steve's mini sstc isseu heat with the ucc37... 22 and 21 ic's/ EDIT: A new problem arises. / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

steve's mini sstc isseu heat with the ucc37... 22 and 21 ic's/ EDIT: A new problem arises.

previous 1 2 3 next

Move Thread

LAN_403

Steve Conner

Thu Aug 13 2009, 12:22PM

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

R v.d. Tuuk wrote ...

o dam it's yellow :(

Well there's your problem then. Yellow toroid cores are made of iron powder, which is the wrong material for GDTs. This will cause the transformer to draw too much magnetizing current, overheating the gate drive chips.

You want a ferrite core (grey colour)

Duality

Thu Aug 13 2009, 12:27PM

Registered Member #1951 Joined: Sun Feb 01 2009, 01:59PM
Location:
Posts: 105

so this is the gdt i use now:

and the picture under it is the one i need to use?
just checking

Steve Conner

Thu Aug 13 2009, 03:27PM

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

Well the core in the top picture is obviously the wrong kind. It's not really obvious what the one in the bottom picture is made from: it's ferrite, but there are several grades of ferrite and not all of them are good for GDTs. However you might as well try it since you've got it.

Duality

Thu Aug 13 2009, 06:40PM

Registered Member #1951 Joined: Sun Feb 01 2009, 01:59PM
Location:
Posts: 105

ah thanks i'll try it :P

hvguy

Fri Aug 14 2009, 07:06AM

Registered Member #289 Joined: Mon Mar 06 2006, 10:45AM
Location: Conroe, TX
Posts: 154

Ya, get the right core on there first

GDICs do run hot (80c) in some coils. Here are the determining factors:
IC current rating, more specifically its Rds on, the higher the current the greater the losses in the part
IC bypass capacitors, should be EVERYWHERE in your circuit, preferable solder right across the terminals
Gate resistors, high values equal cooler parts but slower gate rise/fail times
Gate capacitance, the higher Qg the higher the amount of power required to charge it
Transformer core material, must be a good high perm core material
FREQUENCY, ya, this is obvious, higher frequencies equal MUCH hotter parts. I try to keep my coils under 200KHz...
Logic voltage, higher voltage means higher current, higher power, and hotter parts
Use a series DC blocking cap, which out this your parts will likely just be dead.

Think that covers the basics...

FireBird

Fri Aug 14 2009, 08:27AM

Registered Member #1104 Joined: Tue Nov 06 2007, 07:38PM
Location: Wisconsin, United States
Posts: 34

I have this same problem with my ucc's but for the most part I ignored it because my ferrite core was just some random one I found (for a choke). I later discovered that when thereâ€™s no power going to the bridge, the uccâ€™s get extremely hot (too hot to touch anyway). However when my bridge is powered, my uccâ€™s stay somewhat cool. Oh and I donâ€™t use gate resistors â€¦ but I though that only affected the mosfets.

Duality

Fri Aug 14 2009, 12:07PM

Registered Member #1951 Joined: Sun Feb 01 2009, 01:59PM
Location:
Posts: 105

i had the sstc working with the new gdt,
it really gave better output i noticed.
but then the whole thing stopped so i checked some stuf.
and i blew a regulator? because the regulator was not really giving the 12 volts as it should.
it only was giving 2 to 3 volts.

anny way i am getting a new one today

CT2

Fri Aug 14 2009, 02:20PM

Registered Member #180 Joined: Thu Feb 16 2006, 02:12AM
Location: Ontario, Canada
Posts: 187

If your regulator was a Lm7812 or something like that then its possible that your UCC's overheated, or shorted, and what happens is the 7812 is only rated 1 amp, so when it tries to pull more it overheats and shuts itself off befor damage can be done. I've had this happen, try it now after its cooled down and it might still be good, it just shuts off when it overheats.

Marko

Fri Aug 14 2009, 02:49PM

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

Well, that's why I never use a regulator on gate drive IC's power supply. Just a transformer of around 10V AC output, rectified and filtered with a large capacitor works more than well enough. A few volts of ripple is no problem for gate drive, and other loads like fans as well.
This greatly simplifies the design, and also saves a lot of power that would otherwise be wasted on the regulator.

Only drawback of this is that the UCC's are likely to blow up impressively when you short them - but I had them blow up anyway with a regulated power supply too. If you want to save them from such accidents, some sort of active protection system is needed.

To R v.d. Tuuk: I don't know what kind of wire did you use for your new GDT, but if it's enameled wire like in the first case, I'd strongly recommend to re-do the transformer with plastic insulated wire, such as from UTP cable. Enamel is rather easy to damage while winding it onto core, and windings could arc mutually or to core under high voltage stress.

Marko

Duality

Sat Aug 15 2009, 10:46AM

Registered Member #1951 Joined: Sun Feb 01 2009, 01:59PM
Location:
Posts: 105

ah i see, i'll rewind the core then.
as for the regulator.
i'll try to get it to work again.
anny way if not.....
i may consider using a 10vac transformer.
as marko sayed

then i'll modify the board. but that is easy

litlle edit:
yes it was the shut down meganism of the regulator.
looks like it's back working again.

i think i am gonna put a little heatsink on it now.
atleast until i have found a good transformer as supply.

previous 1 2 3 next

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