Smoking flyback resistors / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Smoking flyback resistors

1 2 next

Move Thread

LAN_403

Fnord

Sun Apr 05 2009, 04:03PM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

Hi,

I've been working on my flyback cap charger recently, and I'm looking to fix a bit of an issue I'm having.

I'm using Dr. sparks "improved" single transistor driver with a 2n3773 and it works brilliantly. I'm putting out like 50+ watts with 18v input. This driver uses a diode and foil/film capacitor across the emitter/collector, along with a big 25v cap between the +- terminals.

The problem is my little 22ohm resistor starts smoking and turns black within 2 seconds of operation. Now, it still works, but I dare not run it for much longer.

Now, I know I could just use a high wattage resistor as a brute-force approach to the problem, but I was wondering if there is another solution that doesn't involve so much wasted power.
I know the REAL problem is I am not an EE. I really don't understand precisely how this circuit works, and had enough trouble figuring out how the original stfd works (though it does make sense now).

Also, I've had two 2n3773's die for no apparent reason, and they were running quite cool. Another died when my rectifier diode blew up, but I think that was because the capacitor I was charging(200J+) dumped itself into the secondary and let some ungodly current fly through the primaries.

Dr. Dark Current

Sun Apr 05 2009, 04:08PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Hi,
The resistor should be rated at minimum of 5 watts.
If it still smokes, try decreasing feedback turns, possibly down to 2.

Fnord

Wed Apr 08 2009, 10:19PM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

Got a bigger resistor and it seems to be working ok (though it isn't even 5 watts).

Now if I could just stop blowing up diodes:)

Arcstarter

Wed Apr 08 2009, 11:18PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Well, the diode is protecting the transistor. If the diode explodes, that means it would have been the transistor otherwise. Just get a high current ultrafast diode.

Fnord

Thu Apr 09 2009, 12:17AM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

The rectifier diodes are what dies, not the one across the transistor. It seems to be fine so far.

I'll be ordering a bunch soon and probably put a few in parallel. I was using some cheapo 1kv radioshack diodes and they burn up pretty easily. I doubt they were even rated for such fast switching applications.

Arcstarter

Thu Apr 09 2009, 01:15AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Well the rectifier does not have to switch fast. They should be rated higher than one amp, if that just happens to be what you where using. At my radio shack they had some 6 amp 50 volt diodes that would work well. Or you could go with a bridge rectifier already made.

Dr. Dark Current

Thu Apr 09 2009, 06:50AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

The circuit will surely draw over 1A, get some 5-10A bridge rectifier.

Fnord

Fri Apr 10 2009, 02:12AM

Registered Member #2004 Joined: Sat Feb 28 2009, 11:43PM
Location:
Posts: 39

Let me rephrase that... the OUTPUT diodes are what what keeps dieing. I'm just using an 18vt battery so there is no mains rectifier involved (and for the record, I'd say it draws at least 4 amps; I'll measure it later). The output voltage is probably under 1.5kv, with 10 primary turns and 150 secondary turns.

I'm pretty sure I've heard you can't full-wave rectify a flyback output, but I'm not sure I understand why. Right now I've just been using* single diodes for testing purposes.

*setting fire to

Arcstarter

Fri Apr 10 2009, 02:49AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Well the diodes do indeed have to be fast. The voltage might be higher than you think, and the diodes might not be able to handle the voltage and are reaching the avalanche voltage.

Hard to say without being there.

Dr. Dark Current

Fri Apr 10 2009, 06:31AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

You need ultrafast diodes, such as UF4007 or BA159 (a bit slower).
When using this circuit (single transistor), you can rectify the output with just one diode. However you must get the polarity right so that it catches the flyback pulse NOT the forward one.

1 2 next

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