SS bias pot tweaks?

I have a Audoninics cc3 and one channel is warmer then the other . I checked the dc offset and the cooler side was 18mv and warmer side was 3mv so i was told to get them as close to 0 as possible. They are now set at just about 0.to 1.0.
Now i do have bias pots ,but where do I take the measurements from and what should be?

Thanks

Be careful - my amp is in the shop because I very stupidly shorted something out when I was measuring the bias voltage.

Mark

I should mention that having said all that, I would definitely not attempt this procedure on an amp with multiple pots on each channel.

Points are very well taken, guys. I do agree you must be very careful but I'm just convinced by my own experience that the only way to pinpoint it is by listening. I've listened to lots of high-end stuff over the years but never heard the kind of sharp spatial positioning between instruments and vocals that I hear on my own stuff. Limitations in source material is simply apparent and not obtrusive or fatiguing. Ambient sound is so present that I can tell I'm hearing "all" of the information. I've been doing this for many years and have never blown an output or fried anything. Incidentally, I wonder if the self bias function in my Meridian 559 is why my other amps sound so much better?

Without specific instructions and a quality DVOM to make accurate measurements makes for a foolhardy endeavor. Turning screws and tuning by ear = bad idea. I have done bias adjustment on Classe Dr-8's. The measurement for that amp was a range of 70-96 ma per channel divided by the four bias pots per channel. An infrared gun is handy to measure heatsink temp as well. The amp needed to be powered several hours at idle prior to making any adjustments. I preferred 88ma or 22ma per pot.

Csontos - I agree that bias might not be set to optimum on particular amp especially for the best channel match. If you can find sweet spot by listening - great. You might want to measure just to know where you are. It is voltage drop measurement on one of two large resistors (usually around 0.3 ohm) near power transistors to determine current.

Heat sinks in class AB amps are undersized (average music power is only few percent of peak power) so be careful.

I'm aware of the overbias issue. However I was referring to channel balance. I discovered the bias pots in my Meridian 105's back in the late 70's and their significance to the overall performance of the amp. I'm no technician but confident enough to undertake such an adjustment. I've consulted a few techs regarding this and was assured the damage threshold is far past the degrees I've been turning them. I took my Bedini 100/100 in to have the bias adjusted once and found it had deteriorated the sound a little so I took the top off and simply "matched" one side to the other, barely moving the screw back and forth on one channel until it was "tuned in to the other side" while listening to music. The result was phenomenal. Heat is the killer here. Typically clockwise is the direction of least resistance, hence more heat. Many amps are "underbiased" for warranty purposes which serves to undermine their potential performance. Optimizing such an amp "speeds" it up, makes it noticeably more linear and deepens the sound stage. If you are not noticeably increasing the heat of the outputs your not doing any damage and just nudging the screw does not when your matching one side to the other. From my experience with doing this excercise it has become clear to me that using instruments to set bias merely ballparks it and then using your ears you can fine-tune it. The further off one side is from the other, the flatter more lifeless the sound will be to the point of no soundstage at all, just right and left channels. Transient response will also suffer practically in that the amp will sound "slower". You also will lose the ability to hear it's full frequency extension at both ends of the scale.

AFAIK bias should be set to particular value. I asume you expect better performance setting bias higher. Increasing bias most likely won't improve the sound (otherwise everybody would do that) and is called overbias. Bias is usually set around 50-100mA and increasing it, while reducing slightly nonlinearities(that are reduced by negative feedback anyway), increases area of larger transconductance (gain) where two transistors work instead of one causing nonlinearities as well.

The problem is not with the bias but with negative feedback. Once gain before feedback was set high (like 4000) to make output linear then damage is already done and tweaking bias will not help much. Class A has gain before feedback as low as 200.

There are measurements of amplifiers' distortions with different bias setting that show clearly that increasing bias above some point will increase distortions.