Thanks @carron for your external analysis !carron wrote: ↑Wed Aug 14, 2019 10:09 amGreat detective work!
Yes, aluminum is not the best, and mumetal is expensive. Perhaps you can try to cover some of the aluminum enclosures in copper tape to see if that helps. Though two types of metal together are a recipe for corrotion.
In principle a toroidal transformer should minimize the leakage of magnetic flux, so I would have thought it should have minimizde hum relative to a classic transformer design.
It is true that the radiation has a "more" omnidirectional structure than a classic transformer, where it is even more concentrated in one direction. Nevertheless the radiation of a toroidal transformer is not perfectly omnidirectional. It should be larger along the axis perpendicular to the toroid "circle", because the inner side of the coil and farther side are at different distances to the center and the effect does not perfectly cancel out, and the effect sums from every direction along the center symmetry axis. Perhaps there is something just directly below that transformer, along that axis that is acting as an antenna and picking that noise. Have you tried turning the Toroid 90 degrees on its side, to see if anything changes?
Reducing hum is a difficult art for any expert designing circuits, and I am no expert by any stretch of the imagination.
Again, amazing build.
Yes, like you, I was prone to think that toroidal transformer would be better in terms of magnetic hum field, combined to a welcome known significant gain in weight... It's not the case here, at least...
Consecutively to your fine thoughts , I now consider the central screw of the toroidal transformer : wouldn't it be a "good magnetic hum field spreading accessory" ? It's made of steel - hence magnetic, so replacing it with a brass one is maybe a test to do.
And yes, good idea : I will try to turn the toroid transformer at 90° also.
Thanks again for your suggestions !