The problem of shielding 50/60Hz magnetic fields is very old an well known in the field of oscilloscopes with cathode ray tubes. Long time ago I attempted to build my own CRT based scope and found out that the stray field of the mains transformer in the vicinity of the crt widened the focus. I built a steel housing for the transformer, but this showed little or no effect. Magnetic conductance of steel is far too low to shorten the 50Hz stray field. This can only be achieved with Mu-metall. And so I learned that CRTs in professionnal oscilloscopes were screened by a mu-metal housing.
To make a long story short: There is little or no magnetic field reduction to be expected by steel transformer housings, but they are nice anyway.
To make a long story short: There is little or no magnetic field reduction to be expected by steel transformer housings, but they are nice anyway.
My 2 one box attempts with a Phono preamp have featured a toroidal transformer, steel transformer can (the one I linked to above from Antek) and steel dividing wall and hum was negligible with transformer about 12” from the input. Both featured the same supply, general layout and grounding configuration.
However due to lack of extensive experimentation with different circuits or in-situ adjustment and subsequent measurement protocol it’s unclear to me what are the largest contributing factors.
Seems to me mumetal would be best suited to a DIYer pasted upon a steel divider.
Attempting to form the mumetal much at all seems to compromise results, from what was said.
From my one experimentation with it, it’s obnoxious to work into any useful shape and best kept flat / straight.
Are you suggesting that derating a transformer for its duty will reduce flux emitted?
For example drawing 1A from a 6A rated toroid, let’s say?
I do, with the caveat that my knowledge in this area is not advanced.
Could you provide links, instructions or other general information that a weekend hobbyist might not be too intimidated by?
Don’t toroids emit EM mostly horizontally, not vertically?
Hence the practice seen mounting the toroids on a side wall?
What differences are there in dispersion patterns for the different core types?
Many have advocated R-cores as being one up on the toroid for EM, I believe.
Thanks for the links and info.
And yes, exactly.
Let’s for example take the cost of second chassis,
Umbilical cabling and connectors.... not to mention extra space occupied.
High quality umbilical connectors rated for higher voltages are not cheap. The amphenol ecomate is at the lower end of the spectrum and thats around $75 in M/F connectors alone.
Two chassis are usually two smaller chassis equal roughly to the size of a normal chassis (let’s say the Pessante in the diyaudio store) given that most want it to fit on the same shelf space. Hifi2000 options seem to indicate these buyer habits are true. Usually they are (in my observations) set side by side with a similar total transformer distance that could be achieved with a wider single chassis build. Many want to minimize the length of the umbilical to avoid hum pickup as the PSU is usually in the chassis with the transformer.
So the second chassis is essentially two steel dividers (each chassis wall) and some air.
To me this is quite similar to a regular sized chassis with two steel dividers or one steel divider and a can.
Perhaps in most instances this effort does not yield the best ROI.
Compare that with the cost of a can, a steel divider or two and a strip of mumetal..... or whatever your weapons of choice are.
But knowing this ahead of time is key to the savings in time, money and sanity.
Or having a way to ensure success in a one box configuration with about 12” in space (what has seemed roughly about the best possible distance from input to me in a 300mm deep diyaudio store 2U chassis.)
I wonder if thin steel plates and mumetal strips in alternating layers would be the way to go.
The mu-Metal is difficult to stick, drill, bend etc and so the steel would make up for a few of its limitations in a practical sense.
Also adhering to steel in alternating layers would ensure electrical contact of layers as well as minimal deforming of the mumetal while creating contact with chassis base.
Since the mu-Metal would I assume operate best grounded?
Would likely look tidier as well.
One could then measure without anything, then adding layers of each until no further improvement is possible?
I suppose this whole thread is leaning towards the grand question of:
——
Considering Mu-Metals apparent vast superiority, what is the best practical (most uniform results, easiest chassis attachment, least deformation, clean appearance) method of utilizing Mu-metal layers to confidently eliminate transformer in a diyaudio store chassis of standard width and a transformer with unknown EM behavior.
——
Last edited:
I mounted a transformer on a “vibrapod” once when I was in a similar situation.
They are turntable feet... were for a project that wound up being too heavy for them.
It worked extremely well for me, fit like a glove with no alteration required. They have a hole through the center that fit my mounting hardware.
But yes sorbothane does wonders. My washer and dryer are silent as a church mouse now.
I believe that a sorbothane pad for example would be technically called a mechanical snubber.
Also I think a mechanical issue can stem from a faulty diode as well for example, not necessarily dc on mains if all your transformers are in the same box and only one is humming.
Likely it’s a problem with the transformer itself, it’s mounting (maybe just a little loose?), or an issue with just the circuit of that transformer.
You won’t know until you pull the thing up off the ground and get it open, which can be half the battle.
I used 4 pole speakon and van damme 4 core 4 sq mm speaker cable, works well, not an issue for me since I don't use the same connectors for speakers
Thanks for the vibrapod suggestion.
I used sorbothane under my U frame woofers, the only thing that stopped the cabs from twisting at low frequency high excursion, fantastic stuff.
Can you explain the annealing issues?
The more the mumetal is deformed the less effective it is?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.