Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
This should be interesting




 
The AM Forum
December 07, 2021, 10:46:36 PM *
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]   Go Down
  Print  
Author Topic: This should be interesting  (Read 2428 times)
0 Members and 1 Guest are viewing this topic.
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4420



« Reply #25 on: November 09, 2021, 06:38:29 PM »

 This was a pain
Logged

What? Me worry?
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4420



« Reply #26 on: November 21, 2021, 04:47:51 PM »


Quick update.

The BC 610 mod transformer talks to me. Not much of a problem for now. I wired up the P Dahl plate transformer, ~ 1650 V with choke input & 33 uF. Then the zorching began......Got rid of the clip leads,  and used ring terminals with 5 KV rated test lead wire. Now a lot quieter, and less chance for incontence.

A nice pattern on the scope, then the 31kHz parasitic decide to ride the waveform.  I'll assume that the original builder had this thing running correctly, so adding resistors or even ferrets will wait. Increasing the bias has a small effect on the parasitic ( or is it decrease the bias if I make it more negative???)

My scope has decided to play games with the volts/div, so I cant yeah a low to see what happens; the plate meter shows 50- 100+ mA when swinging. Somewhere i have a 100:1 scope probe, so more to come.

I do wanna get started on the RF side of things, but thats a NoGo untill I get this thing to behave....

BTW, there feedback is from one pin on the S-9 driver transformer back to the
1st stage of the 12au7.

klc

Logged

What? Me worry?
K8DI
Contributing
Member
*
Offline Offline

Posts: 243


« Reply #27 on: November 22, 2021, 12:05:38 PM »

Hello --

I'm looking at the schematic a couple posts up, and saw a few things that may (or may not) be contributing to your frustrations...

R2 has no point. 10M paralleled by the 470K R1?

first triode has no cathode resistor, and no grid leak. With no DC reference for the grid, it will just conduct. Maybe that's why the plate R is so unusually large? Look at the second stage, 100K plate/510R cathode/some part of 350K grid leak, vs 1200K plate/0 cathode/infinity grid leak. If you mod this section make sure that C2 ends up tied to ground and not the cathode, it's an RF bypass cap on the mic input to keep RF out of the audio.

Phase splitter. Usually the grid goes directly to the plate, OR a grid leak and an extra cathode resistance are included. Yours does neither. See https://en.wikipedia.org/wiki/File:Tube_push_pull_poweramplifier.PNG vs. https://robrobinette.com/images/Guitar/Tube_Load_Lines/5E3_Cathodyne_Phase_Inverter_Schematic.png

6CA7's screens have no power source, just a cap to ground. The cathodes are tied together, but appear to go nowhere. They need fixed bias on the grids, or some amount of cathode resistance to set a bias point.

Maybe most of this is mistakes in the creating of the drawing, but, maybe worth checking...

Ed

Logged

Ed, K8DI, warming the air with RF, and working on lighting the shack with thoriated tungsten and mercury vapor...
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4420



« Reply #28 on: November 22, 2021, 03:30:03 PM »



I removed the embarassing schematic. What a waste of time playing around with the DigiKey freebe...... following Blutos advice to Flounder, time to start drink heavily.

I'll replace it with something hand drawn that should be more accurate as to what I actually have. Ed, I do apreciate the input.

klc
Logged

What? Me worry?
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 429



« Reply #29 on: November 23, 2021, 06:36:00 AM »

I used to use EasyEDA on the web. It worked ok but did not like my schematics being hosted in China. I now use Kicad. Pretty straightforward if you have a cursory understanding of schematic editors. For some of the simple circuits, I take it all the way to having PCBs made. For the tube designs, I just use the schematic editing. All free.

It is really pretty easy. Install it. Create a project template. Find a suitable component. Drop it. Find another, and drop that. Change to line mode and connect the two. Do another. Rinse and repeat. Easy.

There are tutorials on youtube if you prefer. PM me if you want to know more. Be glad to help.

John
Logged
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4420



« Reply #30 on: November 23, 2021, 02:10:40 PM »


J,

Thanks fer the offer. For now, I'll stay with the hand drawn schematics. I've waisted far too much time with the digiKey freebee with very little return. I wanted to get this mod deck finished by November, and get the RF end going.Maybe sometime in the early spring when I've got some more time.

I found my 100:1 probe and hooked it up to my scope. I'm begining to think that some of the osc was from my rats nest of clip leads, scope probes and jumpers. Last night I was able to get usable output before the parasitics showed up, and they seemed to be somewhat better behaved. I was able to play around with the 810s bias a bit. I think that I'll remove my choke to get the HV up a bit and play with the bias more. With the "Q" point set  ~ 25 mA, the output is cleanish most of the time.

klc
Logged

What? Me worry?
KB2WIG
Contributing
Member
*
Offline Offline

Posts: 4420



« Reply #31 on: November 30, 2021, 11:09:33 AM »

Another Over Weekend Update.
 I added two 100 Ohm, 2 W resistors into the grid leads of the 810ís. Iíve been led to believe that this isnít the best way to do things on a Class B amp. Anyway, almost no improvement in the osc department.  To recap, the amp didnít always go parasitic, but as the mic audio was increased it would go when it wanted to. I then added 2 of the same 100 Ohms to the plates of the 810ís. There was a marginal improvement, in that the oscillations sometimes could be stopped by changing the grid bias. It also appeared that the oscillations didnít occur as often as before Ė although this couldnít be quantified.
I have a 1.5K, 10 W resistor, so in desperation this morning, I placed it across the grids of the 810s (which is affectively across the driver transformer). For a short test, No parasitic oscillations observed with a few yeah a loowss and moving the bias around. So it looks promising.

klc
Logged

What? Me worry?
K8DI
Contributing
Member
*
Offline Offline

Posts: 243


« Reply #32 on: November 30, 2021, 03:19:13 PM »


I have a 1.5K, 10 W resistor, so in desperation this morning, I placed it across the grids of the 810s (which is affectively across the driver transformer).


This.

Tube amps run with no load (like a disconnected speaker) often oscillate.....


Class B modulator grid impedance goes from very high to kinda low when it starts drawing grid current. If it is fed by a transformer output tube amp, the previous stage is working into an essentially open circuit until grid current happens. Thus the need for a beefy driver to avoid distortion when suddenly the current jumps, and the practice of minimizing the jump by swamping the grids with a moderately low resistance so current is always flowing.  That's basically what you've done, added the swamping resistor. In my HB class B 812A modulator chassis, I have a 4K resistor across the driver transformer. One has to balance lower values meaning less of a glitch when drive gets to the level of bias (class B entered/current drawn/impedance drops like a stone) vs. higher values not using up driver power so the driver doesn't clip before the class B stage.


Ed
Logged

Ed, K8DI, warming the air with RF, and working on lighting the shack with thoriated tungsten and mercury vapor...
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 7923



WWW
« Reply #33 on: December 02, 2021, 12:05:42 AM »

For 10 bucks you can get these on amazon...

https://www.amazon.com/gp/product/B087PRYGRT


Anything from China or amazon gets tested, especially cheap high performance rectifiers that should be expensive. Do they get hot when 2A DC is run through them for an hour? Had a bad Amazon Chinese diode experience before, parts turned out to be counterfeit. Solution is use old USA parts or buy new diodes from authorized distributor like digikey or equivalent.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 7923



WWW
« Reply #34 on: December 02, 2021, 12:51:13 AM »

klc:

Putting 100 Ohm resistors in a class AB2/B grids of a modulator is a poor thing to do because the peak grid current is quite high on those tubes and so you weaken the drive, create a grid leak sort of non-linear bias voltage but at audio frequencies. It's like a high level audio compressor effect.

When you drive the modulator hard, as with 180V peak to the grid, the grid can consume 250mA peak, and that means you lose drive voltage of 25V across the 100 Ohm resistor.
page 6 of https://frank.pocnet.net/sheets/049/8/810.pdf


Also, what you and Ed just posted about swamping/loading the 6CA7 speech amp stage with a resistor across the 810 grids. 100% agreement!!

I skulled 1.63K for the swamping resistor to consume 10W based on the 2KV plate supply class B audio data from the above data sheet, which seems to correlate with the Terman-Jones tradition of burning 1/2 of the driver power in a swamping load for class B modulators. YMMV. Bump it to a 20W part for margin though? -or measure later anyway and see.

(amusing because a 3-500Z modulator can consume ~700mA peak grid current so that swamping is more like 500 Ohms and I spent a couple months figuring out the class B driver transformer there was a terrible match for those low-Z grids, but would have been good for 810/304/250T/833s)

Looks like you got it.  8-)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Pages: 1 [2]   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 0.038 seconds with 19 queries.