Yep each phono input just through the coax to the PA board. The mains to the 0V of the PSU board. The PSU board to both 0V of each PA. The 100k volume pot (if you have one) should have each cold end to the phono input earth only. That way not current from the mains earth or the power supply earth can get into the phono inputs. With the phone inputs shorted but not connected to an external pre-amp, the amp should be quiet.
You suggest away, my friend. All ideas are welcome. This is driving me nuts now. I could indeed try that. If it works there is a way to replace the connection to earth that lifts it enough to suspend the hum but reinstates it in the event of the chassis going live. It uses a bridge rectifier. Details on page 24 of this article.
http://hifisonix.com/wordpress/wp-content/uploads/2019/02/Ground-Loops.pdf
http://hifisonix.com/wordpress/wp-content/uploads/2019/02/Ground-Loops.pdf
No you can put a 10R 2W resistor between the mains earth/chassis ground and the 0V. Maybe time to try something else. So if you just short the input to ground directly on the PA board you still get hum with nothing connected on the input or its ground. I guess if you remove the EF86 then the hum goes (that will disable the LTP as well). If you have a oscilloscope a waveform across the speaker outputs would be good - failing that a .wav file. It may also be worth running the heaters of the EF86 and the LTP (if that's easier) off a bench supply with DC.
How odd maybe that does happen. Problem is the ECC83 grid goes to HT so its not a valid test. Do you have a bench supply to power the heaters off for the ECC83 and EF86 - I see they can be powered separately. I notice the original elector had 50Hz 60dB down on 1W which is audible - so it may not necessarily be a fault of the construction.
Last edited:
I am just wondering if its not a ground loop - but the heater supply getting into the EF86 input. Not sure how we can check. It does appear to be floating which is bad practice. It should have the same arrangement 2x100R with the centre to 0V as V3 and V4 supply. The only thing I can think of is to put a 10uF 200v cap from the plate of the EF86 to ground just as a check. This would remove all audio too, but if the EF86 stage is a culprit the hum will also go. I would disconnect the NFB if you try this.
Last edited:
Yes with the NFB disconnected but with no other mod you could check for hum on the plate of the EF86. It will still not tell you the exact cause - but you will have a lot more to go on. Actually even a modern meter on AC should be able to measure the 50Hz component. Often tracking down hum issues is often a process of elimination and it may not be a fault of your build.