. Even NFB didn't help much. The lows were not affected at all and still looked good. But now things look quite acceptable as an interim fix until I come up with another driver.

Stu, I'll look into the ferrite core idea. Also, I'm waiting for Bruce to post the redrawn 807 schematic and Steve to post the MOSFET one to get additonal ideas for the next step.

Derb - that's FB - that's what I'm looking for OM. You were right.  It's hard to hear the problem in the monitor, but when I fixed the highs, it was so obvious that they were lacking. Now I can hear more hi-freq blower noise and my breathing. Before it was muted.

More tests today into the dummy load and hope to get on for the weekend. Talked with Joe, PJP late last night but skip was long.

T

Interesting point Frank.

I wonder if I should put some more loading at the 4X1 grids.. like 10K to ground or across the winding to CT, etc?  Also 100 ohms in each grid lead for stability?  I have parasitic chokes in each plate lead.

Right now I have a 10K, 10w resistor across the xfmr secondary for safety. Also a 500 pf doorknob at each modulator grid pin to ground, that's all.  The 10K is wirewound - maybe this shud be carbon?

I do see a slight instability on the modulators when I unkey shown as some hash on the scope when watching the grid pins.

 Any suggestions for values, etc?
T

Tom,

The subject of using driver transformers for driving class B grids is a complicated one.

I have an article from the 1936 Proceedings of the IRE that goes into this subject at length from an detailed engineering standpoint. Unfortunately, the quality of the copy is poor and not suitable for scanning, etc., but there is a wealth of information on this very subject. It was entitled, "The Minimization of Distortion in Class B Modulators", or something to that effect.

There are two modes of thought with regard to loading the secondary of the driver xfrmr. In some cases it can improve the regulation of the driver, but at the expense of available voltage swing at the modulator grids. Others believe that the less loading on the driver tubes (i.e. no resistive loading), the better the driver regulation. I think it has to be determined expermentally, in each specific driver amplifier circuit.

As you are using a solid-state driver amplifier with a large amount of internal negative feedback, the regulation (or output impedance) of the driver is probably excellent to begin with, so the loading on the driver transformer secondary becomes less critical.

I think you need at least some resistance across the secondary of the driver xfmr, to minimize any ringing inherent in the xfmr. I shunt a 15K, 25 watt resistor across the secondary of my driver xfmr for this purpose, and it clearly makes an improvement. In the case of a vacuum tube driver, you also want to use the highest impedance available on the primary side (plate-to-plate impedance of the driver tubes) and the lowest available impedance on the secondary side, so the step-down ratio presented to the class B modulator grids by the driver xfmr is as high a possible. This improves the driver regulation and reduces the bias shift as the modulator grids are driven positive, but at the expense of available modulator grid swing. Obviously, this all has to be taken into consideration when designing the driver/modulator circuit.

I question how well conventional high-quality audio output xfmrs connnected backwards work as class B driver xfmrs. I did this for years and was generally satisfied with the results, but it may not be the optimum solution, particularly when a low impedance output tube amplifier is being used as the driver, due to the cascading of the two xfmrs and the attendent phase shift issues, etc.  Driver transformers designed for class B service must have a very low value of leakage reactance and a low DCR in the secondary winding, to minimize the effects of bias shift as the class B modulator tube grids are driven into grid current. Designing such as transformer is very difficult and costly as a result. And as I had stated in a previous post, when using some high-power solid-state direct-coupled audio amplifiers as drivers with backward-connected output transformers, the amplifier must be properly loaded with a high value capacitor and suitable resistor to maintain stability of the amplifier when it is looking into an extremely low impedance or low value of DCR such as the xfmr primary.

The best solution is to use a driver transformer designed to and suitable for the task at hand, but they tend to be expensive and quite scarce nowadays. I was lucky; the driver xfmr in my rig came out of a Gates BC-1F, and was designed to match a pair of class A push-pull 845s to a pair of class B 833As. But Gates stopped using that xfmr in the mid-1950s, so aside from a Peter Dahl replacement unit, they are essentially unobtainium.

That said, I think the optimum solution is to use the cathode follower or FET driver and be done with it, as it eliminates the troublesome driver xfmr issue once and for all. This is especially true if you want to apply some aggressive audio processing to that rig; this becomes very difficult to achieve in those plate modulated  rigs with an abundance of cascaded audio coupling transformers, due to ringing and overshoot issues. I have an Orban Optimod 9100A, and this processor is basically unusable with my rig for this very issue; it is difficult to have consistent control of the negative-going peaks of modulation. These later generation processors are suited and designed for use in transmitters (class E comes to mind) where there are no complex impedances anywhere in the audio signal path.

Just my 2 cents!

73,

Bruce

Tom,
those 500 pf caps could be rolling off your highs. If you really need them I would add some shunt R to raise the time constant. Resistors could go back to the CT as long as you have a heavy bypass for RF and audio at the CT. Make sure the 500 pf caps are right at the tube sockets.

I agree with Frank on this one.

Personally, I do not see the need for bypass capacitors at that point in the circuit, nor am I familiar with any broadcast or other rig that used them at the modulator grids-to-ground either. I think they are superfluous. At that impedance point and value of capacitance, I'm sure they will have some negative impact to the high-frequency response, and that capacitance could be reflected back through the driver transformer and cause some potential ringing and overshoot issues.

I can also see a potentially negative effect of putting series resistors between the each secondary leg of the driver xfmr and modulator grid, in an attempt to control potential audio parasitic issues. You want to keep the resistance as low as possible at that point in the circuit; as the grids are driven positive in class B service, a voltage will be developed across these high-vlaue resistors and the modulator grid bias point will shift accordingly. The load the driver looks into, as well as the grid bias supply, must be as low an impedance as practical, including (as I had discussed above) a driver xfmr with leakage reactance as low as possible.

If you are concerned about audio parasitics in the modulator stage, the place to put resistive suppressors would be right at the plate connection of each modulator tube.

73,

Bruce

The reactance at 10Kc in a 500pf cap is >30K. I don't think it's a problem.

(See recording below of 4X1 rig by Al/ VTP)


Interesting.  I did the C calculator before putting them in there, so figgered they wud be transparent for audio. The fact that the modulators are on the same chassis made me do it.

I just got a recording from Al/VTP.  For a first shot I thought it sounded FB.  Though, Tron said earlier he heard some slight distortion. He's got such good ears for this stuff maybe he's hearing the effects of that driver transformer, dunno.

OK on the info, Bruce and Stu. Good points and I'll keep them in mind as I continue here.

Hope to get back on 75 in a while while the skip is still short.

run that time constant with the output Z of the driver transformer

I never used any capacitance at the grids of the mod tubes, no parasitic chokes, just a resistor across each secondary winding of the driver transformer to give it a load to look at.

I was home for lunch and heard you on Tom, with others, very strapping signal into South Jersey, I did not notice the audio (a good sign) as I was doing antenna work, getting the swr down a little and running some rg214 to the roof.

I got the swr down to 1.3-1 at 3880....

From what I heard, it sounded good, but I am no judge at all.

It seems like you are getting to the nit picking phaze, which if you let it, can go on forever....


Brett

I was not compairing the caps to oxygen free cables.
Dont all audio amps look into a dc short?
Is that not what a voice coil is in a speaker?
Much lower resistance than a transformer I would think.

I suspect really odd things to happen if you used half a winding, or tried to run audio through 2 windings in paralell, or even if you ran it into a much higher impedance than it was designed for.

Brett

That's exactly the way I look at it as well.

73,

Bruce

OK, I'll take out the 500 pf caps and see if it helps the high end.  Tnx.

I'll replace them with some 20K resistors to ground.

Had a ball on 75M tonight with the new rig.  Plate modulation has a soul- the only way to go...  ;D

T


BTW, Here's a recording from Jeff/W2NBC.  I'm generally pleased with what I hear so far! Hope to fine tune it some more.  Tnx, Jeff.

Sounds good to me!

Brett

Hi

Reading N3DRB report ,receiving  your Transmitter,He wrote that the audio  sound is flat,I have the conclusion that Your Transmitter is Okay,but lags "power" in it(audio),So since You all ready used SS Audio Amplifier,why don't You used a preamplifier with tone control, or compressor or some kind of audio prossesor  to Shape Your Audio,in front of Your SS amplifier.To boost the high tone or the Mid tones or the low tones,suitable for Your "taste"
Just a Suggestion .


Gito

oh it sounds good now.  :D  I mean, like : really good.  8)

no joking, that transmitter is smoking. It's got personality now, Thomas. Good show.

Thanks, Derb!  It was a nice surprise to hear ya break in tonight.  You were vely loud here with the Gonset.  I will have to slide down the band and join you guys this weekend for the WFD Net.

Gito:  Yes, we have used what you suggested for many years. I use an Electro-Voice RE-20 mike into a 32-band equalizer, into a 6-band Beringer audio processor. I think the problem yesterday was I had the wrong taps on the driver xfmr and was killing the highs according to my audio sweep test.

It's sounding better now and I have plans to improve some things further. It's a long haul to get a plate modulated rig just right.  I never have an easy time with it.

Now to pull out those 500pf grid bypass caps and see what happens.

T

   Brett, Sorry I snapped a little there. Im on a business trip that has went awry, and I'm out of clean clothes, and too many 12 hour days.  :'(

Here is a post of spec's on a 4 ohm impedance 800 watt RMS 15" Woofer:

http://www.parts-express.com/pe/showdetl.cfm?Partnumber=295-420

This dude has 3.8 ohms DC resistance in series with 2.5mh, and a rating impedance of 4 ohms. No worry here driving a big amp into it. A big transformer on the other hand on the low - Z primary can be much less than 1 ohm DC resistance even though the impedance rating could be 4, 8, or 16 ohms.

Tom, That rig sounds fantastic. Congrats!

Jim
WD5JKO

Thanks for the link, Jim - I'll check it out.

Well, Frank and Bruce - you guys were right!  I took out the 500 pf bypass caps in the modulator grids and the high end now goes well up into the 8-9kc range without distortion!  Those caps were definately affecting the transformer. 

I still notice the mod iron doesn't want to modulate heavily up above 6kc or so, but at least it's clean. The highs start to roll off at about 6kc but the modulator current stays the same and the mod tubes start to show color. Looks like typical iron capacitive shunting. It's an RCA BC 1KW mod xfmr.

I ran into another problem tonight that took me about 4 hours to find. Seems the modulators started idling at about 600ma out of nowhere. What a bear it was to find, but turns out one of my terminal strips that connect the grids went intermitant and the grids were floating on one of the tubes at times.

During the task, I accidently left the Jesus stick on the HV and turned it on. It burned the arms off my HV keying relay. It looks like an amputee.  So screw it.. I may not be keying the HV anymore and just leave it on the particular rig.

All in all, the audio sweeps pretty nicely now in the high end. That's probably what the Tron was hearing for distortion b4 - I'll see what he says tmw.  I can sweep from about 25 hz all the way up to 7 kc and see little variation in the sine wave - looks quite nice... only that it starts rolling off above 6kc. I plan to roll off the audio above 5kc anyway, so that's not a bad thang.

Later -

T

Hi Tom

I  also used also used  Behringer  Compressor for a small Bc Station in my town.
Actually its  6 band  compressor ,limiter ,gating ,which can be set for each band channel individually  ( the frequency band is divided into six band.)
The purpose is that if  a frequency say low frequency is compressed it doesn't compress the high Frequency band,and vise versa  if the high frequency is compressed .it doesn't affect the Low frequency band.

It is as a leveling amplitude of the frequency band,reducing  the dynamic range an make a more solid sound .

 I think the equalizer must be placed after the Compressor :  Compressor- Equalizer- Amplifier.
If you put a signal in the the Equalizer and boost it for instance 1 v and the compressor treshold  is .5 v than you always get a .5 v out put,You boost it to 2V You always get a .5 v output.
So how you try to boost your high frequency band and the end it is limited by the compressor.

So say if you use a pink noise/white noise injected in your Compressor,and after the compressor  You have a flat compressed Frequency response and drive with it Your Modulator/transmitter and at the end You has a lost in the High frequency notes.

It can be caused by the interstage trafo ,You used a 80 ohm impedance trafo as an 4 ohm input,even if the impedance ratio is okay (primair to secunder) a 80 ohm impedance must have much turn winding than a 4 ohm impedance,so the inductance is much higher than needed by a 4 ohm design trafo. the effect can boost the low frequency .and some looses at high frequency.
And maybe caused by some other reason.

So after  having a flat response  audio  after the compressor and using  an equalizer after it,you can change this flat frequency to a frequency with a frequency band with a boosted high Frequency end by the Equalizer,and hoping it can compensate the high frequency lost in the Modulator.

And since You used it after the compressor, the highest peak output of it is still the same,even how hard  You drive  your amplifier( first limited by the compressor than shape by the Equalizer).
Again I can be wrong.

Gito

Definitely time to get RID of those transformers in the driver chain  :)

I have an audio driver project in the works (for push-pull tubes).  When it's complete, it would be very instructive for me to bring it to your site and hook it up to your modulator.  Then we can actually hear and measure the results!

At that point, there will be a practical, proper implementation of something that will potentially be of a benefit to many.

Regards,

Steve

Gito:

Thanks for your info. Yes, the 6-band processor is a very nice device for bringing up the average density independently of freq.

I use my compressor last in the chain ... mic > preamp > EQ > 6-Band processor.   I always thought the low end could be manipulated better that way and was always told to put the compressor /limiter last in the chain.  Does anyone else have an opinion on that?  There was a time when I swapped them back and forth and eventually settled on this config.  I'm always open to new ideas.


T

Hi Steve,

I would be very interested in this circuit as well.

I could replace the audio input xfrmr that splits the phase at the input to my audio driver, and the driver xfrmr that couples the audio from the p-p 845s to the 833A grids.

I still would need to couple the existing negative feedback loops (one for each 833A) in my rig into this audio driver of yours. I assume your design makes provision for this.

Can you briefly describe the topology of this circuit for us? Are you doing just the engineering and design, or are you also providing an unstuffed PCB with a parts list and schematic, etc.?

73,

Bruce

Tom,
Heard your new transmitter yesterday......all I can say is it sounded GREAT....nice job.
I recorded u but so did other people so I wont waste your time sending u the recording.
Hope to work ya soon.

Bill

Yes a pair of 11N90s as source followers driving the grids would motivate the modulators well into HF if done correctly.

Thanks, Bill. Yes, I received four recordings so far - that's plenty... ;D

Last night I improved the high end some more and think I've licked the grunge Tron mentioned. So I could use another recording the next tme you hear me.


Steve/QIX - I sent you an email about working with you on the MOSFET driver project.  I'm in no rush to put together a new driver now that this SS amp and xfmr are acceptable for now.

Bruce (and Steve) - For Audio Negative feedback, consider that KLR cathode circuit I posted earlier in this thread. It has a NFB stage that will mix with the regular low level audio. Bottom line is I have a universal stand-alone NFB board that will feed ANY audio driver -   ready to go.

T

Series resonance?

--Shane

I'm in the process of reconfiguring the 4X1 modulators from triode connected to conventional AB2 tetrode connections. I built a 400V regulated screen supply and bypassed the heck outa the modulator screens. I will drive it the same way with the grid driven SS amp and xfmr. It will be able to draw grid current for power. The screen will fluctuate between 0-95ma.

The Tron said his old 4X1  X 4X1's triode connected had some linearity issues and suggested trying them as tetrodes. He said the screens neeeded  a higher voltage above the grid to function better.  (As usual) We'll soon see.   If anything, they will be much easier to drive and the SS amp / xfmr will have a much lighter load.

I'll post a pic later of the rewired underside of the three 4X1's.

T

Tom,
Try to get away with low screen voltage to avoid distortion when the plate swings below the screen. I noticed something in the low end when you were hitting it hard last night. Not sure if it was over modulation because didn't have the spectrum display on. Maybe you could drive the screens and ground the control grids as an option.

Here's some pics I took of the rewired 4X1's underside. You will see the RF tube at the left end and the modulators at the right. Notice the new screen bypassing in the modulators - setting up for regulated 400V of screen voltage as tetrode modulators. That 600 uf bypass cap shud do it for the lows.  The LC tuned circuit for the final is about 4" away in a sub-chassis. I wanted the area clear for air as much as possible.  The RF deck is very stable with neutralization and a 14K resistor network as swamping. Easy to drive with about 10w.

Also shown is the RF deck with the tubes removed. Terry, there's a view of the aluminum angle/ sheet construction you wanted to see.


Notice the tiny vac variable neutralizing cap.  The C1, plate tuning cap is not visible cuz it is directly behind the 2000 pf loading vac var.

See the weatherstripping that provides a tight seal for the chassis for air flow. I use four large ground straps on the chassis corners to RF ground the subchassis since the weatherstripping is insulated.  The tank coil looks crowded in the pic, but actually has a diameter of space around it.

TEFLON wire used, of course... ;)

OK on the report, Frank. Yes, there are still some minor issues that I need to work out. I think this tetrode mod and an audio driver w/o a transformer will do it.

Later -

T

I've been listening around the past couple of days hoping to hear your new rig on the air. I figure those flashes in the sky off to the southeast are probably occurring when you key and unkey. You do nice work Tom! How come my stuff looks like the picture below?

73,

Rob

heheheh... Rob you bagger - I've seen some of your work and it looks quite anally inspired.. ;D  Meaning you dress it up nicely.

You will find with my work that I don't make any real effort to make it pretty. It is strong, fuctions well, I try to keep things square and at right angles, (sometimes) but that's about it.   There's lots of work out there that will put my work to shame.   I tend to recycle and tear down things a lot, so it's more of a time thing too. But you can drop my stuff from a few feet and it will stay together and it will work well when I'm done - that's really my only requirements... ;D

T

New Problem:

I have the the modulators connected as tetrodes with 400 volts on the screens with a current meter monitoring screen current.

All is working well and I see about 30 ma of fluctuating positve SCREEN current on audio peaks.  But when I unkey, I get a huge NEGATIVE spike of screen current measured at least 80 ma.  It lasts about a 1/8 second and goes away.  This is also with no audio connected to the modulators.

I am keying a 49K resistor in the CT fil as usual for mod on/off..

What could be causing this?  I have a 680 uf cap across the screen pin to ground along with some smaller .01 types.

T

Tom

Sounds like the large screen bypass capacitor is discharging through the screen power supply (or some other path to ground that you are introducing)... and therefore through the meter in the direction of "negative screen current".

Stu

put a heavy bleeder on the big cap so it doesn't freak out the meter. Then you could go lighter at the power supply. Keep something at the power supply for safety.

I relay the screen voltage on my stuff.
The supply is on all the time, and regulated, and a relay applies the voltage and is keyed by the step start, when the step start goes to full power, the screen voltage gets to the mod and rf tube grids.

The screen voltage cap is not big, but I do choke input on the power supply.

Brett

Thanks for the info and suggestions, guys.

Frank, I can't bleed the supply heavily cuz it's using a SS regulator that is almost at the limit now at 400V @100ma.  So I decided to remove that big 680 uf cap at the screen pin and replace it wid a 4uf 1000V oil filled. That will keep the initial charging which drives the meter nuts - and we'll soon see if it helps the screen spike problem or not. 

BTW, I found the spike is in the positive meter direction, just like normal current. I thought it was negative before.

I noticed it only happens when the RF amp is unkeyed with the modulators. With no drive to the RF deck, there's no screen spike in the mod tubes.

Brett - I may try keying the screen suppy as a last resort. But since the screen spike happens only on unkey, I wud need to put the relay on the screen supply's output as a HV relay.  We'll see.

To add to the problem, that big cap held a charge so long when I went to measure the resistance of the circuit I blew out my Asian  VOM. Gotta go down to RS now.... cheezz..   I hate those little meters that can't take abuse. Shud get an old Simpson again.

T

Tom,
When I unkey my 4x1....the screen voltages spikes to about 100 volts above the operating voltage...Jamie said this is because the screen supply no longer has a load on it and  the filter caps  are discharging. Dont forget this is  the screen voltage on the rf 4x1......not modulators.......

Tom,
I forgot to mention........the screen supply for the 4x1 is a separate power supply. I have a 20 henry choke in series with the screen to self modulate it. I do NOT tap the screen voltage off of the modulated b+.

Also I can vary the screen voltage with a variac.......

Bill  

OK Bill -

Yes, I do the same thing with my RF final screen using a self modulating choke.  The RF final is FB.

But since I changed the modulators to tetrode configued with 400V screen voltage, the screen meter on the modulators spike. Maybe it was there all the time before and I din't see it cuz I didn't have a mod screen meter, dunno. Might even be an instability of the modulators taking off.. and is now enhanced cuz of a tetrode's greater sensitivity.


 I'm still working on the problem and will post the solution - soon I hope... ;D



T

Screen spike problem solved rig is getting very tame and reliable.  Almost there.

Tetrode config on the modulator 4X1's -  update:

Well, I took out the 680 uf bypass cap on the 4x1 modulator screen grids and replaced it with a 4 uf.  The screens now have regulated 400V and the bias is about 90 diodes in the fil CT (-45V) for a stiff bias.  The screen spike problem is gone - fixed! I ran some tests with the audio sweep gen and also some voice checks. I think it's the best config so far!  It's also easy to drive with about 10 watts of audio or so.    The screens are drawing only about 15ma on peaks, while the Eimac specs say 95 ma is for full power - this tells me there's plenty of headroom left.  The grid is drawing no current at all. I'm idling them at about 300ma total and the plate meter barely move under voice peaks.  This tells me we are running close to class A, but really about AB1. That's a great mode for cleanliness.  This is on medium power tap, so it will increase the modulator requirements once I bump it up to full strap.

I'm real pleased the tetrode connection so far.

Still have some minor issues to iron out with the input grid tuning, but almost finished with the rig for now. (until I get the next modification bug up my ass... ;D)

More later -

T

I ran some tests tonight. I swept the audio from 30 hz to 8KC and found it looked very good. The only problem was above  5kc when going above 70% modulation.  The driver transformer is causing some minor distortion.  So I'll be building up a transformerless driver for sure.  Otherwise, I think the rig is working pretty well.

I also put the scope probe on the RF final screen and saw the clean audio from the self-modulating choke. However, on the RF grid I looked for audio after the grid leak resistor and saw very little to none. Is the grid leak supposed to have a reasonable audio voltage swinging there or is it just supplying the additional DC bias only?  I always thought it helped to modulate the final too.  

The grid leak circuit consists of a 2500 ohm resistor in series with the fixed bias supply, .001 to ground, in series with a  2.5MH RF choke, and to the RF final grid.  The coupling cap and RF also connect to the grid pin.

Also, I measured the audio voltage at the grid of one modulator tube and saw about 90V peak. So I would need 180 peak to peak to drive two tubes in push pull, right?   I need this info to deteremine the power supply voltage needed for a new audio driver.  Eimac suggests 140V per tube to make 3KW of audio, but my requirements are substantially less.

T

Since you only need 10 watts of audio drive now, a good 8 ohm to 10kct 20 watt transformer would likely fix you up nicely.

Are you sure the high frequency distortion is in the driver transformer and not the rf screen bypass, rf plate bypass, mod trans, etc?

My 813 rig is flat up to 10k then rolls off very quickly, some bypass I suppose, since the audio amp and driver transformer are good to past 20kc. Or it could be the mod iron.

Its bad news to go beyond 5kc with a high power rig anyway...

Some guys sound really good and are not real wide, others sound good and take up the entire window from below 3870 to above 3890....

I hear/see all sorts of combinations, guys who sound somewhat rough and take up 25kc, something not right in the trans or the way they are running it, other guys just dont limit the highs at all, and pass audio well past 10kc, as if anyone could hear it at prime time when they operate mostly....
How often can you use the 20Kc filter in prime time, or how many people even have a 20kc filter?

Just poor or sloppy practice...fuel for the anti AM crowd.
Last nite I heard a group on 3868 moaning about AM, which they enjoy, since they snuggle up to 3870 so they can listen to the splatter from 3870 to 3890. They really seem to enjoy this as they do it every nite...





Brett


 

Tom,
I wonder if you could configure the driver transformer to use 1/2 the secondary? (tube side) to reduce winding inductance? Then feed grid bias in through a pair of load resistors into each grid. Bypass the heck out of the bias side and give it a good load like a power Zener

Brett:  The distortion starts directly at the audio transformer's output. I don't see it on the input cuz the SS amp is holding firm.   I only have one transformer left here. A few years ago I sold them off thinking I'd never use low level audio transformers again... ;D

Frank:  The sec is just one winding with a CT, 7,000 ohms, so no way to alter that. I need the CT.

Looks like the only way is to build up a transformerless driver.


I'm still wondering if I should be seeing AUDIO after the RF final  grid leak resistor?  I see plenty on the final's screen, but little to none on the grid leak. ???

T

My point was use 1/2 the winding and do a simulated CT with a pair of resistors to a common point. They also serve as a load on the transformer.

Tom,

Could you use the CT to one side of the secondary, and then use the trick of a pair of resistors to "simulate" the center tap?  Like the way you get one with no center tap on a filament xformer.

Not sure if it would work at all, but might be worth a shot, if you need the different ratio.

--Shane

I see Frank and I said the same thing.  That IS scary :)

I had a feeling that's what you meant, Frank.

I'll give it a try.

As you said, it might reduce the high freq inductance causing the problem - might be worth a try. The step up will be reduced of course.  What values would you suggest from the winding to CT for the resistors?  It will now be a 1750 ohm winding end-to- end. (7000 ohms /4)    (265 ohms pri > 1750 ohms sec) 


T

1 or 2 K each big enough to handle the voltage. You may have to play with the value to get the best response out of the transformer. you need to bypass the CT well for all audio frequencies so a number of caps in parallel.

Well, after spending the day on Fabio putting in a new grid L/C circuit into the sub-chassis, adding a rotary switch to select the diode stack for modulator bias and running some more tests, I'd say he's about ready for prime time.

The Huzman is sending me a hi-fi driver transformer that will work FB until I get a new direct coupled driver built. The rig is quite stable now for both RF and modulator arcs. The only remaining problem is when I unkey and run less than 200ma of idle current in the modulators, I get a spark gap arc on the mod xfmr secondary. It's simply a matter of the iron needing to dump its power somewhere. Gotta figure that out and work on it.

Here's a pic of the new grid circuit I relocated from a remote outside chassis. I decided to locate it right at the tubes pins like it shud be. The air flow is still superb, so it worked out FB.

The 4X1 modulators definately play well as tetrodes, so I'll leave them that way.

Hope to be on the air with it this weekend.

There's a pic of Fabio in operation tonight running in low tap. The final shows no color at all, while the modulators show a faint orange blush. Fabio is shown in standby. I'll get some plate color shots next time.

T

Well, I ran the rig today on the air and everything is basically working FB.  Had a big arc over, but that was just the 4X1 RF tube's plate cap braid arcing to the cabinet, about 1.5" away.  Seems RF can jump huges distances that DC can't. Easy fix.

Anyway here's the last problem:  When I UNKEY the rig, I get a snap - an arc across the secondary modulation transformer arc gaps only.  This happens ONLY if the modulator tubes are idled less than 150ma or so. If I idle them higher there is no snap, but I still want to locate the problem.

I tried all kinds of relay delay combinations... I put a delay on the modulator key ups and unkeys - antenna relay keyup/downs, RF final key up/downs, etc. Nothing seems to help.

One strange thing: If I ground the modulator grids, I always get the snap on unkey even with the modulator idling high, but when I turn on the SS amp and have it running, the arc happens only when the modulators are idling less than 150ma.

My question is WHERE is the power from the mod iron and heising reactor supposed to be dumped on Unkey?   I thought is was into the final, but holding the final on longer after key up doesn't help. 

Any suggestions for a fix?

T

That is a very strange event. My thoughts are the same as yours, that when you unkey, the energy would softly drop as the tubes pulled down the supply. I wonder how it would behave if you sequenced your "step-start" to "step-down" when you unkeyed. Still, it would be nice to find where the transient is originating.

The rig sounds very nice. For those who have not heard it, here is a very short clip of Fabio, featuring the moment of arc-eology, courtesy of WB2CAU Eric's recording studios. A scary moment in time, but Mr. Vu brought Fabio back to life after a glance at the inside of the top of rack. Hmmmm... was that black zorch mark always there???

Tom,
I had a similar situation with the CVM-4 and 813 rig and not driving enough snot into the rig. For some reason running it light, it did not like it. It arced all over the place but when it was run under the specs of your plan it was FB.  Maybe you haven't found the sweet spot yet?

"uh oh"   HA!  Pretty good catch.

Yep, the plate cap lead arced 1.5" to the rack cabinet top. RF is strange stuff.

I was talking to Tron tonight about the arcing problem. As usual, he made a good suggestion. He said the spark gap arcing on unkey was cuz the RF final was shutting off and resulting in the mod iron not having anywhere to dump the power. But I said I already sequenced the relays so that the RF final stayed on after the mod relay. He axed if I was using fixed bias on the RF final too. I was and the tube was cut off completely once drive from the exciter was gone - regardless of if the RF final's relay was still on... bingo!  I still have yet to reduce the fixed bias so that the RF final idles at a small current when the drive is gone, but I'll bet that fixes it. We'll see.

So remember this one, guys: Use just enuff fixed bias so that the class C final tube idles in class A or somewhere like that with no RF drive. A complete cutoff gives the modulator no where to dump.  (At least I hope that solves it)

T

Bob,

Do you mean RF drive?  Then yes, low RF drive will not give the RF final enuff current to load the mod xfmr to dump the power. (Assuming you are using too much fixed bias too) I've seen it written before to let it idle like a linear, but didn't realize this was one of the reasons.  The downside is the final is now more sensitive to taking off in the time it is running linear due to higher gain.

 In my case, I have enuff drive, but the tube is being cut off anothere way - by too much fixed bias.  I have about -90v fixed and -140v grid leak bias. Should be more like -60 v fixed or so.

hi

Can it be because the drive is gone instantly when the B+ of the transmitter is unkeyed?
I think You must keep the drive for a part of a second after the B+ and Screen Supply of the RF tube is unkeyed.
So when the B+ " slowly,in part of  a second" fading away  it still gets an Rf Output/ used to dump /to bleed the B+ because of the RF drive.

Gito

All set - no more arcing on unkey.

It took TWO things.

1) The fixed bias had to be reduced so that the RF final continued to idle when drive was taken off. (Keeping drive to the final would have done the same thing, thanks Gito)   However, since everything in the shack is being keyed by the transceiver PTT, the exciter drive is shut off as a first step, so the RF tube fixed bias fix was easier.

2) Added a  100 uf cap across the fil keying resistor to keep the cathode bias on longer (and the tube idling) did the trick.   I happen to use cathode keying, but if you use something else, then simply keep the tube idling whatever way you choose.

Now when I unkey, no matter what the modulators are doing or idling current they are at, there is no spark gap arc.  I do see a slight movement UP in the RF plate meter showing the energy is going into the RF final circuit and to the antenna. I unkey the antenna relay last in the sequence.



T

Hi Tom.

 Tom,but  when You idle the plate current,doesn't the transmitter works in different class ( class AB2?) and a expense in the tube efficiency?
It different when You" delay" the driver when You unkeyed.The RF transmitter is working for it was design like the data's book.(pure class C)
The spike is caused when the plate current instantly  disappear ,not because the high voltage instantly disappear.( I think the B+ graduadlly disappeared).

When the drive is still"on" the the plate current slowly disappear ,at the same time The B+ plate " slowly" disappear.
And I think the Spike can be elimatted.

Gito

Hey Tom heard you on and you sounded good. I wonder if the cap is to ground or is one end of the mod transformer secondary to ground? I think its ok for BC rigs that are more or less always on. In the Westinghouse they have it across the B+ feeding the bottom of the mod transformer to the bottom end of the modulation transformer therefor no cap to charge up like a cap in series with an inductor. In my Westinghouse I have it across the top of the secondary of the mod transformer to the top of the reactor. The bottom of the reactor is at B+ as well as the bottom of the secondary of the mod transformer. No arc overs of any type in my arrangement.

No common point to ground I feel is the answer.

Gito:

The final operates in class C all the time when drive is applied because it uses a grid leak resistor. When I unkey, for an instant the drive disappears  but the fixed bias lets in run in class AB for a second so the modulation iron will still have a load to work into.  The idea is to keep the tube on for an instant while the iron dumps its power and that's what is happening now.


Terry: The heising coupling cap is located on the high side.  The Tron seems to think the whole problem is related to the final dropping out into cutoff too quickly, thus the iron was left hanging without a load, so it sparked instead. So far, his info seems correct. (as usual)   ;D  OK on the Westinghouse arrangement. I wud try that, but I think the problem is solved for now, TNX.  BTW, heard youse guys on 160M tonight. Good signal with your generator running.  I have no ant for 160M right now nor does the 4X1 have a tank coil for it. I may have to do some winter climbing, dunno.

T

good that it's fixed. This thread has taught me a few things!

Hi Tom

I get the conclusion,that the culprit is the cutting of the drive instantly.
When the drive stop ,the current of the plate Rf is instantly stop,making a flywheel effect ,just like ignition coil work,when there's current flowing in the coil,the core stay magnetize ,when the current collapse ,there a flywheel effect that makes a great big emf voltage in the coil ,causing an ark jumps a cross the spark plug.
In my opinion when You idle the Transmitter.it does not completely remove the spike,the Spike is still there but with a reduced amplitude/lower voltage,So there's no ark cross the spark gap(not enough peak voltage).

Gito

Hi again

Tom when You idle the Rf tube,there's a magnetizing current flowing in  modulator transformer,for instance there's a 100 ma current flowing there's it means there's 2500 v X .1A = 250 watt power dissipated in the modulator transformer.,makes it a " little" saturated ,so the back EMF is smaller than no "saturated" trans former.
I mean "saturated" is an example ,meaning the Modulation Transformer is partly magnetize.

Gito

There is 100 ma of current going through the transformer. The only dissipation is due the wire resistance....But you have a nice big electro magnet. So when the current stops a large voltage is induced back into the winding since there is no load across it. This is why you generate a big spark.
All dressed up with no place to go.

Then what's the correct solution?

Right now I have the final conducting for about 1 second after unkey. It appears to absorb the mag collapse OK.

Gito suggested to keep the drive going. This will keep the tube on and conducting plate current - just the same as putting it in class AB for an instant to keep plate currrent going. I don't see any difference in the end result.

But I do see the wisdom in Gito's sugestion to let the plate voltage and current die slowly into a load.  My transceiver drive drops instantly and will have to stay that way. Maybe there's another PRACTICAL way to slowly drop off the HV so there will be an EASIER collapse of the field. Right now I keep the HV on all the time and key both cathodes of the final and modulators using a 56K biasing resistor that gets shorted by a relay.   I could add a keying relay if needed for the HV.

BTW, I added a 100uf cap across the 56K keying resistor and that helped to ramp down the bias (thus plate current) in the brief AB mode. Maybe that is the right approach.  Taken too far, I could see the RF plate get red for a second...  :o


Opcom:  Yes, this has been a learning experience for many of us. I post most of this stuff (some trivial) for two reasons. One is for the Newbies to learn and get confidence that they too, can build a big rig. And second, for we OT's that need to be reminded of stuff we once knew - or never knew at all.. ;D   This particular plate modulated project has been especially tough compared to other projects, but that's a good thing in ways to get the problems ironed out early rather than popping up later one at a time.  Plus, it's been stressed so much that the weak parts have been kicked out by now.  I've gotten a bunch of emails form various guys offering parts, help or just questions themselves. It's a good blog.

T

Tom I think the first action should be to bias the modulators off slowly. The cap across the resistor sounds like a KISS way to go. Once you dissipate the mod transformer energy then you can drop the drive then last change the antenna.

Frank,

Surprisingly, the cap on the mod keying resistor had no effect. The spark was there regardless of when the mods turned on and off. But the RF final had the big effect.  It seems the mod iron is dropping the field even with the mod tubes uncapped at the plates.   I figgered since they seem to affect the loading at times (their idle current) it wud help to delay them, but no cigar.

T

Well if there is any secondary current, it also stores energy in the mod core.
You want to use the final to suck the well dry before you remove the drive and antenna as a load. so if there is any final idle current with the modulators off you still could generate a spark. I bet the inductance is pretty high at low current. Maybe this is why guys drop the HV and let the system discharge before the load and drive drops out. maybe you also need a cap on the final CT also. 

Yes,  you're right, RF drive. Sorry. Memory is getting fuzzy.

The final CT is where the 59K resistor and cap is now. All set.

As I mentioned, the drive comes out of the transceiver and I key the transceiver, so its drive drops first on unkey. Not sure I want to time the transceiver keying externally just yet.  I've gotten other plate modulated systems to work this way, so can do it again.

I used to key the HV DC until a short wiped out the arms of the relay. I think I'll put another HV relay in there and drop it out early on unkey. That may help a lot to slowly bleed off the power. At least it won't be supplying additional  power during unkey as it does now.

T

Hi

I think the spark is caused by the instantly stopping current flowing from B+ to the Mod. Transformer to the RF  final tube Plate,because of the stopping of the drive.

I think idle current is different ,it does not stop instantly.it decay in phase with the decaying B+,because it takes "a part" of a second to completely discharge the Capacitor. So the magnetizing Core is decaying in phase with the decaying current( the current is flowing till the B+ is Zero).

 This gives me an idea ,make the smoothing capacitor capacity(for the B+) bigger,so it needs more time to discharge .
Can the power supply of the driver discharge A part off a second slower than the power supply of the RF final,by making a bigger capacitor as the smoothing part of it's B+.
Hoping the driver stays on for part off a second after the B+ off the RF final tube decay.

Gito

I see what you are saying now, Gito.

This is based on the HV being shut off before the filter caps so the caps can discharge slowly, otherwise the current will remain the same regardless of how long the drive is kept on.  Right now my HV is always on.


I'll have to think more on it.

T

May, 2013 ---

I see this old topic has almost 18,000 hits over 3 years.  This is most likely due to outside general web search engine readers.

UPDATE:  Since this thread, I've retired Fabio I  -   and built a new Fabio II. Might as well have some continuity here.

I look back at Fabio I construction and cringe at some things, so hopefully the newer version will give new readers better ideas for their own construction.

Fabio II:  (not completed yet)

http://amfone.net/Amforum/index.php?topic=33816.0

73,
Tom, K1JJ