Does anyone wind their own toroid inductors/transformers? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Does anyone wind their own toroid inductors/transformers?

1 2 3 next

Move Thread

LAN_403

flyrod

Wed Jun 13 2018, 07:02PM

Registered Member #61905 Joined: Sun Nov 12 2017, 03:27AM
Location:
Posts: 23

I've seen a few pictures of people winding their own toroids for GDT but is anyone winding their own power inductors? Anyone have tips? I've done a few in single layer to keep the parasitic capacitance low, but I'm looking for ways to improve the results. For buck/boost converters it seems like multiple parallel conductors would help some with skin effect, but should they be braided or twisted?

I'm sure there are limitations of what can be done with the machine wound toroids, but as a hobbyist are there techniques that can improve the results?

Patrick

Thu Jun 14 2018, 03:52AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

I wind special inductive circuits ALL the TIME !

and you should too! Its really hard to get power or frequency or saturation that you want, whether it be power or signal. And buying pre-made inductors and transformers is the most expensive way to get them.

The only issue ive found with making your own is the length of each unique strand. Assuming your core was made as one piece or is glued and now inseparable.

Calculate your skin effect there are calculators on google for that. but honestly smaller strands in parallel will cool better, bend better and make better packing efficiency through the window. above 150 kHz i think is where skin effect begins to be a problem in power inductors.

Not sure how far you want to go or how much advice you want but i have more info and examples if you like.

I will give you one piece of advice above all else: buy solder-able magnet wire. If not you'll regret it.

Let me show you some of my arsenal :

liberated black ferrite, added all thread to each end to mount.

some wire

belt sand them open.

450 watt ATX PSU core

winding jigs

3d printed bobbin wet wound with red epoxy.

Purchased planar cores.

flyrod

Thu Jun 14 2018, 01:44PM

Registered Member #61905 Joined: Sun Nov 12 2017, 03:27AM
Location:
Posts: 23

Thanks for the great reply! Is that first picture the circuit that you use to test your inductors/transformers?

This is a picture of a factory wound toroid and a hand wound version that I made:

Toroids

It uses 3 parallel strands, which are easier to bend, as you mentioned. I used a drop of super glue to secure the ends against the core. The 3-strand windings also have the advantage of being able to "bunch up" on the inside diameter and spread out on the outside diameter. I don't know how well the 3 strands share the electrical load, but the hand wound replacement runs cooler than the original part.

Patrick wrote ...
The only issue ive found with making your own is the length of each unique strand. Assuming your core was made as one piece or is glued and now inseparable.

Happily, the length I calculated was pretty close to what was needed. I left the strands a bit long to be safe, but I didn't end up wasting much wire. The inner 3rd strand is only a couple inches longer than the two beneath it.

ritaismyconscience

Thu Jun 14 2018, 06:28PM

Registered Member #62937 Joined: Wed Jun 06 2018, 08:20PM
Location:
Posts: 32

I take ferrite transformers from old power supplies and heat them up with a soldering iron to melt the glue.

johnf

Thu Jun 14 2018, 07:14PM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

I remove used ferrite transformers from the pcb and submerge the pin out pins in a brine solution in a soup bowl and microwave the transformer for about a minute. Careful as the core absorbs the microwave radiation like a sponge and gets extremely hot you can then pull the core halves out as the heat softens the potting varnish

Patrick

Fri Jun 15 2018, 12:14AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

your already off to good winding skills. Next study powdered metal, ferrite, iron tape/lamination and geometry.

johnf wrote ...

I remove used ferrite transformers from the pcb and submerge the pin out pins in a brine solution in a soup bowl and microwave the transformer for about a minute. Careful as the core absorbs the microwave radiation like a sponge and gets extremely hot you can then pull the core halves out as the heat softens the potting varnish

wow, never heard this ! looks like i learned from this thread too !

flyrod wrote ...

Is that first picture the circuit that you use to test your inductors/transformers?

No thats a full featured custom SMPS, meant to put out 500 or more watts from a wall socket. its just for signal testing the ICs at the moment it does use that 250 watt atx core.

its the first module of a future power factor corrected 240 VAC 2kW smps which will be isolated and put out 20-30kV in a single step. hopefully with a fixed load it can stay near 95% eff. that means 100 watts of heat, still alot.

but thats a lot of ifs' and maybes' its already been 2 years.

Ash Small

Sat Jun 30 2018, 04:07PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I just finished winding the one on bottom right of picture last night.

I'm using 3" type 26 D toroids (the D stands for double thickness)

wire on both is 14 SWG, if I recall correctly, 1.8mm diameter, 2.5mm^2 cross section, good for 9.6 Amps, apparently (10 amp)

the one on lower left is, if I recall correctly, 1.75mH, and has 250 grammes of wire

The one on the right has 500 grammes of wire. This worked out to be exactly four layers.

Winding it was a nightmare, the first layer was very slow going, with lots of bird's nests to untangle. Part way through the second layer it became a bit more manageable. By the third layer the wire had started to work harden, this just bacame worse and worse, but at least the length of wire was more manageable.

I started winding it weeks ago, then had a break while I was doing other things, then finished it off over the past week.

I've another half a dozen or so to wind, but not in a rush to start the next one yet.

It's not as pretty as it might be, my brain had turned to jelly by the time it was finished.

They will be used for conditioning the power supply for a single ended class A amp, along with about forty 4700uF stud mount electrolytics

EDIT: I've not yet measured the second one, I'm aiming for around four mH but will be happy with whatever I get, I only need around 2-2.5 mH, anything else is a bonus. I'll post details of the test over on my thread regarding powe supply simulators.

flyrod

Wed Jul 11 2018, 05:26PM

Registered Member #61905 Joined: Sun Nov 12 2017, 03:27AM
Location:
Posts: 23

So in this picture of a power supply, there are a variety of toroids:

L1600

In some cases the parallel strands are twisted together before being wound on the core. Why would you do this vs. parallel strands not twisted together?

johnf

Wed Jul 11 2018, 07:14PM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

It maximises surface area for skin effect and twisting together as they in parallel reduces the interwinding capacitance so raises the self resonant frequency

hen918

Wed Jul 11 2018, 10:14PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

It also reduces the leakage inductance, as each strand has almost exactly the same average distance to the core

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob