Old wives tale. What about a stereo amp reproducing signals panned hard to one side? The two halves of a stereo amp are as independent as their common power supply allows. You can treat them as two separate units and drive main loudspeakers from one and monitors from the other.

I sense another angle to this question.
Would you use the amp to drive different speaker loads?

Example: I rig Aux 1 to amp Input A, Aux 2 to Input B.
I rig Output A from the amp to a 8ohm speaker, and Output B to an 8 ohm speaker.
All well and good.

But what if I added a second 8 ohm speaker in parallel to the one attached to Output B?
The load on the amp becomes 4ohm, yes?

I see one of the following happening:
- Amp remains functional according to desires.
- Amp heats up more.
- Amp pushes one side incorrectly.
- Amp spontaneously combusts.

I vote for spontaneous combustion. :lol:

The two sides of all power amps that I have worked with function completely independently. There are OP amp pres, and MOSFET power drivers or maybe a push pull PNP NPN amp, and they really don't give a hoot what the other side of the amp is doing. The side driving more load will get hotter, and sag the power supply, but the other side isn't going to care. The sag in the power supply may make the output slightly less linear, but you aren't going to blow anything up or burn anything.

So if you have a stereo 500W program Amplifier, expect that the max you can drive each side is 250W.

Boswell wrote: Old wives tale. What about a stereo amp reproducing signals panned hard to one side? The two halves of a stereo amp are as independent as their common power supply allows. You can treat them as two separate units and drive main loudspeakers from one and monitors from the other.

Good point. A great power amp is actually dual mono with discrete power supplies. Unfortunately, most crap made for a price point today is shared power supply garbage.

The problem with shared power supplies is this. If you are running mains on channel A and monitors on channel B, and mains are requiring more power, then it is band and power limiting channel B. Can't get more water out of a glass than what it will hold.

In some situations this works. In some it doesn't. There are infinite variables to say yes or no.

GeckoMusic wrote: I vote for spontaneous combustion. :lol:

The two sides of all power amps that I have worked with function completely independently. There are OP amp pres, and MOSFET power drivers or maybe a push pull PNP NPN amp, and they really don't give a hoot what the other side of the amp is doing. The side driving more load will get hotter, and sag the power supply, but the other side isn't going to care. The sag in the power supply may make the output slightly less linear, but you aren't going to blow anything up or burn anything.

So if you have a stereo 500W program Amplifier, expect that the max you can drive each side is 250W.

I beg to differ. I can give you many examples of amps that will crap completely because of shared power supplies. Just about all manufactures who build for a price point in the MI market have this problem. It is what it is.

sheet wrote: The problem with shared power supplies is this. If you are running mains on channel A and monitors on channel B, and mains are requiring more power, then it is band and power limiting channel B. Can't get more water out of a glass than what it will hold.

I agree that a shared power supply induces some problems, but don't see your logic on the band and power limiting of the low power channel.

Do you mean "band limiting" as in the frequency band? Are you saying that if I have a heavy load on channel "A", this will make channel "B" have a poor bass and treble response? I could see it being slightly non-linear due to sag, and possibly some bass cross talk, but not band limited.

If channel "A" is drawing enough power to sag the power supply, and "B" is running lower than "A", then "A" is power limited, not "B." If you crank "B" up to where the power supply is limiting it, "A" will droop a little in power, and both "A" and "B" will be at the same power and both be limited. Not typical in a monitor / mains situation.

Yes. Some manufacturers (Peavey with their DPCs, etc) even went so far as to market this feature as a benefit.

Hmm...

If you run a power amp at nowhere-near-to-properly-gain-structured levels where the power supply isn't being a problem, then loading each side differently matters?

Codemonkey wrote: Hmm...
If you run a power amp at nowhere-near-to-properly-gain-structured levels where the power supply isn't being a problem, then loading each side differently matters?

No, we're not talking here about gain staging, we're talking available output power. I've designed many power amps in my time - you may even have used some of them.

Any power supply has voltage limits, current limits and a source impedance for each rail. If it supplies two power amps driven independently, then the antics of one amp can change the instantaneous limits that the other amp has to work within. Note that it is within the limits. Most tolerable operation of power amps is within the limits, not on the limits. Reasonable use includes your being able to treat the two channels as two completely independent amplifiers and hear no difference. That applies to loading (loudspeaker impedance), rated power levels, gain, heat dissipation or whatever.

As a slight aside, the case of running a two-channel amplifier in bridged mode is interesting, as it guarantees that the positive and negative power rails are loaded equally at all times. This enables the designer to specify a greater output power in bridged mode than the sum of two independent outputs.

My interpretation:

You meant that loading one side at 8ohms and one at 4ohms makes no difference if the amp is pushing less than it could, it still doesn't matter what the load is, it will still work.
BUT you also imply that if loaded differently, the limits of max output power change on each because one side is being driven harder therefore less power is available to the other side.

"I've designed many power amps in my time - you may even have used some of them."
Unless you make cheap stuff that ends up in Phonic mixers, probably not. ;)

Thankfully I haven't ever had a call from Phonic in my career, as I hate saying "no" to clients.

All power amps at this level are designed to a marketing price point, and that dictates the component budget, forcing the designer into compromises. Sagging in the power supply is not generally a problem, as long as you are careful with the cable routing, star points, connector contact resistance etc.

In my designs, I at least made sure that the rated power was available from each channel no matter what the other channel was being asked to do. It usually resulted in being able to get more than rated power under normal operations, especially at the higher load impedances where the currents are lower.