Just turn it down. At full class A bias the circuit is capable of 8.9 watts into 8 ohms. Not all transistors could take that so many were set lower. R ealistically it is hard to hear the difference between 8.9 watts and 5 watt s. Plus, there is some amount of DC on the speakers so if they are not stoc k they might not like that DC. I know I blew a pair like that a long time a go.
If you check the DC resistance of the output choke you can do a few calcula tions. It is probably 1.2 ohms and they call for one volt across it. For th e stock speakers that were actually ten ohms, the DC did not bother them, b ut if they have been replaced you want that bias lower or else put a blocki ng capacitor in series.
I would say that you don't want any more than a volt across that output cho ke. If you want more power get a power booster. Also, since it failed ad th ey never do, and the age, I would suspect the cap across that output choke. Those are alot higher value than we are used to these days, I don't rememb er specifically but they might be 10 uF or something like that. If that goe s open the output transistor is very vulnerable to inductive spikes. Being across the choke, very few caps checkers will check their ESP, bitching abo ut a short circuit. So you have to disconnect it. My scope based one will b ut those mean some building. You have to have a known impedance of a known voltage of a square wave and that requires prints. It might be useful later but not enough for just one project.
Anyway, using the output choke DC resistance as a guide, set it to somewhat less than one amp and see how it works. It might be just fine. I the speak ers have been replaced they might be four ohms or at least have a lower DC resistance in which case you have to set it even lower. If stock speakers w hich will be ten ohm, put a bit less than a volt on it and your output tran sistor will be dissipating about 12 watts all the time. Even the cheaper on es can handle that.
There were some of those cars that had faders and more than one speaker. Th ose had the bigger output transistors and the heftier chokes, and were of c ourse biased higher to get that couple of extra watts.
Check that cap, and then get say 750 mV across that coil and see how it sou nds. If it sounds good, ship it.