Inductor Core and Copper Losses

[iamakda]

Dear all,

I have to build an inductor of 32nH for which I have chosen toroidal core from Micrometals T12-10. I calculated the number of turns as 5 turns. #26 AWG Wire with a length of about 30mm.

DC Current of inductor is 100 A dc where as ripple current of 40 A (pk-to-pk), operating frequency is 90KHz.

Now how I can calculate the core and copper losses associated with the inductor.

Please help.

Thanks.

[kikkoman]

Re: Inductor Core and Copper Losses

wait, what? #26 for 100A peak? no way!
you've got to do it the other way around.
first choose your wire (i'd guess #4 or thicker for that current?!), then you choose the rest.

[iamakda]

Re: Inductor Core and Copper Losses

wait, what? #26 for 100A peak? no way!
you've got to do it the other way around.
first choose your wire (i'd guess #4 or thicker for that current?!), then you choose the rest.

100 A is the dc current and ripple current is 40A(pk-to-pk).....

[tom66]

Re: Inductor Core and Copper Losses

#26 is rated 2.2A in chassis. It will incinerate at 100Arms.

Also, I've never wound an inductor myself, but would the skin effect prove problematic as it's only ~650Hz (maximum depth) for #4? So you may need more parallel windings to get the current you need through the inductor and to reduce the losses.

[iamakda]

Re: Inductor Core and Copper Losses

#26 is rated 2.2A in chassis. It will incinerate at 100Arms.

Also, I've never wound an inductor myself, but would the skin effect prove problematic as it's only ~650Hz (maximum depth) for #4? So you may need more parallel windings to get the current you need through the inductor and to reduce the losses.

Dear Tom66,
I am using one online software from Magnetics (http://www2.mag-inc.com/calculators/...ign-Calculator) and they suggested 5 x #12 (AWG) wire.....so what does this mean?

[tom66]

Re: Inductor Core and Copper Losses

AWG #12 can carry 41A so 5 strands is fine.

Maybe you could wind 5 x 160nH inductors on the same core and put them in parallel?

[Evil Lurker]

Re: Inductor Core and Copper Losses

You do realize that a single strand of #14 gauge wire will not fit thru the center of that toroid don't you?

[PeteS in CA]

Re: Inductor Core and Copper Losses

5X AWG #12 would be a little better that AWG #8, marginal for 100A IMO. I'd use #4 or #2.

[PeteS in CA]

Re: Inductor Core and Copper Losses

I have to build an inductor of 32nH for which I have chosen toroidal core from Micrometals T12-10. I calculated the number of turns as 5 turns. #26 AWG Wire with a length of about 30mm.

DC Current of inductor is 100 A dc where as ripple current of 40 A (pk-to-pk), operating frequency is 90KHz.

Thinking about this as a whole, my BS meter is pegging! Five (5) absurdities in this post:

1. As has been pointed out, AWG #26 will fuse with 100A flowing through it; skin effect will make even 5X #12 wire insufficient;

2. As has been pointed out, a T12 core is way too tiny;

3. 32nH is insignificant as a filter inductor;

4. Micrometals -10 material is an RF material, not a power material;

5. Micrometals doesn't make a T12 toroid in a power material.

B-freaking-S!!!

[tom66]

Re: Inductor Core and Copper Losses

Would add though I did use a 240nH inductor in a 30A DC-DC design (for a future project, among a hundred others), but:

- it was a three phase design (30A per phase, 450W output power max)
- it was a boost converter
- it operated around 500-600kHz
- it used some high power Coilcraft powdered iron inductors (50A saturation current! Micro-ohm resistance!)

I don't know if you could even reasonably use 32nH in a DC-DC design, because you'd either cook your transistor from using a frequency that was too low (too high peak currents) or you'd have too much lost in the transistor due to the need to switch at very high frequencies.

[iamakda]

Re: Inductor Core and Copper Losses

Thinking about this as a whole, my BS meter is pegging! Five (5) absurdities in this post:

1. As has been pointed out, AWG #26 will fuse with 100A flowing through it; skin effect will make even 5X #12 wire insufficient;

2. As has been pointed out, a T12 core is way too tiny;

3. 32nH is insignificant as a filter inductor;

4. Micrometals -10 material is an RF material, not a power material;

5. Micrometals doesn't make a T12 toroid in a power material.

B-freaking-S!!!

If 5x#12 is also insufficient as you mentioned, then what wire should I use for this application and with how many strands?

Thanks.

[iamakda]

Re: Inductor Core and Copper Losses

AWG #12 can carry 41A so 5 strands is fine.

Maybe you could wind 5 x 160nH inductors on the same core and put them in parallel?

Dear Tom,

How come you know #12 AWG can carry 41A?

Thanks.

[iamakda]

Re: Inductor Core and Copper Losses

wait, what? #26 for 100A peak? no way!
you've got to do it the other way around.
first choose your wire (i'd guess #4 or thicker for that current?!), then you choose the rest.

Hi Kikkoman,

how many strands of #4 wire should be used?

Thanks.

[tom66]

Re: Inductor Core and Copper Losses

Dear Tom,

How come you know #12 AWG can carry 41A?

Thanks.

http://lmgtfy.com/?q=awg+current+rating&l=1

[iamakda]

Re: Inductor Core and Copper Losses

http://lmgtfy.com/?q=awg+current+rating&l=1

Thanks a lot Tom

Have you ever ever designed the output filter for DC DC Boost Converter?

Thanks.

[budm]

Re: Inductor Core and Copper Losses

" The Maximum Amps for Chassis Wiring is also a conservative rating, but is meant for wiring in air, and not in a bundle. For short lengths of wire, such as is used in battery packs you should trade off the resistance and load with size, weight, and flexibility. NOTE: For installations that need to conform to the National Electrical Code, you must use their guidelines."

[tom66]

Re: Inductor Core and Copper Losses

Thanks a lot Tom

Have you ever ever designed the output filter for DC DC Boost Converter?

Thanks.

No formal education (yet!)... I just add lots of capacitors until it works :P.

For a 100A supply with a 40A inductor ripple current it is absolutely essential you use ceramic capacitors or ultra low ESR poly capacitors. (You will need a lot of them... at 3Arms ripple each, that's at least 12 but that won't last long, so figure twice that for long life.)