This should hopefully be a working amplifier and not an oscillator. And yes I know its horrible so looking for feedback. >🙂
A few thoughts, just personal opinion, (you asked for it ) 😉
By the way, could you post a picture of the schematic so I don't have to conjure connectivity from the layout?
I assume Q14/Q15 are the output drivers, and C9/C10 are the bypass caps for the 45V rails? Is J4 the output? or J3?
Usually a star ground is used, to prevent the high current drivers from inducing an AC IR drop that can bounce the pre-amp ground. Power and pre should each have an independent path to the AC decoupling ground. You are using a ground plain, which is pretty good, but you still have the predrivers between the power input and the power consumers forming a ground trampoline. Might consider putting the power connector over closer to the output devices and bypass caps so the return current does not need to snake through the predriver area. The way the +/- 45V are routed you are forming a big inductive loop with your sensitive circuitry right in the middle of it. More after I can study the schematic. Great start though! -W
By the way, could you post a picture of the schematic so I don't have to conjure connectivity from the layout?
I assume Q14/Q15 are the output drivers, and C9/C10 are the bypass caps for the 45V rails? Is J4 the output? or J3?
Usually a star ground is used, to prevent the high current drivers from inducing an AC IR drop that can bounce the pre-amp ground. Power and pre should each have an independent path to the AC decoupling ground. You are using a ground plain, which is pretty good, but you still have the predrivers between the power input and the power consumers forming a ground trampoline. Might consider putting the power connector over closer to the output devices and bypass caps so the return current does not need to snake through the predriver area. The way the +/- 45V are routed you are forming a big inductive loop with your sensitive circuitry right in the middle of it. More after I can study the schematic. Great start though! -W
I would keep the power and loudspeaker connections at the high current end of the pcb so you don't wrap those large high current loops around the input stage.
Hey guys, thanks for the feedback, I have made numerous ammendements to the schematic before I have another crack at making a PCB. Seeing as I plan to sandwich 7 of these amps I had an idea of a backplane for connecting all the PCBs to a sort of IO board and have UPC1237's watching over these amps, hence the overload detection circuit to the right of the schematic.
Its based off the John Audio Tech JAT501, but adjusted for slightly higher voltage operation and compacted a little bit using some nice sanken darlington BJTs.
One thing I am looking at doing is incorperating a VI limiter alongside the overload detection circuit for added redundancy in terms of short circuits and to help reduce the likelyhood of the output BJTs going "What's this? Overcurrent? I melt now good luck speakers.". I also plan on having a system in place where if one channel faults the power supply is tripped.