Are transistors, BJTs and Mosfets affected by the magnetic fields from step down transformers used in linear power supplies? I appreciate that there are such things as hall effect devices which deploy such characteristics but do conventional small signal devices suffer such affects to any meaningful degree?
There are reported effects on the characteristics of a junction field effect transistor (JFET) in applied magnetic fields up to 30 T.
https://pubs.aip.org/aip/rsi/articl...ic-fields-on-junction?redirectedFrom=fulltext
A modern well-designed 60 Hz power transformer will probably have a magnetic flux density between 1 and 2 T inside the core.
However the external magnetic field will be very small by comparison.
I therefore conclude that there are no meaningful effects in your scenario.
https://pubs.aip.org/aip/rsi/articl...ic-fields-on-junction?redirectedFrom=fulltext
A modern well-designed 60 Hz power transformer will probably have a magnetic flux density between 1 and 2 T inside the core.
However the external magnetic field will be very small by comparison.
I therefore conclude that there are no meaningful effects in your scenario.
The general effect is to slightly increase the Vbe, which in theory could modulate the signal, but in practice this effect will be dwarfed by the emf induced in the tracks and wires leading to the device, and this can be very sizeable, which is why it is a good idea to keep sensitive circuits like preamps far away from any transformer
Thank you for your replies. Yes it was my understanding that the most likely cause of interference would be by coupling to longer lengths of conductors PCB tracks or cables. From my work on power distribution transformers the design of transformers is for high efficiency and thus the magnetic circuit is very efficient which has the benefit of minimizing EMF 'leakage', indeed parallel runs of single core cables carrying high current are much more of a challenge in this respect.
But I digress! in terms of HiFi, I would expect the control of 50 /60Hz noise on your average line level circuity to be entirely manageable, whether by distancing or by use of screening plates.
I have long since wondered whether locating the transformer in a separate enclosure connected by the inevitable longer cable lengths is of doubtful benefit, let alone necessary and possibly disadvantageous due to the possibility of coupling through the interconnecting cables, additional impedance etc?
i.e. Is interference from mains done and dusted, I don't recall ever having an issue, although I don't design phono stage amps ?
But I digress! in terms of HiFi, I would expect the control of 50 /60Hz noise on your average line level circuity to be entirely manageable, whether by distancing or by use of screening plates.
I have long since wondered whether locating the transformer in a separate enclosure connected by the inevitable longer cable lengths is of doubtful benefit, let alone necessary and possibly disadvantageous due to the possibility of coupling through the interconnecting cables, additional impedance etc?
i.e. Is interference from mains done and dusted, I don't recall ever having an issue, although I don't design phono stage amps ?
Last edited:
Because the fields diminish very quickly by distance, reaching back into the darker, less used recesses of my mind, is this a case of inverse square law, by chance?
I heard an audible improvement wrapping the transformer on my tube headphone amp with low frequency high permeability mu metal. And further improvement when I wrapped a second layer of mu around it.
An umbilical is the golden standard regarding induced noise, etc. and it poses no particular problem if you locate the regulator inside your audio device and the rectifier/filter in the remote supply box.
The main inconvenience is the practicality of having to deal with two boxes and their interconnecting cable
Something I noticed from my junkyard diving. (Buying high end audio scrap). Sold by the KG. I then pull the heat sinks and transformers and in some case even components off these. I also study what transistors they use. None of them use BC547 for sure lol.
For some reason all the amps I build sounded way way better when I used a junk yard transformer. In fact my EI core junk yard transformers were working better than my Toros. They just sounded better volt for volt amp for amp.
Or locally made transformers.
This is when I decided to take a closer look at these. What I noticed is that they had very good shielding.
In fact there are at least 3 layers of shielding. Two metal based and one fiber.
I discussed this with my audio guru.
He said it will be hard to copy the materials they have used. But at the very least you can copy their idea of using copper shielding. i.e. all of them were wrapped in copper.
So I have placed an order for adhesive backed copper from amazon. This is what he uses.
I would rather measure the thickness of the copper sheet these guys have used and buy something similar.
Its not as thin as the tapes. If I had to guess its about .2mm thick. Tape is way less.
I also need to figure out how to wrap my toroidal transformers. Sadly I have no junk yard example of a toroidal. But I'm sure google will find me something.
Ps: Are transistors, BJTs and Mosfets affected by the magnetic fields from step down transformers used in linear power supplies?
For some reason all the amps I build sounded way way better when I used a junk yard transformer. In fact my EI core junk yard transformers were working better than my Toros. They just sounded better volt for volt amp for amp.
Or locally made transformers.
This is when I decided to take a closer look at these. What I noticed is that they had very good shielding.
In fact there are at least 3 layers of shielding. Two metal based and one fiber.
I discussed this with my audio guru.
He said it will be hard to copy the materials they have used. But at the very least you can copy their idea of using copper shielding. i.e. all of them were wrapped in copper.
So I have placed an order for adhesive backed copper from amazon. This is what he uses.
I would rather measure the thickness of the copper sheet these guys have used and buy something similar.
Its not as thin as the tapes. If I had to guess its about .2mm thick. Tape is way less.
I also need to figure out how to wrap my toroidal transformers. Sadly I have no junk yard example of a toroidal. But I'm sure google will find me something.
Ps: Are transistors, BJTs and Mosfets affected by the magnetic fields from step down transformers used in linear power supplies?
For magnetic field shielding what you want is low frequency high permeability mu metal, available on eBay, at least it used to be. Mu metal is a special thin easily bendable metal alloy. Wrap transformer once with mu metal for very good shielding, for even better shielding use a 1/4” spacer like wool or cotton, then wrap a second layer of mu metal.
Last edited:
easily bendable, but the bending reduces it's properties considerably which may require re-annealing.
Mu metal is intended to be bent, that’s why it’s thin and flexible, malleable!. Its properties will be fine.
Last edited:
manufacturers don't think so https://mumetal.co.uk/?p=100 even tapping a can down onto a circuit can stop it working properly.
Would sticking a transformer and amp in the same steel box make matters worse?
I've noticed that I like the sound of alloy encased amps better than steel, and plastic best of all, however this could very easily be a coincidence. I was thinking of using two alloy enclosures, one for two transformers, and another for a DAC and amp, the two enclosures bolted together, one on top of the other. Would the magnetic field from a transformer have an effect on capacitors as well?
I've noticed that I like the sound of alloy encased amps better than steel, and plastic best of all, however this could very easily be a coincidence. I was thinking of using two alloy enclosures, one for two transformers, and another for a DAC and amp, the two enclosures bolted together, one on top of the other. Would the magnetic field from a transformer have an effect on capacitors as well?
Nope, bending it greatly reduces the µ value by work-hardening - once bent into shape you have to re-anneal... Its soft because its annealed with extremely low density of dislocations, necessary for getting large µ (relative permeabilities of 100000 are achievable IIRC).
One layer of mu metal reduces B by 75% as I recall and two layers of mu metal reduces B about 90% as I recall. So in my opinion it’s worth doing. I did it and it was worth it. I’ve also used mu metal around video cables. That’s good, too. Remember The Little Engine that Could.