Reducing gain in a SRPP circuit with ECC82?

[dwjames]

Hi all,
I currently have this SRPP circuit using ECC82 tubes

An externally hosted image should be here but it was not working when we last tested it.

I'd like to reduce the gain, ideally quite significantly to act as a buffer stage with as little gain as possible, but my knowledge of tubes is basic at best.
I understand that the ECC82 is considered a medium gain tube. Is it possible I could swap them out for a very low gain alternative and would I need supporting modifications to the circuit or is it just not that simple?
Thanks in advance,
James

 

[Dimitris AR]

Use negative feedback to reduce the gain .

 

[nigelwright7557]

You could remove the second valve and put a link in from the output of 1st stage to output of second stage.

 

[kevinkr]

I'm assuming this is stereo.. And there really isn't a lower mu triode that would work in this circuit. The most reasonable answer unless you are looking to turn this into a cathode follower with near unity gain would be just to add a pad at the input to reduce the input signal amplitude - not optimum but if the signal level is adequate it will be fine.

This strikes me as the gain block for a simple line stage, why not use it that way? I made something similar commercially about 2 decades ago.. (300V B+ and 870 ohm cathode resistors, about 14 dB of gain.) IMO this would be far better than trying to buffer that cd player (or whatever) to get some tube sound into a solid state pre. Likely the simple line stage on its own will sound better. Ask for more advice if interfacing with a solid state amp.

 

[Merlinb]

You could remove R12 and R22 (i.e., replace with short circuits). This would reduce the gain to perhaps x7 (17dB), but not much less.

 

[DeathRex]

Can't short R12 and 22, they provide bias for the upper triodes. You can move the output from pin 3 to pin 8, then you would have 0 gain. Easy enough to try.

 

[45]

Can't short R12 and 22, they provide bias for the upper triodes. You can move the output from pin 3 to pin 8, then you would have 0 gain. Easy enough to try.

Not a good idea as well because the lower tube, used as a cathode follower as it is, will have rather limited output because of the small cathode resistor.
Best ways for me are:

1)loop feedback

2) cathode follower removing the upper triode, connect +170V directly to the plate of the lower triode, (I would) change R13 to 1K instead of 270R, add a 15K cathode resistor in series with R13. R14 remains connected at the node between R13 and the 15K resistor. The 15K resistor goes to ground. This way the tube should run at about 105V/4mA. This modification can be done because there is an input cap.

 

[kevinkr]

You could remove R12 and R22 (i.e., replace with short circuits). This would reduce the gain to perhaps x7 (17dB), but not much less.

No you can't this leaves the upper tube in the SRPP without bias.

 

[tinitus]

funny, but I thought about using SRPP for driving output stage, and just now found this 6n6p curcuit

but isnt the SRPP often followed by a buffer because it should drive a fixed known low impedance load ?
not sure, but is it something about loosing its gain if the load is very high impedance....ehh, but that is actually what you want

 

[kevinkr]

Load impedance (reasonable values obviously) only affects distortion, used as a line driver the SRPP really doesn't need buffering.. Load impedance can be chosen for the distortion minima if that is really a concern.

 

[Merlinb]

No you can't this leaves the upper tube in the SRPP without bias.

Yes you can. Shorting out the resistors converts it into a diode-loaded gain stage. OK, it won't be an SRPP any more, but since this is a line stage I presume it will be driving a relatively high impedance load, in which case there isn't much point in an SRPP anyhow.

 

[lcsaszar]

The gain is about 20 dB. Select a different tube, the attached calculation sheet might help.

 

[Miles Prower]

Hi all,
I currently have this SRPP circuit using ECC82 tubes

An externally hosted image should be here but it was not working when we last tested it.

I'd like to reduce the gain, ideally quite significantly to act as a buffer stage with as little gain as possible, but my knowledge of tubes is basic at best.
I understand that the ECC82 is considered a medium gain tube. Is it possible I could swap them out for a very low gain alternative and would I need supporting modifications to the circuit or is it just not that simple?
Thanks in advance,
James

There isn't much you can do about it other than add an attenuation pad at the input. The whole deal with the SRPP is to increase the gain by means of bootstrapping the load of the bottom triode plate. If you want a buffer with minimal gain, you need a cathode follower, or the White Cathode Follower variation, instead.

Either that, or cut down the gain with NFB as a form of anode follower. That might cause problems as anode followers have a rather low Zi.

Also, the SRPP works best if it sees a constant load that allows the source and sink currents to be equal, otherwise, it goes out of balance, and distortion rises. That wouldn't be a problem if the input impedance of what it drives is high enough, then simply load the output with its optimum resistance.

There is one small signal type that has lower gain, the 12B4A, but that's a singleton triode (you'd need four) and you would also require a complete redesign in order to make it work. Even then, it wouldn't drop the gain to unity.

 

[kevinkr]

Yes you can. Shorting out the resistors converts it into a diode-loaded gain stage. OK, it won't be an SRPP any more, but since this is a line stage I presume it will be driving a relatively high impedance load, in which case there isn't much point in an SRPP anyhow.

Linearity??

 

[SY]

Linearity??

Probably not the goal or that circuit wouldn't be the one chosen.

 

[dwjames]

Thanks for the help guys 🙂
Unity gain would be ideal. I think losing -20db from the signal would put it about where I want it.
I think the proper long term answer is to build a new tube buffer in an external case with a new board, different tubes and higher output power supply than my current 170vDC. I'll probably go for Aikido Cathode Follower kit from John Broskie's site.
That way I can do it properly without compromises and I also end up with something I can use with a DAC or Squeezebox too.

So I think the quick fix for now is an attenuation pad on the input. Thanks for the suggestion 🙂
I've done a bit of reading and it seems there's a few types, am I right that a simple L-pad would do the job?
Could someone help me identify what values of resistors I'd need to get a 20db drop please?
thanks,
James

 

[tinitus]

dont know if it makes sense to use tubes at all if you only need unity gain

even Broskie had to realise a passive pot could sound better, sometimes

maybe take a look at Salas 6V6 preamp
I think it has 6db gain

 

[ruffrecords]

As someone else said NFB is the obvious answer. Lowers gain, distortion and output impedance. Easy to implement. 100K from output cap to grid of lower tube and 47K in series with input cap gives about 6dB gain.

Cheers

Ian

 

[tinitus]

sounds like he may not know where the grid is....

since he already have this preamp ... maybe someone could show him the feedback on schematic ?

 

[dwjames]

Thanks Ian,
I did have a quick search for info about feedback loops, but didn't get very far...
That sounds like it'd do the job nicely and 'Easy to implement' sounds just like the type of solution I'm after 🙂
Could you confirm whether I need the tube-side or output-side of the output cap and whether the 47k should go before or after the input cap please?

Tinitus, I know which the grid is, but only just..... lots to learn!
I'll try and get an amended schematic up and approved before I get the soldering iron out 🙂

thanks again,
James