BNOS 30V 30A PSU questions - Page 2 - UK Vintage Radio Repair and Restoration Discussion Forum

vintage port · 6th Nov 2025, 9:07 pm

The driver is a TIP120 so good for a couple of amps, given the mediocre heatsink it's attached to. That should supply enough, though the incumbent blew when soak testing. New one in, all running via a single 2N3055H, a bit more output ripple but the device is still disassembled so may improve when reassembled?
The floating negative output has some 136V AC RMS on it and the chassis is a bit tingly?? The thing works fine when the output is earthed, and the chassis is of course properly earthed, so do I assume a random leakage or go hunting?
Thanks to all participating!

Silicon · 7th Nov 2025, 11:46 am

That last paragraph gives me the impression that the 'chassis is a bit tingly' even though the chassis is properly earthed.

A mains transformer built without a Earthed screen between the primary and secondary windings can give AC on the low voltage DC side due to the leakage through the capacitance in the transformer between the primary and secondary windings.

It would be beneficial to have a DC power supply that could have the + and - isolated from the mains Protective Earth but also having an case/enclosure that is safe.

Could you please tell me if I have misunderstood the situation?

vintage port · 7th Nov 2025, 7:04 pm

Quote:

Originally Posted by Silicon

That last paragraph gives me the impression that the 'chassis is a bit tingly' even though the chassis is properly earthed.

A mains transformer built without a Earthed screen between the primary and secondary windings can give AC on the low voltage DC side due to the leakage through the capacitance in the transformer between the primary and secondary windings.

It would be beneficial to have a DC power supply that could have the + and - isolated from the mains Protective Earth but also having an case/enclosure that is safe.

Could you please tell me if I have misunderstood the situation?

Hi Silicon,
thanks for the reply. You understood as well as you could what I had garbled, so time to make it more comprehensible!
The chassis is well earthed (metal) and shows 0.00V between it and mains earth. The tingle appears between the PSU output and the chassis, the 136V AC is between the PSU output and the chassis. If I earth the PSU output to chassis it causes no exciting events and makes no discernable difference to the operation of the PSU, so it would appear to be a floating potential. Nevertheless, 135V seems quite a voltage offset in a 0-30V supply. The multimeter reckons its 50Hz and shorting the PSU output to the chassis through the ammeter shows 0.16mA. I hope this helps?

vintage port · 7th Nov 2025, 7:22 pm

I have measured the PSU output to chassis impedance and it's OL on both positive and negative terminals. The chassis has two beefy 500VA torroids mounted on it with 30V secondaries. The output wiring passes alongside these trafos, it looks like it always did. For reference I have only mounted one transistor on one side and where the the other bank of four normally go the wiring is floating. Is it possible that this sort of current can be induced by this proximity?

Silicon · 8th Nov 2025, 12:21 am

The leakage impedance seems to be very roughly 1 MegOhm.

A capacitance of 3nF would be needed to give that impedance at 50Hz.

I doubt if this comes from the wiring.

I would try connecting a 0.1uF capacitor from the secondary windings to chassis to see if it would reduce the AC voltage on the DC terminals.

vintage port · 9th Nov 2025, 4:39 pm

Hi Silicon,

I only just got a chance to have a go at your suggestion.
When I tried with a 150nF polyprop the voltage dropped to around 3.4V RMS, so with a 1uF polyprop its around 0.5V. On the scope the output floats on a fairly clean sine at 50Hz. I've tried unbolting and lifting the trafos a little to decouple them from the adjacent material but there's not a lot of room to manoeuvre and it make little if any difference. The floating voltage appears after the two diode bridges.

I assume it's OK to leave the cap in, seeing as how the PSU has no current regulation anyway and, if so, can I ignore the floating 0.5V RMS after using a 1uF cap as being acceptable, or should I dig in a little deeper to see what is causing this and try to fix it? One option is to rebuild with my eight new spangly RCA 2N3055H from Langrex. There might be some minor leakage/coupling from those circuits to ground. Please, can anyone offer a clever place to start?

Thanks to all once again!

vintage port · 10th Nov 2025, 7:12 pm

Hi all,

Rebuilt, the leakage it is reduced to 70V RMS, still not exactly great. Curiously the 1uF capacitor reduces this to the same 0.5V so maybe this is as good as this machine gets?

The ripple on the regulated supply is a bit lower, about 1.4mV RMS, and the new transistors are keeping much more equal temperatures so, all in all, successful.

6th Nov 2025, 9:07 pm	#21
vintage port Tetrode Join Date: Dec 2010 Location: West London, UK. Posts: 58	Re: BNOS 30V 30A PSU questions The driver is a TIP120 so good for a couple of amps, given the mediocre heatsink it's attached to. That should supply enough, though the incumbent blew when soak testing. New one in, all running via a single 2N3055H, a bit more output ripple but the device is still disassembled so may improve when reassembled? The floating negative output has some 136V AC RMS on it and the chassis is a bit tingly?? The thing works fine when the output is earthed, and the chassis is of course properly earthed, so do I assume a random leakage or go hunting? Thanks to all participating!

7th Nov 2025, 11:46 am	#22
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,567	Re: BNOS 30V 30A PSU questions That last paragraph gives me the impression that the 'chassis is a bit tingly' even though the chassis is properly earthed. A mains transformer built without a Earthed screen between the primary and secondary windings can give AC on the low voltage DC side due to the leakage through the capacitance in the transformer between the primary and secondary windings. It would be beneficial to have a DC power supply that could have the + and - isolated from the mains Protective Earth but also having an case/enclosure that is safe. Could you please tell me if I have misunderstood the situation?

7th Nov 2025, 7:22 pm	#24
vintage port Tetrode Join Date: Dec 2010 Location: West London, UK. Posts: 58	Re: BNOS 30V 30A PSU questions I have measured the PSU output to chassis impedance and it's OL on both positive and negative terminals. The chassis has two beefy 500VA torroids mounted on it with 30V secondaries. The output wiring passes alongside these trafos, it looks like it always did. For reference I have only mounted one transistor on one side and where the the other bank of four normally go the wiring is floating. Is it possible that this sort of current can be induced by this proximity?

8th Nov 2025, 12:21 am	#25
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,567	Re: BNOS 30V 30A PSU questions The leakage impedance seems to be very roughly 1 MegOhm. A capacitance of 3nF would be needed to give that impedance at 50Hz. I doubt if this comes from the wiring. I would try connecting a 0.1uF capacitor from the secondary windings to chassis to see if it would reduce the AC voltage on the DC terminals.

9th Nov 2025, 4:39 pm	#26
vintage port Tetrode Join Date: Dec 2010 Location: West London, UK. Posts: 58	Re: BNOS 30V 30A PSU questions Hi Silicon, I only just got a chance to have a go at your suggestion. When I tried with a 150nF polyprop the voltage dropped to around 3.4V RMS, so with a 1uF polyprop its around 0.5V. On the scope the output floats on a fairly clean sine at 50Hz. I've tried unbolting and lifting the trafos a little to decouple them from the adjacent material but there's not a lot of room to manoeuvre and it make little if any difference. The floating voltage appears after the two diode bridges. I assume it's OK to leave the cap in, seeing as how the PSU has no current regulation anyway and, if so, can I ignore the floating 0.5V RMS after using a 1uF cap as being acceptable, or should I dig in a little deeper to see what is causing this and try to fix it? One option is to rebuild with my eight new spangly RCA 2N3055H from Langrex. There might be some minor leakage/coupling from those circuits to ground. Please, can anyone offer a clever place to start? Thanks to all once again!

10th Nov 2025, 7:12 pm	#27
vintage port Tetrode Join Date: Dec 2010 Location: West London, UK. Posts: 58	Re: BNOS 30V 30A PSU questions Hi all, Rebuilt, the leakage it is reduced to 70V RMS, still not exactly great. Curiously the 1uF capacitor reduces this to the same 0.5V so maybe this is as good as this machine gets? The ripple on the regulated supply is a bit lower, about 1.4mV RMS, and the new transistors are keeping much more equal temperatures so, all in all, successful.