Gainclone...No gain, YES pain.

[JAZZ2250]

I tried gainclone with gain=1, and it got pretty hot in a very short time. I think, according to the others' comment in this thread, it is due to an oscillation. I tried both T and TF package, and both of them got hot. Actually, I was thinking to use a gainclone as a power amp without any gain, but it doesn't seem to be possible because of the nasty oscillation.
What I'm thinking right now is that if I set the gain of gainclone to, say, 20 and put a voltage divider at the input before the coupling cap, and hook it up to a tube preamp with a maximum gain of 20 dB, would it be worth trying? As I heard, the main benifit of using tube amps is at the voltage amplification. In the configuration stated above, voltage amplification is done first at the tube stage, and signal gets lower by the voltage divider, and then voltage is amplified again by gainclone. In such a case, there wouldn't be any benifit of using a tube line stage, IMO.
I would like to know what others think about this configuration. Is it gonna be worth trying or a waste of time and money?
Thanks in advance.

[Peter Daniel]

Doesn't make much sense, such configuration. First you amplify, than attenuate and then amplify again.

[analog_sa]

The only way to get gain around unity from these chips is to use inverting topology.


cheers

peter

[JAZZ2250]

Could you explain more about it? Thanks.

[JordanG]

Originally posted by JAZZ2250
Could you explain more about it? Thanks.

I'd recommend reading this. It provides a nice introduction to the inverted gainclone. Also, here is a forum where people like to talk about and tweak IC based amplifiers.

hope that helps

--Jordan

[tiroth]

Amplify, attenuate, amplify is just going to raise your noise floor.

You realize that the LM chip is not spec'd stable with gain < 10?

[JAZZ2250]

Yes, I realized why the gain less than 10 is not recommended by NS. Definitely it doesn't seem to be a good idea to run the amp with such a low gain though I cannot hear the oscillation. It should also be applied to non-inverting configuration, right? I gotta try it.
Mmmm, then I'll have to think about which power amplifier, which is simple to build, will go well with tube preamp. Another tedious search is about to begin now. Any recommendation, please?

[ThorstenL]

Hi,

I tried gainclone with gain=1, and it got pretty hot in a very short time.

Then you did badly wrong. Big booboo. You CANNOT get unity gain in non-inverting topology, you can extremely easily get unity gain in inverting mode. Simply use a pair of 220k Resistors, one from the Amp input to the inverting input, the other from the inverting input to the Output.

NOW observe the magic. Place a 22k resistor between the inverting input and ground. Then place an 18k resistor between the positive input and ground. Unity gain and stable. Easy as pie.

Sayonara

PS, this same sort trick lets you use the OPA637 as I/V converter without osccillation. It sounds notably better than OPA627 in this position but is not unity gain stable....

[JAZZ2250]

I used all 47k resistors. Let me see what I did wrong... I think I didn't use one resistor which is 22k in your recommendation. I directly connected the signal to a resistor which is connected to the inverting input. I thought the one (22k in yours) was not necessary because inverting input could be thought as a virtual ground. Wrong? Sorry for my poor knowledge, but I thought I had read it somewhere. Do you think putting that resistor will make a difference? Please let me know, Sayonara Sang.

[analog_sa]

If i may volunteer a reply the 22k resistor is there to provide stable dc reference. A problem of 'textbook' inverted amps is that the offset voltage varies according to the dc resistance the inputs see relative to ground.


cheers

peter