The AM Forum
March 28, 2024, 05:06:36 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1] 2 ... 6   Go Down
  Print  
Author Topic: Simple Audio Driver for 833A modulator  (Read 94275 times)
0 Members and 1 Guest are viewing this topic.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« on: October 30, 2005, 07:16:58 PM »

I wonder if anyone has given thought to simple modulator audio drivers. Most guys use solid state amps driving backwards audio output transformers. Some use a line up of tubes using direct coupling or transformers.

Some guys have used the IJY/QIX mosfet board too, though that requires a lot of power supply voltages, heat sinks, complexity.

I'm designing up a single 833A final, plate modulated by a pair. I want a silent rig with no blowers. I was wondering if a single Mosfet in the cathode would drive each 833A modulator? If the tube needed a positive bias on the grid to come to life, that could be easily done to give it the ability to draw grid current.

A p-p fet driver into the two cathode Mosfets sounds like a real simple method.

Will this work?  Any other suggestions for simplicity?

[Frank/GFZ , I'm doing metal work on the 4X1  X 4X1 PDM rig now.]

T

Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #1 on: October 30, 2005, 07:24:04 PM »

Quote
I wonder if anyone has given thought to simple modulator audio drivers

Probably.  Not me, though.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
Bacon, WA3WDR
Member

Offline Offline

Posts: 881



« Reply #2 on: October 30, 2005, 10:50:55 PM »

Might be better to use zero-bias tubes like 3-500Z for the modulators.  Then you could avoid the dual-supply issue, because the filaments only need to be equal to, or negative, with respect to the grids.

Grounded grid, and series fets to a single negative supply could then work if the filament transformer capacitance is low enough.  Maybe a 200 KHz filament supply?  Otherwise a totem-pole of fets is required, but that's not too bad if only a single driver supply is needed.  But then with a totem-pole, and a grounded filament and a single positive driver supply, you could drive the grids with much less current.  In any event, it's much simpler with zero-bias tubes, I think.  Darn, can the 3-500Z take the heat without a blower?  Maybe with a quiet blower.  Push-pull parallel 805's instead? Multiples of them?  Grasping at straws here.
Logged

Truth can be stranger than fiction.  But fiction can be pretty strange, too!
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #3 on: October 30, 2005, 11:28:08 PM »

You need to get back on the air Bacon.
3-500Z's are really not zero bias. No big deal though all you need is a Zener in the cathode return.
Same thing would work for an 833 but you would need a lot more Zener Voltage.
Cathode drive could be done with a pair of discreet fil xfmrs. Drive the center taps. Shunt capacitance might not be as much of an issue as the impedance is most likely less than 100 Ohms.
Grounded grid 833's cathode driven. I like It Vu. 833's stink in cathode driven service at R.F. Anything other than high Mu triodes will not fly for cathode driven service up at radio freq. Audio might be another story. I like the whole cathode driven idea as the degenarative feedback cleens up the output.
You are mastal innolvatol.
Logged
WA3VJB
Guest
« Reply #4 on: October 31, 2005, 09:00:53 AM »

Quote
You need to get back on the air Bacon.

Hey Bacon !
Logged
Bacon, WA3WDR
Member

Offline Offline

Posts: 881



« Reply #5 on: October 31, 2005, 09:12:19 AM »

Arrr, I've been monitoring... but all I hear is my computer.
Logged

Truth can be stranger than fiction.  But fiction can be pretty strange, too!
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #6 on: October 31, 2005, 09:38:11 AM »

Tom,
I've often thought about driving modulators with FETs in the cathode.

What is the cut off voltage of the 833 and how far positive does the
grid need to swing? This should be easy and the 900 volt fets used in the class E rigs should work fine with a good sized heat sink. i'm thinking 3 FETs one for each tube and a third as a phase splitter.  3 FETs, 6 resistors and 2 caps. 
also how much cathode current per 833? Also need 2 fil transformers.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #7 on: October 31, 2005, 10:09:37 AM »

Tom,
I've often thought about driving modulators with FETs in the cathode.

What is the cut off voltage of the 833 and how far positive does the
grid need to swing? This should be easy and the 900 volt fets used in the class E rigs should work fine with a good sized heat sink. i'm thinking 3 FETs one for each tube and a third as a phase splitter.  3 FETs, 6 resistors and 2 caps. 
also how much cathode current per 833? Also need 2 fil transformers.

Interesting and looking do-able... Shocked

Check out these 833A curves on page 6 to determine viabilty for cathode driven service:

http://www.tubebuilder.com/images/tubedata/833a.pdf


Dave, you brought up a good point that I didn't think of... built in negative feedback of the GG config!  I wonder if that wud be enuff or will it still need a loop around the fet driver line up to help that area out with neg feedback too?

Frank, I checked the 833A curves and it shows that at 3kv, the tube draws 900ma at zero grid volts. And it gets cut off at about -70V. So that's 70V of monkey swing from the fet. That's kinda ideal, I think, cuz a fet can turn on in the cathode and generate enuff current without having to add a positive bias on the grid. So true grounded grid simplicity.  BTW, my supply is 0-5kv HV at about 1.2A, but I expect to run it around 3kv so that I need just a silent muffin fan.

I have two matched and  separate 10A/10V fil transformers.  I wonder if it wud best best to mount the fil transformer above the chassis  on fiberglass to minimize audio coupling/capacitance to ground??  The fil bypass caps wud be about .1 ufd or something else?

Frank, if you think it will work, please detail out that fet driver you described. I will build it up. I have line level audio to drive it with.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #8 on: October 31, 2005, 10:31:17 AM »

Tom,
I think this is very doable if 0 volts on the grid gives you enough current and 70 volts cuts the tube off. Line level you will need about a gain of 50 or so let me think about it while i'm working. I have something later.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #9 on: October 31, 2005, 11:35:26 AM »

Tom,
Setting up for a gain of 50 in the driver 50 ohm in the drain and 1 ohm in the source would be a problem. This will put about 25 volts of bias at .5 amps.
A higher voltage audio source would make it easier so there is less positive bias on the cathode. You can't avoid a bit of voltage on the cathode because it is an analog signal. Need to keep this voltage under 10 volts or you will a negative bias on the driver or a bit of positive on the grid. A stage of gain is possible ahead of the phase splitter. That would take another FET.
Logged
Steve - WB3HUZ
Guest
« Reply #10 on: October 31, 2005, 05:32:53 PM »

I don't know how well this circuit may work, but it's interesting. Take a look.

http://www.qsl.net/wa5bxo/driver1.html
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #11 on: October 31, 2005, 06:22:27 PM »

OK, Frank.

We'll have to see what you come up with.  So far, how many fets are you looking at counting the one in the cathode?   Pre splitter, splitter, 2 drivers and 2 cathode fets?

Thanks for the info on the driver, Steve.  Well, that one uses a transfomer [ Shocked] to drive the fets, a 600V? supply and drives the grids of the tubes.   I was hoping for a low level supply to power the fets maybe < 50V, no transformers and cathode driven.  Simple if it can be done.  If it gets too complex, I'll just PDM the 833A's like the 4X1 rig. But was hoping to have a plate modulated rig again with 12"X 12" viewing windows for both the final and modulators..  for your viewing pleasure, caw mawn.

T

Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
kc2ifr
Guest
« Reply #12 on: October 31, 2005, 06:29:16 PM »

Tom...looks like there is no free lunch.....
Logged
Steve - WB3HUZ
Guest
« Reply #13 on: October 31, 2005, 06:40:16 PM »

It has a transformer only because the builder/author chose to use one. I think the circuit could be modified to be transformerless (is that a word?).
Logged
W2VW
Contributing
Member
*
Offline Offline

Posts: 3489


WWW
« Reply #14 on: October 31, 2005, 06:43:43 PM »

Stick with the cathode drive if you can. It eliminates the problem of dead FETs when the tube arcs internally. Sooner or later they will burp.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #15 on: October 31, 2005, 07:38:30 PM »

Stick with the cathode drive if you can. It eliminates the problem of dead FETs when the tube arcs internally. Sooner or later they will burp.

Yep, no free lunches for sure, Bill. It's amazing how it's all a compromise, otherwise everyone would be doing the same technique and the rest wud be forgotten... Grin

Dave: Talk about internal tube shorts... I remember the time when Gary/INR had the IJY fet board driving the grids of his 833A modulators with a 4X1 final. G had sold the full rig to some African govt for a hefty price and was gonna ship it asap. Tron was visiting the shack and let out one last atomic yallo into the mike. A loud buzz rang out as an arc ripped it's way across the fet board, destroying every solid state device.  Grin Grin Grin   

BTW, I may have scored a nice 1 KW broadcash mod transformer. It was originally from a pair of 4-400's modulating a pair.  Now I GOTTA do the plate modulation 833A scheme.  I wonder how it will match up to a pair modulating a single 833A? I don't want to do TWO finals cuz this is supposed to be the "medium" power rig.  I don't know the xfmr ratio or impedances yet, but was just wondering. 

From what I know, a 1:1 ratio is good for one final modulated by a similar pair. And a 2:1 step down is good for two modulated by a similar two. Any truth to that?

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #16 on: October 31, 2005, 08:43:42 PM »

Tom,
3 FETs 2 go in the cathode circuit 1 per tube. third fet a phase splitter.
Cap the couple the phase splitter to each gate of the two drivers.
Say you did a gain of 10 with 10 ohms in the driver drains and 1 ohm in each source
At 1/2 amp peak there would be positive 5 volts on the cathode. The closer to gain of 1 the closer to zero volts you can get. A resistor is required in the drain and source for it to be linear. the ratio of the resistors sets the gain. There has to be some voltage in the cathode circuit to be linear. wattage of each resistor IXIXR=W
a fourth fet could be added as a voltage gain stage ahead of the phase splitter.
The phase splitter should only need low voltage 12 or 24 volts to operate.
then a gain stage ahead of it the same voltage if AC coupled
The bias point of each tube would be set with a pair of pots setting the resting gate voltage on each driver fet.
Steve's find does not eliminate transformers or driver tubes just adds followers to lower drive z.
I wonder what kind of plate current you would get with 5 volts on the cathode.
I'll try to draw a schematic if I get a chance.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #17 on: October 31, 2005, 09:12:10 PM »

It sounds reasonably simple, Frank.  I especially like the single 12V or 24V for the pre-splitter and splitter.  That's the only supply for the whole driver - wow.

You axed about 5V on the cathode. Did you look at the 833A curves posted back a few above?  It would seem to me that it would draw about 800ma with +5v on the cathode. [equiv to -5V on the grid]   800ma is PLENTY of current at 3kv, in fact way too much overhead... Grin    The mod transformer is good for 1KW.  Sounds good, OM.

If this works out, I'm betting many guys will try this circuit to eliminate transformers, tubes and other devices to drive their modulators.  A good audio driver has always been a pain for many of us, especially with very hard to drive triode tubes like 833A's etc. Dave[APE] mentioned that there can be transformer ringing when driving low impedance grids. The GG, cathode driven fet solution wud be quite FB - simple with  built in negative feedback and bulletproof to internal tube arcs.

I'm looking to see what you come up with for a working schematic. Specify actual fet devices, if you can, that we can order from the suppliers.

Tnx again, OM!

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #18 on: November 01, 2005, 08:16:12 AM »

Tom,
I'm pretty busy today but will try to email you something. The class e fets will be perfect for this. 900 volt parts in a 100V circuit 11A parts in a 1 amp circuit.
I figured 5 volts of bias would be OK. fc
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #19 on: November 01, 2005, 09:33:07 AM »

Tom just mailed you a schematic have someone simulate it The 2 k pullup on the input fet may need to be 1 k for maximum voltage swing of the phase splitter.
Gain=10. BMP file
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #20 on: November 01, 2005, 10:10:19 AM »

Thanks, Frank .

Looks so simple, it's amazing!

I tried to upload it here, but it's too big of a .BMP to handle.  So, I'll have to see if anyone can simulate it and I will send it email.  What load should someone use for the driver output?

Could someone here simulate this driver circuit for us? It's three fets and 18 parts. I can email it directly to you.

Or, Frank, do you have a website to access it via URL, or put it in a format that is less than 300K to post here?  It's presently a 1 meg file.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #21 on: November 01, 2005, 11:54:12 AM »

Sorry Tom no site.  Also I would put a 15 volt transorb Gate- Source of each FET. 12 Volt bypass at each pot and bypass the wipers to keep RF out. All the normal stuff. I would drive the CT of each Fil transformer with good film caps each side to CT. Fets will need to be mounted on a heat sink since biased class A. You want to keep the fets cool so RDS/ gain are stable. first stage 2 K pullup on gate will need to be a little lower if you need more voltage swing.
lets hope someone can simulate it for you...
Logged
nu2b
Guest
« Reply #22 on: November 01, 2005, 12:18:47 PM »

Hey Guys,
Interesting discussion, this sounds similar to a circuit I had worked up for a pair of 6L6's long ago.
The zero bias problem is a little different but a similar fet-in-the-cathode should work. this is essentialy similar to biasing up a shunt regulator to work like a series cathode or screen modulator. For audio you need two of these, one for each tube and driven out of phase. You could probably bias the grid up at +200volts and put the shunt fet circuit in the cathode. Gain would be approx 200/4=50.
Would have to check gain sensitivity vs temp to estimate idling current drift.

I'll try to attach a simple schematic.
Regards,
BobbyT


* ShuntKiss.jpg (36.76 KB, 581x480 - viewed 1560 times.)
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #23 on: November 01, 2005, 12:48:12 PM »

OK Frank -

Good on the transorbs and the RF bypasses.  What value film caps for the fils?

Would the fil transformer sec winding have capacitance to the 110ac primary problems for audio? 

And, how about case to ground stray? Would mounting it above chassis and inch help there?

I tried uploading the schematic to my home page, but it will not take .BMP files.

So, maybe I can email it to someone here and they can post it. Anyone?? A group effort will speed things up to get a finished product.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #24 on: November 01, 2005, 01:22:59 PM »

what kind of file do you need i can convert
Logged
Pages: [1] 2 ... 6   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.045 seconds with 19 queries.