Mattes (pronounced "Mat-ease") was one of the earliest (if not the earliest) to produce a 100 watt per channel transistorized amplifier with respectable specifications. In about 1968, I was able to buy several of their power amps, an integrated amp and one of their preamps on a special deal because one of our family friends was buddies with one of the Mattes engineers. Back then this kind of power was likely to blow out your speaker even if it claimed to handle 100 watts. I was in a rock band in college and our tube Bogen PA amp sounded terrible, you could hardly tell the words being sung. I showed the band that these amps could not only replace our PA, but we could simultaneously run the rest of the band through them too. I made some plywood ported cabinets for four Electrovoice SRO-15 (15 inch) speakers which sounded great using the Mattes amps. One problem we did run into was that if we played in a larger venue, the amps would get hot and the protection circult would cut off the sound for a second or two. So I re-engineered the circuit to use silicon instead of germanium transistors (the weak link) and added a fan. This solved the problem. I took an electronics course at the university just to get access to equipment which would let me make this fix. The only drawback to the fix was that it introduced a pop at power up which I didn't get around to eliminating until recently (see notes below.) Before the fix, we would occasionally blow out transistors. The germanium 2N2148 would go and take out the 2N3053 along with it. If the Westinhouse 156-10401 output transistors blew, that was real trouble. Those were specially selected 2N3055's to handle the higher voltage, as I understood it with a custom part number for Mattes use. We tried 2N3773's but they didn't last long. Same with MJ15015's of that era. They were probably too slow in turning off. Around that time the 2N3055's changed technology from hometaxial to epitaxial which were supposedly less immune from secondary breakdown, but I am not sure if that is the real problem. Anyway, the newer MJ15015G ON Semiconductor transistors with fT of 6Mhz, high hFE and higher voltage actually seem to work well.
The circuit was quite interesting because it had about a 12 watt intermediate amplifier (using the 2N2148 + 2N3053). The two output transistors would not even turn on until you got higher than the supply voltage for the intermediate amp. Once the output transistors turned on, a pair of diodes would disconnect the intermediate amp from the speakers and only drive the bases of the output transistors via a transformer. Feedback fixed any non-linearity and harmonic distortion at the high end was essentially inaudible anyway. The circuit as a result was quite efficient, only the class-A intermediate amp drew power all the time. In fact, the switch on the back adjusting for 8 or 4 ohm speakers only changed the power supply voltage for the output transistors. There was a test position which basically never let the output transistors do anything. If you want the full power rating of the amp, you need to supply 125 AC volts to the power cord. Check your amp because some were configured for 117 volts. So if you are testing the amp, keep that in mind. It is sort of a class A amplifier connected to a class C amplifier. There was an interesting article about the "Madan Sharma" circuit in Electronics World here.
Their preamp was very nice too. The integrated amp worked well too. However, over many years the switches started oxidizing and getting noisy to the point were they would need to be replaced. The circuitry was so crammed in there that it was a bit much to rework, not to mention trying to find the special custom rotary switches. So I parted with those some years ago. But I still have three, 50 year old, working SSP-200's with their original capacitors.
Here are some of the original brochures and documents: