NE555 interrupter and GDT problems / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

NE555 interrupter and GDT problems

first previous 4 5 6 7

Move Thread

LAN_403

Tonic

Tue Dec 04 2007, 11:37AM

Registered Member #528 Joined: Fri Feb 16 2007, 10:32PM
Location: Warsaw, Poland
Posts: 166

And what value would you suggest? Is 1kohm in series with collector fine?

Tonic

Thu Dec 06 2007, 04:39PM

Registered Member #528 Joined: Fri Feb 16 2007, 10:32PM
Location: Warsaw, Poland
Posts: 166

Bah, messed a bit with it..

I didn't notice the hint about values from Steve in previous post, or he did edit in secret, but the trick with capacitor connected parallely with base resistor helped a lot. The base resistor is 9kohm, seems to work fine.

The first prototype of RTL inverter (Resistor Transitor Logic):

The upper waveform is a inverted one, the second comes from NE555's output, which is very similar to one after BD's totem pole. Notice that inverterd signal has rounded begin of on-time signal part.

That's a waveform after plugging probe to two BD's totem poles outputs. Little ringings are visible, but it's possible to fix. The peak to peak voltage has doubled and seems to be working as it should be.. that's what I thought until I did plug GDT with loads. At this moment waveform was really messed and I don't remember why. Whatever, after few hours I had some decent waveforms, but soon after driver started to work horrible. In case I don't find reason, I will build another prototype of whole driver.

Second prototype of RTL inverter, didn't work. Third one was a prototype that gave decent waveforms I said about earlier. No photo of it, this time.

When I was trying to get proper waveforms, I've decided to relax a bit in meantime. GDT plugged to one totem pole, two IRFP250's, 70kHz and +30V supply voltage for half-bridge

Also, I have problem with figuring out, why transistors heat to a level where my fingers can't resist. Supply current is smoothed by 10 000 uF capacitor, fall and rise times are small (few tens of nanoseconds if I remember correctly), I don't see switching losses on oscilloscope, transistors switch at 0V. Then why? I wanted to use it for bug zapper, but if it won't last at least half hour, then no-no. IRF820 heats to high temperature in half minute. IRFP250's heats very slowly, I haven't fully tested them, but they heated a bit in few minutes and I have feeling they might heat a lot more, if they had more time.

Dr. Dark Current

Thu Dec 06 2007, 05:14PM

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

The transistors heat up because of on-resistance (I think this is obvious). If you are powering 200V transistors with just 30V you're not using their full potential.
Another thing to note is that the reactive current circulating in the primary circuit can be up to many times higher than input current,.

Tonic

Thu Dec 06 2007, 05:50PM

Registered Member #528 Joined: Fri Feb 16 2007, 10:32PM
Location: Warsaw, Poland
Posts: 166

Earlier IRF820 weren't heating so rapidly, so there's must be another reason. I've changed transistors to new ones, but it didn't help. I know it doesn't use full power from main, but what's the reason for it if at 30V transistors heats so much?

Dr. Dark Current

Thu Dec 06 2007, 08:21PM

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

Tonic wrote ...

Earlier IRF820 weren't heating so rapidly, so there's must be another reason. I've changed transistors to new ones, but it didn't help. I know it doesn't use full power from main, but what's the reason for it if at 30V transistors heats so much?

Well.. MOSFETs are best used at a voltage a bit under their breakdown voltage, for example for 320V you can use 400V or 500V FETs, any higher and it becomes ineffective. This is because higher voltage FETs are 1)more expensive 2)have much higher on resistance for the same current than lower voltage ones.

Tonic

Thu Dec 06 2007, 09:28PM

Registered Member #528 Joined: Fri Feb 16 2007, 10:32PM
Location: Warsaw, Poland
Posts: 166

But momento, I've had two tests with IRF820's. The first test was done when I got half-bridge working at first time. Transistors were heating slightly. Second test was done after I upgraded my driver. The same transistors were heating much more and faster. Just now I swapped IRF820 to IRFP250. I hope I clarified situation

Are you saying that IRF820 are dramatically heating because (besides higher R(ds)on) they're working on much lower voltage than they're supposed (max is 500V)?

I've done some comparing between IRF820 and IRFP250:

IRF820:

Drain to source voltage : 500V
Maximum power dissipation : 50W
R(ds)on : 3 ohm
Maximum continous drain current: 2.5A (at 100*C it's 1.6A)
Pulsed drain current : 8A

IRFP250:

Drain to source voltage : 200V
Maximum power dissipation : 180W
R(ds)on : 0.085 ohm
Maximum continous drain current: 33A (at 100*C it's 21A)
Pulsed drain current : 130A (wow! I hadn't idea those transistors are so good)

It's obvious that IRF820's are inferior to IRFP250's. The interesting thing is continous drain current for IRF820 which is only 2.5A, but datasheet says the pulsed current can be 8A, so I think problem isn't there, but I'm going to check, how much currents half-bridge eats.

By the way, now I got driver working giving nice gate waveforms (yeah, finally have a 1:1:1 radtio GDT)

Now it's about time to test with main, but I will do tomorrow. Do you have advices for me? It will be my first time to plug main directly for high voltage. I will use variac, which can lower 230 VAC to about 170VAC, after rectifing 240 VDC.

Dr. Dark Current

Fri Dec 07 2007, 02:04PM

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

Tonic wrote ...

By the way, now I got driver working giving nice gate waveforms (yeah, finally have a 1:1:1 radtio GDT)

Of course you must use suitably rated FETs, 400V minimmum.
When tuning, I suggest you to put a small light bulb in series with mains, the risk of something blowing up is greatly decreased. When you have it running well with small (~25-60W) bulb, then you can go larger, 100W, 250W and 500W if you have, if you are the sure it's working well you can leave the bulb out. It is also suggested to plug the gate driver's supply before the bulb that it gets full voltage all the time.

And of course- use a fuse!

You want to drive a flyback with it right?
For a standard flyback, 25-50 turns on the primary is about right, and frequency 20-60 kHz.
With the small bulb in series with mains, tune the frequency to where the bulb is the dimmest (with no load on the flyback). This will ensure you are safely away from the resonant frequency.
If you power it up without tuning and without current limiting (the bulb) and you are unlucky to hit the flyback's resonance (or any of its harmonics), you can be sure that your flyback bursts in flames and your driver exploes.

Good luck

first previous 4 5 6 7

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