CR308 looks like it says 1N473A. Info I found says it's 4.7V +/- 5% 200mW. Google keeps coming up with 1N4731 when you search for 1N473A, but that's not the correct part.

CR308 normally shouldn't  conduct. When it does it sets the current limit on the plus side (which flows through R322) by clamping the base of Q305 to the output (my math says @ 35A) . It also sets the current limit on the minus side when it conducts in the other direction through CR306 and Q304 (my math says @ 20A). This is where CR310 comes into play because it sets a current limit on Q303 so it doesn't pull too hard when a minus current limit is reached.

Back to the other issue. Look at the feedback network consisting of R314, R313 and R315, and R312. Do the math on this Voltage divider. At DC, C306 should not affect anything. If the output is stuck at 12V, the base of Q302 should be around 3.3V (without doing the math I guessed 4V). If R312 is not doing it's job, you might see 10V at the base of Q302.

Experiment: If Q303 is off (base at 0V), remove it from the circuit and substitute a 6.8K resistor from emitter to collector. That should allow the output to float up to 45V. That would prove that there is not an issue with the output stage, Q305 to Q314.

If that experiment doesn't work, remove all the power transistors (mark them so you can put them back in the same place) and try again. Q305 and Q310 can be bad, but may partially work at less than full Voltage.

Q304 doesn't do much except bias the output stage. It's voltage drop should be about 2V all the time. First look at the input stage, Q301 and Q302. Is the base of Q302 near +11V ? If it's low like it should be (~4V) with 12V on the output, then Q301 should be off and Q303 should be off (base = 0V) allowing the output to go high. What might be pulling the output down ? The base of Q310 pulling current through R340. 1V drop across R340 would pull the output down to 12V.

No, you might see what the feedback is doing to the signal in an attempt to make the output look like the input.

Since the feedback seems to be working at DC:
11V on the input of Q301, Q302
45V on the plus side of C404

The amp should be able to make a clean sine wave on the output with NO LOAD. Get it working with no load first, then try it with dummy load. Try 8 first, if good, 4 and 2.

Check Voltage across R341, should be about 120mV by my calculation. Check CR310 with DVM diode check out of circuit. It should look like whatever your meter reads with two diodes in series. These components sort of set the current limit on the minus side.

Check CR308(?), CR306 and CR307 with DVM. I'm coming up blank on 1N473A, I'm assuming it's a low Voltage zener, they can be pretty flakey. It sets the current limit on the plus side.

Check the Voltage drop across R339 and R340. They should have similar Voltage drop. Check resistance of all the big low value resistors as well as you can with your DVM. Usually these will be open when they fail.

Good Luck !

I think you have it right. The only problem I see is the diodes from the output to the rails should be 1N400x.

Positive Gate Voltage turns Q315 off. This comes in through R330. Q316 is normally off, but it has a slightly positive base Voltage, so it is ready to conduct. When signal comes in via R335 or R337, Q316 starts to conduct lowering the Gate Voltage on Q315 which turns it ON lowering the Source to Drain resistance. C315 gives the whole thing a fast attack - slow release action.

R331, C314 unknown function, probably for fast attacks.

Note: When replacing semiconductors in an amp this old, double check the Base Emitter & Collector connections. Sometimes the original parts had non-standard pin-outs. Same with the JFET.

Normally the diodes CR301-304 don't conduct. If they do, the input signal is too big. When the signal at the output of the power amp gets high enough, the JFET turns on shunting the input which in turn reduces the output.

A too big signal on the input makes me think the gain of the power amp is too low somehow.

The gain of the power amp is set by R314, R313, R315 and C306. R312 helps set the DC gain (operating point) of the output. The pot R304 is used to adjust the DC (45V) on the output. The low frequency corner is about 30Hz, set by R313, R315 and C306.

Which side of R301 is the waveform from ? One side is essentially the same as the Line outputs, the other might be limited by the JFET compressor circuit.

I think it was mentioned in this thread that many of the connections on the PCB were re-heated and/or re-soldered. Was any flux from re-soldering removed ? I would use a chemical intended for flux removal or isopropyl alcohol, a tooth brush and paper towels. Pay attention to the areas around channel switch IC's.

Can the same be said about some of the more recent Marshalls ? A couple of threads about JCM2000's over at MEF right now. Seems like the quality of transformers used by Marshall has gone down. Has just the reliability changed, or has the sound changed also.

Currently, no-can-do... But I might be able to upgrade my backup phone. Fighting a winter storm right now,maybe when things thaw out next week.

I wish I did, but I'm just not set  up to do it. I could buy some equipment, but I'm not sure what the best way to go is. Got a couple of mics and a few PA mixers. Don't know anything about the digital end of things.

You need to go on a bender:

https://www.ssguitar.com/index.php?topic=3732.0

Google is telling me the schematic file you linked does not exist.

Found this thread: https://www.ssguitar.com/index.php?topic=3605.msg27527#msg27527

Some amps use a scheme like this to detect when something is plugged in to mute the channel when nothing is plugged in. On this amp, it looks like they are providing power to some kind of active guitar or perhaps an accordion.  A normal phone plug will short it to ground, but it will send a jolt to your guitar if you plug in while the amp is on. Probably a good idea to disconnect it.

I'm not sure how much the Frontman 65 has in common with the Frontman 212R, but the 212R has problems with solder connections fracturing on the PCB. A close inspection under bright light and magnification might reveal the problem.