2
\$\begingroup\$

I realize that the most common way to mount a toroidal-core transformer is by sandwiching it between two rubber-padded plates with one long screw holding them together and to the case (or whatever it's mounted to).

But I was thinking of another way to mount a toroidal core, in particular the kind that's made of a roll of iron alloyed with silicon (which is the best for low-frequencies, such as 50hz/60hz power transformers). What if I were to drill and tap 2 adjacent holes into the core (perpendicular to the laminations), and then use screws to connect it to a case. If the case is the perfect width, I could even put another screw into the opposite side of the core to attach it to both sides of the case. Exploded

Mounted

This would also effectively ground the core. That's a good thing, right? Doesn't it somehow add stability to the transformer?

The only problem that I can see is that the screws would electrically link the plates together.

Other than that, I cant see why toroidal cores aren't ever already mounted this way.

Edit: Note that I am referring to the toroidal cores that are made of a roll of silicon steel (notice the spiral shape in my pictures), mostly for sizes ranging from 4-15cm in diameter as in mains transformers. Not the ferrite or iron powder cores used for higher frequencies, which tend to be too small for much benefit from this method anyway.

I also changed a sentence that used to say "Doesn't [grounding the core] somehow reduce problems when the core is getting saturated?"

\$\endgroup\$
9
\$\begingroup\$

You mention 'laminations', so you're obviously talking about an iron core.

The major issue is that toroidal cores are wound with copper all the way round, for packing efficiency. If you drilled into the side of a standard wound transformer, the drill hole would cut through every winding. This means that a toroidal designed to be mounted this way would need relatively large areas left not wound to allow access, sufficient to maintain insulation integrity to the high voltage windings. That would be a fairly large hit in VA and efficiency, for the 'benefit' of being able to be mounted this way.

A second issue is the reduction in core cross sectional area, which would directly result in a reduction of volts/turn, so further reduce the transformer VA.

A third issue is the one you mentioned of shorting laminations together, which will increase core losses, reducing efficiency.

Electrically grounding the core has absolutely no effect on core saturation.

There's not a lot broke with the clamping method of fixing a toroidal transformer, certainly nothing to be gained trying to fix it this way.

\$\endgroup\$
  • \$\begingroup\$ Would the unwound section of the core really reduce efficiency that much? I estimate that this would be 1/6 to 1/8 of the core (45-60°); still much better covered than an EI transformer. Would this really bring the (electrical & VA/size) efficiency almost down to EI levels? \$\endgroup\$ – Electric-Gecko Jan 14 '17 at 0:58
  • \$\begingroup\$ I should also ask about what you said about cross sectional area. Is the cross-sectional area of only the narrowest part relevant? Anyway, thank you for the information, as you also clarified some other things I didn't understand about transformers. As for the mounting method itself, I was thinking that my method would be easier in many cases, better for cooling, and also not require any rare parts. \$\endgroup\$ – Electric-Gecko Jan 14 '17 at 1:06
  • \$\begingroup\$ The relevant cross section for saturation is the narrowest part. When that saturates, the permeability of that bit falls, putting an increased reluctance in series with the rest of the core. Look for a toroidal winding machine vid on youtube, that's how they all work, gaps would be tricky to leave. If you drill the holes first, you have to avoid them, if you drill later, there's magnetic swarf in the transformer. There are so many downsides, damageing the magnetic circuit, compromising the electrical insulation, reducing the packing. \$\endgroup\$ – Neil_UK Jan 14 '17 at 6:09
  • \$\begingroup\$ It's rather like suggesting keeping a toddler in a high-chair by putting a bone-screw through their pelvis. Are you aware of the other methods of fixing? Epoxy a nut into the hole in the middle, or mold a nut with plastic in the middle. Ty-wrap it into a U-section, which then gets screwed down. Slot it into a molding of the correct size pre-formed in the case. \$\endgroup\$ – Neil_UK Jan 14 '17 at 6:12
1
\$\begingroup\$

The material the core is made from is a sintered ferrite and physically almost behaves like a ceramic. Drilling it will be difficult if not impossible without breaking it and putting a screw in it will not give a mount of any kind of strength.

If you need to mount it vertically you can usually get a plastic bracket to put it in.

\$\endgroup\$
  • 1
    \$\begingroup\$ Laminations != iron powder \$\endgroup\$ – winny Jan 13 '17 at 11:22
  • \$\begingroup\$ You are correct, I missed that both answers add up to a more complete whole and a lot of people are using ferrite so I will leave it up. \$\endgroup\$ – RoyC Jan 13 '17 at 11:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.