Skip to main content

If anyone here can answer it, it will be Boswell.
I read over on GS that transformers (notably Studer) may need to be demagnetized. If this is true, I would expect all transformers would be suspect over time. Is this true?


paulears Fri, 01/02/2015 - 03:08

Sowter transformers in the UK have a carefully worded comment on magnetism


Mumetal has a very high permeability so it magnetises very easily but it also has a very low reminance which means when the field is removed the "permanent" magnetism is effectively zero (very small compared with the saturation flux density). The core cannot be damaged or permanetly magnetised by applying a dc current.

We have a number of highly respected customers who have observed that the quality of sound associated with Mumetal transformers improves after a period of "burn in". Some customers even have particular pieces of music they use. We do not know of any scientific evidence for this but it is thought that this could be due to a reminance effect or perhaps "conditioning" of the listeners ears or brain.

Boswell Fri, 01/02/2015 - 03:29

If you are talking about microphone and line transformers, then, yes, they can become magnetised, in the sense that they may develop a small (non-zero) magnetization vector in the core. However, unlike a tape head, a magnetization of a transformer core is a small effect and usually not noticable in the audio that passes through it, particularly if the flux densities are such that the core is never run into saturation. If you are deliberately running the core into saturation for an effect, then you may find that the saturation effect varies slightly with magnetization of the core, even if no difference can be heard in the linear range.

A tape head is different from a transformer in that the tape forms part of the magnetic path. Magnetization in the head core can cause permanent change to the magnetic information on the tape, and hence unrecoverable audio distortion. In professional studios, it is common practice to demagnetize tape heads at least every day, and often before loading a reel of tape each time.

For my own gear, or if I get other transformer-based pre-amps in for service, I do take the effort to check for magetization, usually by examining a B-H curve on a scope if I have unfettered access to the transformer windings. If I can't do it that way, I simply check that saturation levels appear the same in the positive and negative directions, and if there were significant imbalance, I would perform a de-magetization procedure. I have only had to do this once in the last 30 years; it involved applying a 1KHz sinewave that reduced in amplitude from saturation level to zero over 10 seconds or so with the transformer output loaded. The procedure balanced up the saturation levels in that particular transformer, but the normal audio was, as far as I could tell, unchanged.

I haven't seen the GS thread, so don't know for sure whether they was talking about tape heads (the Studer reference) or mic/line transformers. Knowing GS, they were probably all confused anyway.