Op-Amp Self Destruction

[Tim__x]

I've been told that some op-amps (specifically the NE5534/2) will self-destruct if placed in a unity gain configuration with a short between the out and -in, and that they need a resistor around 5-10k in the loop. Is this true? It doesn't sound true but I just wanted to check before I go burning up a bunch of op-amps.

[Centauri]

I've never had any problem with them in this configuration.

Cheers

[benny]

I've been told that some op-amps (specifically the NE5534/2) will self-destruct if placed in a unity gain configuration with a short between the out and -in, and that they need a resistor around 5-10k in the loop. Is this true? It doesn't sound true but I just wanted to check before I go burning up a bunch of op-amps.

yes, i think it's true. i'd say because when the signal comes out, it feeds back to the input where it started, only it's amplified, and this amplifies more, and feedsback again, more and more, until it eventually causes self destruction in the op amp. yes, i'm pretty sure it is true. the resistor is there to drop the signal down enough so it doesn't present an oscilation problem.
I know it probably seems like it would clip, and limit from overloading, but as it amplifies more and more, all the wave peaks would be flattened out to straight DC... which i think you might find is bad to put on the input for your device.

cheers

[sreten]

Possibly if you leave out the compensation capacitor of the 5534,
the 5532 is internally compensated for unity gain, so no problem.

sreten.

[sreten]

Originally posted by benny


yes, i think it's true. i'd say because when the signal comes out, it feeds back to the input where it started, only it's amplified, and this amplifies more, and feedsback again, more and more, until it eventually causes self destruction in the op amp. yes, i'm pretty sure it is true. the resistor is there to drop the signal down enough so it doesn't present an oscilation problem.
I know it probably seems like it would clip, and limit from overloading, but as it amplifies more and more, all the wave peaks would be flattened out to straight DC... which i think you might find is bad to put on the input for your device.

cheers

Sorry but simply not true, and the resistor will have miminal effect,
its only included to match the other inputs effective impedance.

What is important is the op-amp is unity gain stable, some are not.

sreten.

[Centauri]

Originally posted by benny
because when the signal comes out, it feeds back to the input where it started, only it's amplified, and this amplifies more, and feedsback again, more and more, until it eventually causes self destruction in the op amp.

We are talking about a unity gain configuration here - where the input is fed to +in and output is tied to -in, not shorting signal output and signal input.

Cheers

[benny]

thanks for letting me know... what i said was just something i thought up at the time why it could be a problem, but i guess i'm wrong.

i'm not actually aware of what a unity gain config is. could someone please inform me?

cheers

[Devil_H@ck]

Unity gain = 1x gain = no amplification
One would use it in a buffer circuit.

[RetroAudio]

Tim_x - you might surf the web and get some data sheets on these opamps. Often they will have that kind of info. You sure it's not unity gain inverting you're talking about?

[peranders]

Tim_x the very simple answer is NO, but some opamps may need some attention and the best way to find out about this is to read the datasheet and application notes is there are any.

LM3886 is typical case where some parts seems to be unnecessary but in fact serves a purpose which is very hard for even a trained person to see.