Impedance matching woes / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Impedance matching woes

previous 1 2 3

Move Thread

LAN_403

Sulaiman

Sun Sept 23 2007, 11:46PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

Joe, what frequency are you operating at?

If it's below about 500 kHz I would also vote for gapped ferrite,
mainly because I have loads of ferrite cores and more experience with ferrite,
I have also used MolybdenumPolyPermalloy (MPP) from Magnetics Inc.
which is really good for audio to 100s of khz, but it's rather expensive.

Marko

Mon Sept 24 2007, 07:28PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

According to Richie, iron powder is too lossy. It's only for DC with relatively small ripple. The matching inductor current is pure HF and would cook an iron powder toroid in seconds.

That's interesting because I often used #26 as a HF ballast without problems. Maybe I wasn't putting enough power through it. Really, what can be the problem?

Explanations like 'it's for DC with small ripple' lead to grave misconceptions, which I fell into too.

#26 hydrogen reduced toroids are actually one of most used cores for power converter inductors, for frequencies like up to 1Mhz and don't work with anything like 'small ripple'. In a discontinuous mode converters it stands very large voltage swings from 0 to converter voltage of without problems.
And it is very low loss in apps where it's used.

I never imagined why wouldn't it be good for symmetric AC use too. BTW didn't you say it would be best inductor choice for a SLR converter?

AFAIK carbonyl cores are for high frequencies >10Mhz an like, not something to be useful here.

uzzors2k

Mon Sept 24 2007, 09:00PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

I tried some different cores I had on hand, including an ATX psu core, and all of them heated. The air core didn't seem to heat, but neither did the work piece. The only core type left to try is ferrite with a laaarrgge air gap, something like what Richie used. It must be a simple matter of tuning, but I haven't been able to tune much because the cores heat so fast.

Anyway, I switched to series resonant with a matching transformer. When tuned for resonance I get the greatest power consumption yet. The work piece starts smoking after seconds, at which point I turn off the heater since the matching transformer is doing the same. I need to wind some really heavy litz wire on the secondary before I let it run long enough to turn bolts red hot. The power of this system is starting to surpass the safe limit of the two polyester 470nF caps I'm using for the resonant circuit.

Dago

Tue Sept 25 2007, 11:46AM

Registered Member #538 Joined: Sun Feb 18 2007, 08:33PM
Location: Finland
Posts: 181

I'm using an ATX ferrite on my LCLR heater and only the winding (need proper litz-wire) + ferromagnetic metalclips over the bobbin get hot.

uzzors2k

Wed Sept 26 2007, 04:53PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

I tweaked it some more and set up sufficient cooling. The matching transformer is from a monitor smps and the secondary is three turns of tetrafilar 18 awg wire. With the fan it stays cool enough to touch. I choose series resonance and matching transformer over LCLR just because it worked the best first.

1190825567 95 FT24318 Series Resosont Macthing Transformer

After 20 seconds the middle of the bolt is dull red, and after a minute the whole this is glowing. My cell-phone camera had slight problems with the IR.

Marko

Fri Sept 28 2007, 05:56PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Why do you guys just disappear when something this important comes to point?

Why material #26 isn't suitable for high frequency chokes?

It's definitely very efficient in DC/DC converter applications into hundreds of kHz. There it stands massive triangle ripple. Why is it bad to push sine current through it after all?

Sulaiman

Fri Sept 28 2007, 06:56PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

Marko, the answer is ... #26 is very good as a low frequency energy storage core but very lossy at high frequencies
have a look at POWER CONVERSION NOTES 1) HERE

Steve Conner

Sat Sept 29 2007, 07:32PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Well, I have better things to do with my time than argue over stuff like this. It's not like I'm getting paid for it.

The triangle wave ripple current that you mentioned in an output choke is typically about 10% of the DC current. That's not "massive" compared to a 200% peak-to-peak sine wave.

Hydrogen reduced iron powder cores are only used in applications like these, where they filter a small AC ripple on top of a large DC current. They are never used in discontinuous-current converters, because gapped ferrite works better.

I only realised this recently in discussions with Richie, and I apologise if I gave bad advice in the past.

Marko

Sat Sept 29 2007, 07:50PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Well, I have better things to do with my time than argue over stuff like this. It's not like I'm getting paid for it.

Steve, who said anything about arguing? I may have been a bit more provocative just not to get ignored and spread misconceptions further due to my ignorance.
At any time I can't forget where would I be if there weren't you.

It's not for a discontinuous converter either? I have a lot more to learn about these things

So I apologize: definitely DON'T use the #26 core for your series inductor. I would now probably go for air core as you should be able to pass with smaller (>100uH) inductor if higher power is required.

previous 1 2 3

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