Do you know how much of that hum was due to the electric, rather than magnetic field?
Well that's not fair! I'm sure a pipe of 1.5mm-thick mu-metal would be even better your steel pipe! I was talking about an isolated piece (not a loop!) of flimsy 0.15mm material. A similar piece of steel would be practically invisible to the magnetic field.
So do I. At work (here at Neve) I spend a lot of time trying to shield mic input transformers from SMPUs. It's a nightmare. But one thing I have learned is that regular steel is largely useless. If you have to have the PSU and IP transformer inside the same chassis, then only mu-metal gives useful results. (However, our transformers are already shrouded before I get to this point; I'm not talking about naked transformers).
K&J seem to be mainly interested in high flux densities which may actually saturate mu-metal shields. I don't think ordinary transformer leakage flux is dense enough for this to be a problem. But I'll qualify: Steel is useless unless it's very thick. We rarely have the luxury of installing quarter inch steel bulkheads and shrouds. But mu-metal is effective even when it's very thin.
Last edited:
I have an older (15 yrs) Jolida 502, and I stopped an initial power-on "hum" that was generated by startup vibration from the power transformer coupled to the chassis. The hum faded as the unit warmed. I mounted the transformer bolts on rubber grommets. This corrected the problem.
I note this because the noise could be mechanical, as well as the many electronic alternatives discussed above.
I note this because the noise could be mechanical, as well as the many electronic alternatives discussed above.
Last edited:
There is definitely a mechanical component to it. There is a clearly audible hum from the chassis/transformer with the speakers disconnected. As I mentioned in an earlier post, with the transformer placed beside the chassis but not in contact with it, the hum virtually disappears. I tried moving the transformer towards the chassis and as it get closer you can hear the mechanical hum begin and you can even feel the magnetic attraction between the transformer and the chassis.
Today I took delivery of a toroid mains transformer which I expect will not suffer these problems.
Cheers
Ian
Here is an example of the copper "Gauss"/"Belly"/"Flux" band. It should be insulated from the transformer laminations as well, in order to not compromise the lamination insulation coatings. It has its ends soldered together. It acts as a shorted turn (for the leakage field only) to prevent flux radiating from the center back side of the E lams as well as from the exposed winding.
The end bell covers can prevent magnetic leakage flux radiating from the winding core itself as well as electrostatic coupling.
The toroid xfmr may eliminate flux leakage if it is wound with full 360 degree windings, otherwise flux leaks from the winding ends. The toroid arrangement is notorious however for transmitted noise from primary to secondary.
Shielded R core (image attached) can prevent the primary to secondary noise transmission if the windings are on separate legs. Otherwise they need good insulation between primary and secondary (like conventional E-I) if the windings are balanced on both legs. The R core needs the exterior copper shorting band to prevent magnetic leakage off the ends of the R core as well as from the windings, especially for the separated winding version.
Mu metal shielding is very sensitive to bending, so has to be re-annealed after forming into a shield box. You might be able to gently curve a straight OEM sheet around into a tube around the whole xfmr. Useful for diagnosing the problem at least.
A magnetic leakage flux probe (a high turn coil on a Ferrite rod) for the O'scope is quite useful in locating the offending radiation source points. You will have a better idea of what you are up against.
The end bell covers can prevent magnetic leakage flux radiating from the winding core itself as well as electrostatic coupling.
The toroid xfmr may eliminate flux leakage if it is wound with full 360 degree windings, otherwise flux leaks from the winding ends. The toroid arrangement is notorious however for transmitted noise from primary to secondary.
Shielded R core (image attached) can prevent the primary to secondary noise transmission if the windings are on separate legs. Otherwise they need good insulation between primary and secondary (like conventional E-I) if the windings are balanced on both legs. The R core needs the exterior copper shorting band to prevent magnetic leakage off the ends of the R core as well as from the windings, especially for the separated winding version.
Mu metal shielding is very sensitive to bending, so has to be re-annealed after forming into a shield box. You might be able to gently curve a straight OEM sheet around into a tube around the whole xfmr. Useful for diagnosing the problem at least.
A magnetic leakage flux probe (a high turn coil on a Ferrite rod) for the O'scope is quite useful in locating the offending radiation source points. You will have a better idea of what you are up against.
Attachments
Last edited:
Success!!
Yesterday the replacement toroid mains transformer arrived. I fitted it in last night and the hum has completely disappeared. The amp is so quiet that at first I thought there was something wrong with the HT but as soon as a plugged in a signal source I knew it was OK.
Many thanks to everyone who replied with suggestions, observations and theory. I have learnt a lot.
Cheers
Ian
Yesterday the replacement toroid mains transformer arrived. I fitted it in last night and the hum has completely disappeared. The amp is so quiet that at first I thought there was something wrong with the HT but as soon as a plugged in a signal source I knew it was OK.
Many thanks to everyone who replied with suggestions, observations and theory. I have learnt a lot.
Cheers
Ian
- Status
- Not open for further replies.