My 50W, 8ohm amp can't drive my 86db, 4ohm speakers? WTF??

tboooe,
I've written about this before but let me write this again - there are watts & then there are watts. All watts, IMO, are not created equal. What i'm talking about here is the power amp's ability to deliver load current. 
For example you could have a 50W/ch into 8 Ohms using a 200VA transformer. A 50W/ch amp most likely uses a 20VA secondary voltage in the power transformer. So, a 200VA power transformer can deliver 10A to 2 secondaries (stereo amp). Thus, each channel gets 5A max. You could have another 50W/ch power amp that uses a 400VA power transformer. The ability to source load current just doubled meaning that the power amp with a beefier power transformer will most likely drive a lower impedance speaker better.
And, this is the difference you are seeing with your Evolutions vs. your Sonos amps.
Yeah, you are right that the amp would most likely clip at higher power levels.
There are 2 types of clipping - voltage clipping when you turn up the volume & the voltage swing on the output stage exceeds the DC power rail of the power amp. And, the 2nd is current clipping when the power amp does not have sufficient current to deliver into a low(er) impedance & (current) clips.
I don't know what that buzzing sound it - maybe current clipping. It would depend no the size of the Sonos power transformer  which would give you an idea of the max steady-state current possible. Peak current will be higher.
Like a wrote above the 50W/ch channel amp probably has a 20VA secondary which tells me that the DC rail voltage is +/- 28.3VDC unloaded. It probably drops a few volts when there is a music signal passing thru the amp i.e. when the amp is delivering a load current. Let's say the rail voltage drops down to +/- 25VDC. So, if your program material has peak voltages exceeding this rail voltage at 70-75dB SPL then, yes, the amp will clip each time the music signal exceeds 25V. With the SF being 4 Ohms & 86dB efficient it does not look likely. Thus, it could be current clipping (power supply not robust enough).  
 

tboooe OP957 posts03-26-2016 11:10amLooks like my Accustic Arts integrated amp is repaired.  The parts that failed were:

1) $83.99 - 15V Regulators
2) $49.60 - Rectifier

Total bill with labor came out to around $500.  Ouch!  However, its a lot less than buying a new integrated amp.  Hopefully this repair will keep my amp chugging along for at least 2-3 more years!  Then I think I will step up to a tube integrated!

As I am no electrical engineer I am not sure what these parts do and if these are common failures in amps.  I am just happy to be able to listen to my system again soon!

wow! is that true - that's what was broken?? I'm surprised....

both parts related to the power supply.
A regulator is an electronic ckt that ensures that a steady 15VDC is output to the electronics its supply power to. Steady in the sense that the peak-peak ripple is kept as low as possible, good line regulation i.e. if the incoming power supply to the 15V regulator fluctuates a bit, the output 15V is held as steady as possible under those circumstances, good load regulation i.e. if the client electronics ask for more current, the output 15VDC will not sag voltage-wise.
I'm just wondering how a 15V regulator can fail? It's under-spec'd for the client electronics it supplies power to, an AC voltage surge during a bad electrical storm (that would take out more elements in the power supply elsewhere in the power amp), some ancilliary component failed (like a decoupling cap or a bias resistor, etc) that caused the 15V regulator to malfunction momentarily & fail.

A recitifier is also part of the power supply that converts incoming AC sinusoidal voltage to a DC voltage (with a large amount of ripple, which is eventually reduced/eliminated by those large coke-can power supply capacitors). Recifiers are essentially diodes - sometimes/often high-speed diode known as FREDs or Fast REcovery Diodes - which are often rated for high current (60A, 100Am, etc) & high-speed operation. The need for high current comes when you 1st switch on your power amp - there is a surge of current required to charge up those coke-can & other power supply capacitors. This surge lasts for 2-5 seconds depending on the amount of capacitance & it all has to flow thru the recifier. For example, my amp when 1st turned on takes 37A per channel for those 2-5 seconds.
Again, I'm wondering how a rectifier can fail - it's under-spec'd for the job, an AC power surge during a bad electrical storm, power supply capacitor does not have a DC path to ground to discharge when the amp is turned off thus discharges thru the rectifier diode (bad design) or failure due to age i.e. fatigue due to repeated on/off cycles - for this amp has to be really old. Even then, i dont believe this - my 1977 Yamaha integrated is working with its original rectifier diodes. The power supply caps were changed, of course, due to age of the unit.
Hmmmmm.....unless there is a good explanation for the failure of these 2 power supply parts, there's something fishy going on in your integrated amp unit. I'd keep a close watch on it. If these parts failed due to an anciliary part failing & that anciliary part was not replaced, you run the risk of having the exact failure repeat over time.... FWIW.

Shorted power rectifiers are a common problem in many amps. There is nothing surprising about this, nothing to do with any of the issues you mentioned (although they can be influences). Sometimes they just fail for no reason.

oh! ok.

I wonder if turning on and off my amp daily had a part in this. I think I will consider leaving it on all time, though I really dislike that thought since I only listen to my system 1-2 hours a night.

this could certainly have an effect - there’s a good reason why a lot of electronics (esp. IT related electronics & computers, laptops, external HDD, etc) are left on 24/7. Power cycling electronics does introduce a current surge each time thru the rectifiers & the power supply caps & the AC fuse. And, over time this will fatigue the said components causing failure.

There’s a lot written & discussed re. leaving s.s. electronics on all the time - just search the Audiogon archives & you should have plenty of good reading material.