Big mo-fo ferrite HV transformer driving (another CCPS/SLR inverter) / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Big mo-fo ferrite HV transformer driving (another CCPS/SLR inverter)

previous 1 2 3 4 next

Move Thread

LAN_403

101111

Tue May 20 2008, 06:32AM

Registered Member #575 Joined: Sun Mar 11 2007, 04:00AM
Location: Norway
Posts: 263

All this makes me want to make my own, is Ebay the only good place to buy big ferrites?

uzzors2k

Tue May 20 2008, 05:50PM

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

Pretty much. That or Hamfests, which we don't have in Norway.

Finn Hammer

Tue May 20 2008, 06:39PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Ace7one wrote ...

All this makes me want to make my own, is Ebay the only good place to buy big ferrites?

uzzors2k

Sat Nov 08 2008, 12:48PM

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

I'm going to continue on this project, as I've found a solution to the corona problem, namely oil submersion. It's messy, but the best solution I can think of. So far I've relied on math to estimate the amount of primary turns I need, but is there a way to find this in practice? I placed a 4 turn secondary winding on the transformer and started with 20 turns for the primary. (Here's the datasheet for the core.) While decreasing the amount of primary turns I watched the secondary waveform, but eventually I was at 7 turns or 45V/turn at 50kHz with only minor signs of drooping on the squarewave. Is this the right way to check for saturation, and how does loading a winding affect saturation?

EDIT: Does anyone know the highest flux density and frequency I should operate this core at? (N27)

Finn Hammer

Sat Nov 08 2008, 09:39PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Uzzors wrote ...

EDIT: Does anyone know the highest flux density and frequency I should operate this core at? (N27)

I would guess around 0.15 - 0.17 Tesla.
Magnetics has a lot of good literature

and you should read this one:

This one is a must:

I know it would be better with info from Epcos, but they are really arrogant. I once asked if they had cores that would cover the needs for Daniel and my CCPS, and they just told me to contact a transformer manufacturer, because the design of a 50kHz high voltage transformer needs special skills.

Ferrites can be run to saturation below 20kHz without getting too hot. at higher frequencies, the core is said to be loss limited. This means that the core losses have to be diminished by lowering the flux density.

Cheers, Finn Hammer

rp181

Sun Nov 09 2008, 03:55AM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Epcos isnt arrogant, i almost got a 20kJ capacitor bank sponsorship =p

Ide say the best method is to build a variable drive, and play with frequency.

Finn Hammer

Sun Nov 09 2008, 09:00AM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

rp181 wrote ...

Epcos isnt arrogant, i almost got a 20kJ capacitor bank sponsorship =p

I guess I put it bluntly. Of course Epcos are many persons, and the responce you get, depends on the person you happen to get in contact with. However, I have noticed a different, more friendly, attitude from american companies, when approaching them for information. Americans more often respond with: Why don`t we just send you a sample, then you can find out if it works. European style more like: If you don`t know already then go study more before wasting my time. That`s what I ment with "arrogant".

When you say "almost", what does that imply: Did you get the caps or only almost, but not?

rp181 wrote ...

Ide say the best method is to build a variable drive, and play with frequency.

There has been ferrites around for a long time, and the television flyback transformer moved the application of them out of the black arts and into the well understood enginering tasks. It is possible to arrive at better than ballpark values with less than starting from scratch. No need to invent the wheel over again.

Cheers, Finn Hammer

rp181

Sun Nov 09 2008, 02:18PM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

I say almost because, they offered, but my school's district legal department said it would open them up to all kinds of liability issues =(.

What you say makes sense...

Dylan

Sun Nov 09 2008, 10:47PM

Registered Member #1024 Joined: Sun Sept 23 2007, 10:56AM
Location: Northern NSW, Australia
Posts: 95

101111 wrote ...

All this makes me want to make my own, is Ebay the only good place to buy big ferrites?

Inverter microwaves have an excellent ferrite core in them, ready wound with a nice lenth of litz wire for your primary.

uzzors2k

Sun Nov 16 2008, 04:54PM

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

I picked a number of primary turns the easy way. This transformer has to produce between 16 and 18kV, and I have two secondaries with approx 225 turns each. So I need 8 turns.

I was about to pour oil today, when I decided a few quick continuity and inductance checks were in order. Everything checked out except the primary inductance. For the first it's 370ÂµH according to my LC meter, when it was 5.5mH during the last build. The difference in turns is 8 now compared to 10 earlier. This is with no secondaries.
With secondaries installed things get worse. The primary inductance drops to some 5-10ÂµH. Now I'm 90% percent sure these secondaries have no shorted turns. Also, when measuring their (secondaries) inductance while on the core it's in the order of 20mH each. If they had a shorted turn their inductance could surely not be so large?

Did I damage my core somehow while pushing the V/turns? At one point I was up to 0.5T IIRC. Or do I have to keep looking for a shorted turn somewhere?

previous 1 2 3 4 next

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