torodial mains transformer hum

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

marcus4audio

Well-known member
Joined
Apr 23, 2013
Messages
79
Hi guys, I mostly repair equipment but I decided to DIY API mic pre and everything works as a charm but when I put the top of the box i have small hum boost. It's the radiation from torodial tx to output tx and it's not terrible but I would like to keep it clean as it's possible. I measure noise approx -85db at max gain but when I close the top of the enclosure (aluminum) few db's at 50hz goes up. What can I do about it? Any suggestion? I tried to simulate hole at the place of the toroid tx (with few smaller sheets forming the top) and everything works as a charm.  :-\
 
Thanks John. Yes, I found best position for both of the tx's and it's quite as possible (-85db) but when I put the top plate of the box the 50hz (150hz) hum goes up a bit. It looks like toroid is reflecting througn the aluminium sheet  :-\ Strange but true. When I move the top and leave just the rear part of the box open (the part where is the toroid) hum goes away.
 
I have the impression that the enclosure acts as a 'shorted winding', when you put the top plate on.
(In other words: magnetic induction in the enclosure.)
Is the transformer close to the top of the box?
 
RuudNL said:
I have the impression that the enclosure acts as a 'shorted winding', when you put the top plate on.
(In other words: magnetic induction in the enclosure.)
Is the transformer close to the top of the box?
Exactly but it doesn't touch the plate. Also, I made steel/copper/steel ring shield around both of the tx's but top is open.
 
When you close the box you alter the magnetic field inside.

The most difficult design I had to deal with was a low power install amp that had a 70-100V audio output transformer inside the same relatively small metal chassis as the mic input transformers.  No easy solution, but mu metal shielding in the mic transformer helped.

For a power transformer perhaps add an extra metal shield between the transformer and sensitive components to distort the internal magnetic field to provide some hum reduction.

In my old hifi kits a simple aluminum metal plate by the power transformer reduced internal hum levels in the audio path.

JR

 
Ok, here is my layout and what I did. I made a window at the top plate and at first it didn't help. But when I made the gap (see orange) at the back end hum went away. I also added the steel mesh to fill the space and it still works as the charm. I solved the problem but I still don't understand what aluminium plate does to the magnetic field inside aluminium box :/
 

Attachments

  • image.jpg
    image.jpg
    21.2 KB · Views: 29
kato said:
Also, shield your XLR input wires if you haven't already.
Of course, everything is done and all important signal cables are shielded from the beginning . No pin1 problem, just the toroid 50hz (150) reflection over the aluminium top to the output transformer. Now, when I figure out the problem and solved the same I just like to know more about the phenomenon. Some scientific explanation. ;)
 
When you place shields between power transformers and audio circuits, you will always see some part of the noise spectrum increase.  Hopefully to a lesser degree than the parts which are reduced, and in a less obtrusive frequency range.  Sometimes it's a trade-off when it's already a bad situation.  I have definitely seen 60/120 go down significantly and 180 go up slightly on several occasions.  Many times the increase will be above 1K or even 10K. 
 
Thnks for comment Doug. But I'm confused because aluminium is inferior as a shield at 50/100/150hz but radiation (or whatever) shows  just when I close the top of the box.
 
I guess this is on topic enough for the thread if anyone can take a crack at it -

Are there any obvious drawbacks to an internal toroidal transformer which is somewhat "less than round"? This Behringer thing I wanted to try some mods on happens to have one, and if it's likely to be significantly noisier, I may need to decide on some stuff.  :-\

Thanks

 
> toroidal transformer which is somewhat "less than round"?

Oval? Egg-shape? Round-shoulder? Crushed in a wreck?

The "roundness" isn't a key factor anyway.

What would "more than round" be?  :eek:
 
Torroids are notorious for having asymmetrical fields related to winding start/stop, overlap, etc.

In general the external fields are lower, but nothing in life is perfect.

JR
 
PRR said:
> toroidal transformer which is somewhat "less than round"?

Oval? Egg-shape? Round-shoulder? Crushed in a wreck?

The "roundness" isn't a key factor anyway.

What would "more than round" be?  :eek:
Hi PRR, thanks for chime in. Do you have some theory about my issue?
 
I think he may be questioning the "less than round" comment. Torroid cores are generally round and the transformers end up round, give or take winding thickness variations.

JR
 
Thanks guys!

It's sort of an oval, and is easily visible when you pop the top. I had read the stuff before about noise from the winding start/stop points, I just didn't know if having them wound on some randomly asymmetrical shape would create new problems. You sound like it's no biggie.

I have a feeling I may run into some of Marcus' issues with this box. Before I ripped it apart, I noticed substantial interference with a dynamic mic I tried on it. It'll be a while before I see what it does, but I may be in the same boat, so thanks for the info on mine and his.

George
 
Just in case anyone was wondering about the appearance.  ;)
 

Attachments

  • oblong.JPG
    oblong.JPG
    254.4 KB · Views: 25
The winding is round. It's lumpy where the wires come out.

Even if the windings were all-bunched-up, that's not a big deal. 99.9% of the flux is caught in the core, and there's no gaps or corners where flux leaks out.

If you think it matters, loosen the nut and rotate the core, even remove the bolt and set the core at various angles.

I *have* had trouble with a huge torroid core in a hasty-design mixer (I think we got the only one in this configuration before the company closed). My ultimate answer was to extend the leads outside the mixer case and let it dangle behind. Very ugly, but effective; and for other reasons we soon got another mixer.
 
Back
Top