HiFi Subjectivist ¦ Audio Forum ¦ NVA User Group

Ithilstone wrote: ↑Sat Jan 11, 2020 8:09 pm

SteveTheShadow wrote: ↑Sat Jan 11, 2020 8:05 pm Cheers Stu,
I’ll do it the Doc way.
What do you mean by wire resistance?

I would as well split it that way,
I think Stu ment to not connect 0 from both in the same place but rather on both sides of caps bank

Can you explin your last sentence Tomasz ? I don't quite understand what you are saying. Thanks.

karatestu wrote: ↑Sat Jan 11, 2020 8:15 pm I tried joining the two psu's 0V with a very short wire between the two different sets of capacitors and had a hum problem. When i joined them with a much longer wire it went away even though we are talking a fraction of an ohm difference. The extra resistance is the only explanation i can come up with

As Les at Avondale Audio once told me - it's best to have an interface

Ah..yes, I did something similar with the three power supplies on the OTL. With bus bar earthing it was relatively easy. Power stage cap bank was grounded nearest the mains earth end of the bus bar, followed by the driver stage bank further away, finally the input stage bank still further. Hum levels were virtually zero, even into 100dB efficient speakers. The OTL remains the quietest valve amp I’ve built to date.

I think the Doc's 0V arrangement can be viewed as a bus bar. Not a star earth in sight. And with no mains 0V connection (class II) there is no mains 0V end.

There is what i call the dirty end (well two of them in a dual supply) on the rectifier side of the cap bank where i have always connected the speaker 0V return and the "clean" side on the amp board side of the cap bank.

Les at Avondale always said to join the two psu 0V in his voyager amps (which i have built) on the front end regulator board, not straight between one cap bank and another. Make of that what you will

On a mobile so cannot do any drawings and that would be easier
Anyways i would run one wire (0v) across all 4 caps and connected 0v from transformers on each end
and run 0v for boards from middle between both pairs of caps - hope that make sense

Ithilstone wrote: ↑Sat Jan 11, 2020 9:07 pm On a mobile so cannot do any drawings and that would be easier
Anyways i would run one wire (0v) across all 4 caps and connected 0v from transformers on each end
and run 0v for boards from middle between both pairs of caps - hope that make sense

Got it!

The wiring in my nva based amps is a real mess with not much thought gone it to placement. When i redo them i am going to play around with different schemes to see if there are any effects (there probably will be.)

I won't be making wiring looms, running wires together or twisting but i thought i might try and get all the wiring at a level from the base which is half the depth of the transformer. It is said to be the best place to put it as regards stray field pick up. Have you got a view on that Steve ?

May be missing the point but as they are class II why connect the 0v from the two channels together at all?

Hi Nick. We are talking about two psu's on the same channel (front end and output stage) not two separate channels.

karatestu wrote: ↑Sun Jan 12, 2020 10:56 am The wiring in my nva based amps is a real mess with not much thought gone it to placement. When i redo them i am going to play around with different schemes to see if there are any effects (there probably will be.)

I won't be making wiring looms, running wires together or twisting but i thought I might try and get all the wiring at a level from the base which is half the depth of the transformer. It is said to be the best place to put it as regards stray field pick up. Have you got a view on that Steve ?

If in a metal case, then AC wires should be twisted together and pushed into the bottom corners of the chassis in order to minimise the amount of radiation from them. You can then position signal wires up in the air and make sure that if they do cross above an AC feed, they do so at right angles to the offending wires.

If it’s still going to be on an open breadboard I would still twist individual AC wires with their mate, to make up individual pairs and cross signal wires at right angles to them, so that interference is minimised. The signal wires need to be up in the air and obviously not tied into looms. Looms might create unwanted stray capacitances, which as Richard was always at pains to point out, NVA circuits do not like.

Unlike valve amps, which often have AC heater wiring near to low level signal wires, with solid state, there should not really be much of a reason to have AC wire, transformers and signal wiring in close proximity, so in that sense, a solid state amp should, in theory, be a bit easier to keep quiet, if good wiring practice is followed.

karatestu wrote: ↑Sun Jan 12, 2020 12:10 pm Hi Nick. We are talking about two psu's on the same channel (front end and output stage) not two separate channels.

Ok. Point taken.