Changing the gain of an amp stage

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello to Everyone !

I see a lot of integrateds which consist practically in an opamp based line stage + power amp stage, with or without a tone control circuit in between
Usually the voltage gain in the power amp stage is low as a part of the voltage gain is realized in the opamp
Taken by the ambition to get a more transparent sound and being quite ignorant of electronics, I am thinking about bypassing completely the line stage and the tone controls and increasing the gain of the power amp stage
In the end there will be just a input pot followed by the power amp stage with higher gain
Changing the feedback resistor could lead to stability problems ? :rolleyes:
Once I reduced the gain and fried the power amp stage :eek::mad:
I wanted to build a power buffer, high current low voltage gain, to use it along a tube headphone amp as line stage
So I am thinking to go the other way, i.e. increasing it :eek:

Thanks a lot and kind regards, :D

gino
 
Last edited:
Good evening !

I am attaching the schema of the power amp stage
As I said above in front of it there is a opamp line stage and a tone control stage
I like the idea of "pruning" both these stages
Nevertheless I do not know which is a normal voltage gain for a power amp stage
In this case I see a feedback 8,25 K resistor and a 475 ohm resistor to ground
In the end there would be just the 50K volume pot before the power amp stage
Any suggestion would be very welcome and appreciated
Ciao

gino
 

Attachments

  • power amp stage.JPG
    power amp stage.JPG
    58.2 KB · Views: 181
Good evening !

My doubt about the gain being too low is that in a similar power amp stage (that I am attaching as well) with a volume control the overall gain is set much higher
But I have at hand the previous amp
So I am thinking about raising its gain in order to be able to use it as an integrated but just with a pot in front of it (like in this second model from the same brand and similar power)
The first is from a Rotel RA-930AX (that I own)
The second stage is of a RA-820B

Thanks and regards,

gino
 

Attachments

  • power amp stage 2.JPG
    power amp stage 2.JPG
    58.5 KB · Views: 166
Last edited:
in the first amp, the voltage gain is set by the ratio of (R635/R611)+1 and is about 18, the second amp, the gain is (R639/R611)+1 and is about 71. if you really wanted to you could make R635 (or R639 depending on which schematic) a 50k pot, and you would have an amp with a voltage gain variable from 1 to a little over 100. you can treat the power amp as an op amp.
 
Last edited:
As said, you can treat the power amp like an opamp. IMO, you'd be well off to get something like Walt Jung's Opamp Cookbook, or maybe there's lots of info online, but study up on the basics of feedback and gain. The tradeoffs for a power amp are very similar to an opamp and you'll see how stability and other parameters are affected. My worry is that whatever gains you'll make pruning input circuitry, you might lose in fooling with the gain of the power amp. Without some test equipment it's hard to know what's really going on after a change.

Conrad
 
I think that maybe the existence of C611 shows that the designer recognises that all of the grounds have a different job and terminate in different parts of the amplifier/chassis. But, he has typically used a schematic drawing program that does not differentiate between all these different grounds.
 
you would want to be careful about how you add a potentiometer as a gain control, since stray capacitances of the potentiometer and it's wiring could cause stability problems. a better solution might be a CdS photocell and LED opto device with the LED controlled by a remote poteniometer
 
Good evening and thank you very much for all your directions !:)
I understand that to try to change the overall gain could lead to severe problems :(

But still I have this more general question
Here are two integrateds of similar power, designed in different ways
1st) pot+opamp+power amp stage
2nd) pot+ power amp stage

If the feedback (FB) is a bad thing, and the more the FB the worst the sound, the amp 2nd) seems the better design,
because in amp 1st) we have a lot of FB in the opamp and then a lot of FB in the power stage
In the 2nd) we have no opamp and less FB in the power stage compared to the very similar amp stage in the 1st) amp
I do not have a direct experience of the rotel ra-820B, but I read very positive opinion about its sound
Something of a classic
I have the 930ax and while I find it good I think it could be better sounding if modified in the 820B direction
I am finding it a little dinamically "constricted"
In agreement with the KISS principle or the not completely proven rule that less is more sometimes
I understand that top audio designers employ the less feedback they can
(Sorry for the globish :eek:)
In conclusion, as the power stage can provide all the gain that is needed, why put an additional opamp in front of it ?
is it not too much ? :rolleyes:
Even the configuration with a 8K feedback resistor and 470 ohm to ground could be enough with cd level signals, I think

Thank you very much again
Kind regards,:D

gino
 
i don't know where you heard feedback is a bad thing.... it's responsible for the very low THD figures and low output impedance (high damping factor) of most modern amplifiers. the only time feedback contributes to IMD, is when a poorly designed amp has slew rate limiting and the amp can't keep up with the feedback, but such a case is extremely rare these days, since it's a pitfall most designers know how to avoid. be aware that as you reduce the feedback to increase the gain of the amp, distortion will increase proportionally. if the amp already is a well designed amp with THD below 0.05%, the increase in THD shouldn't be audible. some people claim they can hear the difference, but most people's threshold for noticeable distortion is about 1%-2%. what may be slightly audible is the effects of increased output impedance as it interacts with the speakers.
 
i don't know where you heard feedback is a bad thing....

Good evening !
I have to say that I read articles of Mr Pass and Mr Curl stating so
And I have been extremely influenced by these papers
By the way the debate is still open, I understand
In the above mentioned cases it seems to me that Rotel has gone from simpler topologies to more elaborated ones, employing more FB
Clearly the newer models should be the better sounding

it's responsible for the very low THD figures and low output impedance (high damping factor) of most modern amplifiers

and this is very very good of course
Nevertheless it could be the case that an optimum value of FB do exist
If I am not wrong the value in dB of FB is the difference between the open loop gain and the gain when FB is applied
When FB is zero we have high THD and high output impedance
But when it is too much other problems, that I do not know, could appear

the only time feedback contributes to IMD, is when a poorly designed amp has slew rate limiting and the amp can't keep up with the feedback,
but such a case is extremely rare these days, since it's a pitfall most designers know how to avoid.
be aware that as you reduce the feedback to increase the gain of the amp, distortion will increase proportionally.
if the amp already is a well designed amp with THD below 0.05%, the increase in THD shouldn't be audible.
some people claim they can hear the difference, but most people's threshold for noticeable distortion is about 1%-2%.
what may be slightly audible is the effects of increased output impedance as it interacts with the speakers

Thank you very much for your very kind and valuable advice
On this basis I will limit myself to bypass the tone control section, the balance and I will try to place better quality coupling caps between the line and the power amp stages
I already have some PP caps that could work, some 2,2 uF
So FB is not the evil I thought. Very good !

Thank you very much again
Kind regards,:D

gino
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.