QUAD 303 seems a bit poorly

I've rescued a garage find Quad 303 and recapped and rebuilt the PSU and amp boards to the Dada list, switched on with nothing connected it works fine, 67V present 33.5V present and able to set the bias which is at ~6mV on both channels.

Now this is where things go wrong, connecting a signal to it (someone's put phono sockets on the front panel, these are uninsulated) and watching the trace on the scope, it'll get to about 1.5 - 2 volts out then the heat sink will get very warm, set to about 4V out it starts to clip then R115 the 2.2ohm resistors on both amp boards start to overheat and smoke.

Any ideas welcome.

Thanks

Chris
 

Attachments

  • Screenshot_20231016_080641_Adobe Acrobat.jpg
    Screenshot_20231016_080641_Adobe Acrobat.jpg
    194.9 KB · Views: 252
Ok so further testing,

Connect a grounded signal gen using a grounded mains plug and the resistors go up in smoke, ok so looks like it creates a ground loop? Should this then smoke R115 or just hum?

Connect a battery powered signal gen and a grounded mains lead and all is good. Ok so onto the next.

Connect a quad 33 to it and power the 303 through the auxiliary sockets on the 33 and signal to the 33 using the mains powered signal generator and the resistors start to smoke again!

Connect the 303 to the auxiliary sockets on the 33 again so no mains grounded and Connect a mains grounded signal generator and poof of the resistors go again!

What am I missing or doing wrong?
 
The power sully of this amplifier is quite special positive side of filters cap ( C2 C3 ) is connected to the amplifier but négative side reach the amplifier via TR3 so negative side of C2 and C3 should de a few volts BELOW zero. I think this is the first thing to check.
 
It's maybe a silly question, but do you have anything under the boards when they're leaning against the chassis like that? Sorry if that's too basic, but it's good to ask the "is it plugged in, is it turned on" questions just to rule them out... 🙃
 
No nothing under the boards
I guess that was my question. If you don't have anything under the boards, it's likely they'll short out underneath when leaned up against that chassis like you have them. The standard practice is to put a piece of plastic or wood under the boards when they're flipped out like that to make sure they don't touch.
 
Here's mine, thinking that it can't be an individual amp board causing this, maybe a power supply or grounding problem?
 

Attachments

  • 20231019_173513.jpg
    20231019_173513.jpg
    369.3 KB · Views: 96
  • 20231019_173508.jpg
    20231019_173508.jpg
    440.3 KB · Views: 92
  • 20231019_173502.jpg
    20231019_173502.jpg
    447.5 KB · Views: 94