• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

6AN8 Pre & concertina, what seems to be the problem?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
If i lower that 600k resistance to get higher VG2 supply, it affecs and pull the plate voltage to lower value. so I stay with 600k.

Sorry, Junm, misunderstanding ---

I did not suggest a change in the 600K. I am saying, that even with a 10M (high) impedance meter, the screen voltage you read will be slightly lower because by 'attaching' the meter, one also places the meter impedance, 10M, from screen to earth, thus causing the screen voltage to read slightly lower than it will actually be without the meter. This is possibly of academic interest; the voltage counting in the end is the anode voltage, (going to the triode grid).
 
yes, the meter's 10meg resistance may have an effect...
and yes, the pentode anode voltage is what counts as it is direct coupled to the triode concertina grid and so affects the triode operating points...

the triode cathode voltage will be higher than its grid voltage to show its grid is in negative territory..
 
The real bug in these designs is the first stage direct coupling to the concertina, such minor voltage changes plus the resistor characteristics and all goes wild. There could even be a hint of RF oscillation about too. The concertina works with slightly below unity gain so the voltage chaos is with the pentode section. I've known signal pentodes to vary so much as to render the subsequent stages completely out of their operating regions. Those who mention messing with the output stage 0.3uF coupling caps, do so at their peril as they effect the low end loop feedback gain. Leave them as they are. Focus on the 6AN8 resistors. I would change all for newer ones. The design must have worked at one time with the values shown.

One of my verification methods is to actually breadboard the front end with concertina only, with the correct B+ voltages stipulated on the drawings.

rich
 
The real bug in these designs is the first stage direct coupling to the concertina, such minor voltage changes plus the resistor characteristics and all goes wild.
RCA had a solution for that :)
DC-feedback from cathode concertina to screengrid :cool:
Mona
 

Attachments

  • RCA-30W-1959.jpg
    RCA-30W-1959.jpg
    50.8 KB · Views: 129
Richwalters,

Your post # 83: Exactly; I also intimated that in my post #60. When a design uses a cathodyne to directly feed tubes of EL34 ilk and upwards, this problem exists, and in some quite well-known designs. This unless the h.t feeding the phase splitter is higher than usual.

@ Ketje

Yep. I have used that 'trick' ever since I started using the cathodyne, even for lower signal output. One does need to balance the splitter output impedances - but I am confident members would realise that. A slight advantage of this is that direct coupling then allows a slightly higher dc voltage on the preceding anode.

(And OT again; sorry! One soon finds that the often maligned paraphase inverter is then (at lower h.t.) a better option than a cathodyne.)
 
Dynaco SCA35 used this to stabilize the 7199. That is however quite late
in the tube-era this idea got foothold. Still it's a very good idea.

MAny years ago I also tried this and found out higher thd..sonically something adrift to my ears. The 7199 tube is a pricy specimen now, but having a few, (dreadfully microphonic) discovered g2 volts can be varied so much to provide appreciable distortion, with same gain only that all the disadvantages and bugs produced can be ironed out by applying total global feedback for decent linearity ! (oh, how confusing). In short, one pays for the variables for using pentodes, that's why I always use them strapped as triodes. It wraps the undesirables up.
As for the problem stage our colleague submitted, the cue that the cathode voltage of the cathodyne should be roughly 1/3 of the anode. Unfortunately if he insists in not changing components but only tubes, then alot will be required til one with the correct voltages is found. By this time the socket would be worn out ! I am not sure since it was a Jap amp; a Japanese 7199 was ever around. If so, there lies the compatibility problem. My take is with o/p tubes out, to secretly alter that g2/screen resistor. The concertina in this instance has only a 3 external component count with 'zilch' gain.

r,.
 
Gee-whizz ....

Yes, good post by Rich as always.

Junm, staying with my previous advice to not immediately go changing stuff all over the playfield until .....

and thus somewhat OT: Pentode voltage amplifier design has its difficulties (if you want a free lunch, don't do hi-fi). There are many variables. E.g. I found (someone else's research) that the distortion figure vs. screen voltage is somewhat cyclic, although there is a shallow optimum somewhere. But then with spread in tubes ... :eek: Mercifully the variations are small compared to the complete amplifier's contribution. As inputs I prefer pentodes somewhat because of the independance of GFB stability of the input circumstances and useful gain - and low distortion, but depending on what was just said! And not everybody has the benefit of a distortion analyser (there I go again: measurements! :) )

I bow out at this point - am still trying to rig up something using an ECF80 for some investigation - but don't wait up, Junm. And don't become confused! - it is not the hostile territory it seems. By and large, after all said and done there are still many excellent amplifier designs out there.
 
Just checked my old stock of 7199/6AN8's and they were made by Richardson Electronics with Jap symbols. Here we go folks, around in the 1960's vendor circles and fun here as their pin'outs aren't the same as the 6AN8 Philips/Mullard gang manufactured in that era.

In my high power amps I used the ECF80 plus variants (here we go) all in triode mode with an input stepup transformer, to obliviate the swap voltage problem and also the dreaded noise level that the pentode gives. I was after consistency plus the advantage of a balanced input.
In verifying this I tried the equiv 7643; for my records is a nasty little tube made by nearly everyone in the US (for the mass b/w TV market in that era) which behaves like a guitar (F.H.S don't tap it !) whereas the dearer E80CF behaves far more consistently and quieter in pentode as well.

r.-
 
Gee-whizz ....

Yes, good post by Rich as always.

Junm, staying with my previous advice to not immediately go changing stuff all over the playfield until .....

Yes Johan, Tomorrow i will play again with new tubes arrived and restore things on its original values and see. if it will still have the same result well i think i have to do some mod as the one showed above by Jazbo.
 
Have done some testing again today. Placing a new set of 6an8 tubes and restore the resistor values. Voltage measurement has come very close to what is expected. Done some sound testing and now better needs only little adjustment and refinement.

Thanks gentlemen for your help it”s really the weak tubes that cause the problems and voltage at input.which will im going to find .

I wanted a little more gain. Can i adjust the feedback resistor to a higher value?
 
Another problem folks. In the preamp output has 25 vdc reading. Cap coupled on the filter section has voltage readings?

Everything in the amp section had close voltage reading with the schema. When i fed my cd out direct to the amp input the power is still weaker than specified 50w. In my opinion its only around 12w not 50w. Way far with b+ of 410 v loaded. Coming from preamp its a little bit powerful. Pream is isolted with cap so that 25v will be blocked.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.