[sdiy] Prophet 10 op-amp swaps?

[rburnett at richieburnett.co.uk]

It really just comes down to whether a particular op-amp is designed 
for use with low gains (close to unity) or whether it's optimised to 
perform better with higher gain.  Generally op-amps with a high 
gain-bandwidth product are chosen when you either want loads of gain, or 
loads of bandwidth, or a medium amount of both at the same time!  
They're not necessarily a good choice for a low-gain audio buffer.

The datasheet for the LM675 power op-amp for example explains that this 
device is only stable for gains of 10 or more.  Sure enough if you 
choose your feedback resistors to try to get a gain of only 5 it 
oscillates wildly at a couple of hundred kHz, because the loop-gain 
around the op-amp itself is just too high.  The key phrase in the 
datasheet is "The amplifier is internally compensated for gains of 10 or 
greater".  On page 6 they show you the correct way to set up gains of 
unity.  The arrangement of R1 and C acts to bring down the loop gain at 
very high frequencies, thus making the amplifier stable in closed-loop 
arrangments with input-to-output gains of less than 10.

-Richie,

PS.  I'm not recommending the LM675 device be used here, just pointing 
out a device where the datasheet discusses this stability issue and 
shows how to work around it.



> It seems to me the line lies somewhere halfway between 1 MHz and 1 GHz
> - logarithmically. The distance is roughly ten octaves, and if you go
> 5 octaves up from 1 MHz then you're at 32 MHz, which seems to be
> around the value when problems start if you don't add said
> compensation. I'm sure it works differently for different op amps
> though. But if you're closer to 1 GHz than 1 MHz, it seems likely
> you're going to run into trouble.
> 
> Cheers,
> D.
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy