Can anyone recommend some good reading on design/selection of magnetics for SMPS design?
It just seems that there's a lot of crap out there when I search for some good reading on the subject, so if anyone has a great reference, I'd appreciate it.
I'm working on a project, and my preliminary calculations seem to indicate that I need a 1mH inductor that can pass a peak current of
10A (1.5A ripple). Physical size is a concern, too.
What topology? What voltage? That's a mighty big inductor for modern switching frequencies.
It's easy to just use online software, National or Coilcraft or whatever, to design switchers.
John
I'm looking to build a 400-500W PFC Boost that puts out about 400V.
Yikes, that is serious. Is this a one-off, or quantity? If quantity, I could direct you to an outfit that will design and manufacture the inductors for you. Minntronix. The designer is Butch. [1]
John
[1] that's his actual name.Elscimar wrote in news: snipped-for-privacy@g10g2000vbc.googlegroups.com:
1mH is a pretty large inductor even for 65kHz.You could try Coilcraft they have a 570uH PFC choke originally for a NCP1653 app 300W 65Khz. If you go with a 133kHz or 200kHz you maybe able to get 400W or so out of it.
Onsemi has spreadsheets which can save some time if your doing a CCM boost.
I've used both to do up a 300W APFC. It was the easiest switcher I've ever done supriesenly.
I made one of those, BCM, with 200uH (15A peak, 10A average). It ran at
50-100kHz, depending on supply voltage and load.Such an inductor must be made from ferrite, since the delta B is way too big for any reasonable amount of powdered iron. If you insist on powdered iron, the inductance has to be so much larger (also putting it in CCM) that you'll take up more space, though it will still be cheaper (powdered irons the size of your fist are only a few bucks).
Tim
-- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms
(1) Dig up all the app notes that you can find from the various semiconductor companies, and see what they say.
(2) Get on Amazon, and look at all the switching power supply books you can find. Make sure to check the tables of contents for indications that the author talks about inductor selection, then sample those chapters (if you can) to make sure he knows what he's talking about.
(3) Basic physics and a bit of materials science: the voltage, output current, switching frequency and topology tell you what the inductor current profile will be, then you go buy something, or make it, or have it made.
(3a) Spice is nice to get that current profile...
-- Tim Wescott Wescott Design Services http://www.wescottdesign.com Do you need to implement control loops in software? "Applied Control Theory for Embedded Systems" was written for you. See details at http://www.wescottdesign.com/actfes/actfes.html
snipped-for-privacy@g10g2000vbc.googlegroups.com:
...
I was going by TI's UCC28019 worksheet. When I plug the same numbers into the ON sheet, I get an inductor value of less than half.
Kick up the frequency from 65Khz to 200Khz, and it wants a 196uH inductor at 10A max. 6A rms, which seems to be more reasonable.
Would something like a 1140-221K-RC from bournes work? I'm not sure it's the right kind of inductor....
Ferrite bobbin means low core loss, I guess the winding is solid copper though, so you will get high skin effect / eddy current / proximity effect losses. This only matters for the AC component, so a CCM converter will have lower losses than the others.
Other considerations include capacitance, which is fairly high in this style (i.e., multilayer winding), but that doesn't even matter at this current (it'll be around 300pF), and voltage capacity, which is probably good enough (the datasheet claims insulation for core and wrapper, but makes no claim about the wire enamel, so actual volts/turn you can apply is unknown).
Tim
-- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms
Elscimar wrote in news: snipped-for-privacy@q26g2000vbn.googlegroups.com:
The TI part is the same as the OnSemi part a CCM controller.
The higher the inductance the lower the ripple and you will be deeper in CCM throughout a wider line and load range. There is an optimal value most of the literature target a 20 to 40% ripple for CCM boost.
However if your just doing one or a few you might as well use an off the shelf component. The one you link to actually looks OK maybe a little bit bigger hard to be certain based on the little info on the data sheet. It does say 11.4A DC your peak current is only 10A so it shouldnt saturate or overheat with an rms current of 6A. At the rated curent it specs a 5% change and 50C temp rise in inductance and states a ferrite material.
If you check the coilcraft datasheet for a PFC inductor you will see what you should be looking for saturation current PeaK, rms current, and inductance change.
The one you link to appears suitable based on the dc current rateing and is much better then the ones in stock at newark when I was looking. Most of the ones I looked at in stock speced large inductance swings at rated current.
If you search google groups I asked a similiar ? I wound up using the coilcraft part for 300W. You could try and get a free sample off them and maybe pick up a couple from Bourns of the series you linked to and see which offers the best performance.
Does he walk around in the correct outfit?
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Don't know, never met him in person. But he is sure a good magnetics designer, and he doesn't charge for engineering or samples... they just want the production business.
I added the footnote just to head off the inevitable "butch designer" comments.
John
Aren't these the guys who occasionally send along a home-baked cake?
-- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...
Do they do Pie winding? If so I have an interest.
Cheers
Speaking of reading...
Yeah, some Swedish/Minnesota Christmas thing.
John
I think they do anything, or at least they have for us so far.
Call them and ask for Butch.
Tell him Hi for us.
John
"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...
We'll give him a try, or I should say, I'll persuade they guys to quote out the transformer to them. It's a second source thing.
Thanks
I looked at Minntronix after your post cuz I'm always interested in other magnetics companies and sources. Then, shortly after, in another office here, I saw a box with the name Minntronix on it.
Sure enough, they quoted and built some samples of one of our big inductors ! Small world !
boB
Ok. So I tried building something, and it didn't really work...
I figured the logical thing to do, before I try any actual boosting, would be to disable the FET, and see how it behaves. Rather than disconnect it, I just pull the gate to the source, to shut it off.
My FET shorted out instantly.
I've apparently done something wrong. Any idea what? I've checked all wiring. It's exactly as my schematic says.
D1 is a 600V 12A "Turbo2" diode from ST (STTH12R06DIRG) FET1 is an nFET from Infineon (SPW24N60C3) BR1 is a 35A 1000V bridge rectifier from Diodes Inc (GBPC3510) L1 is a 330uH 11.4A inductor from JW Miller A Bourns (1140-331K-RC)
Thanks.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.