Offset servos needed?

[Tsanford]

I am working up the design of an amp thats going to be switchable between bridge/parallel mono and stereo/parallel operation. The AN-1192 bulletin(BPA-200 example) suggests using servos (an integrator)to cancel dc offset.
My question is would it be equally good to put in a precision trimmers and hand trim each output? or does it need to be actively compensated for?
Any help would be appreciated.

[memorex]

I built that circuit. I just used a multimeter and used resistors within +-.1% tolerance and had perfect results. Servos are overkill.

[JojoD818]

Overkill, I agree.

Just try to match resistors as close as possible and your fine.

JojoD

[bobolix]

Trimming and selecting yields a "true DC". DC servo supresses transfer below certain frequency (although it can be low enough) and the complete circuit can have quite long settling time (e.g. after switch-on). But, I don't assume the DC servo to be an overkill - I use it, if the trimming needs to be avoided.

[Tsanford]

Thank you for the responses. I do not understand this response however.

bobolix wrote:
Trimming and selecting yields a "true DC". DC servo supresses transfer below certain frequency

Does this mean the servo will act as a high pass filter?

[johnnyx]

Originally posted by Tsanford
Does this mean the servo will act as a high pass filter?

Yes.
DC should really be called Zero Frequency in this case, because that is what is meant. So the amp will have no gain at zero frequency, it will be attenuated by the servo loop, yet it must have gain at audio frequencies to hear the output. The frequency of transition between the two is determined by the CR network in the servo, usually the servo amp is wired as an integrator or single pole filter.