First SSTC / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

First SSTC

previous 1 2 3 4 next last

Move Thread

LAN_403

Shaun

Mon Nov 12 2007, 03:46PM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

Ok. thanks everyone, just a few questions before I start the next set of changes/tests on this coil.

1. I know the primary must be wound in the same direction as the secondary (it is), but does the way its connected to the bridge matter (in terms of switching the wires)?

2. What is "open loop", and how do I run that way?

Zb, I have checked and rechecked the entire feedback circuit, including the diodes, and everything is connected how it should be. I tested the diodes themselves, and they are fine.

I will try running CW and reversing the primary for now.

Shaun

Mon Nov 12 2007, 04:15PM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

Well, interesting chain of events since my last post. I tried reversing the primary connections to the bridge, and the coil actually worked. The fluorescent bulb was lighting up, and I was getting some corona and noticeable ozone. For about 2 seconds. Then...silence.

On close inspection, both FETs read zero ohms D-S in both directions, and to top that off my variac's 8A fuse blew. I was only at around 60-70 volts into the bridge. I have 2 more IRFP260s, but before I go replacing them I need to know exactly what cause them to fail.

I have a couple theories:

1. Running the logic power transformer from the variac is a bad idea. At 60V in from the variac, it would only be putting 7 VAC into the rectifier for the logic circuits. This may not be enough to successfully drive the gates, causing them to fail.

2. Primary-Secondary arcing

Any ideas?

EDIT: FETs are also shorted G-D; that just reinforces what I already know.

Dr. Dark Current

Mon Nov 12 2007, 04:43PM

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

(1) is right. Run it from full mains.

Did you check the gate waveform (with only logic power)?

Zum Beispiel

Mon Nov 12 2007, 05:09PM

Registered Member #514 Joined: Sun Feb 11 2007, 12:27AM
Location: Somewhere in Pirkanmaa, Finland
Posts: 295

7V is not enough to fully turn on a mosfet like irfp260. Instead, it acts like a low-ish resistance resistor and burns up with heat.

open loop means you figure out the resonant frequency and run the bridge from a fixed oscillator at said frequency (no feedback).

Since you already managed to get breakout, you won't probably want to try this.

Use enough voltage on the gate drive and it should be fine. Also, I noticed that you don't seem to have protection on the gates... I'd put two back to back 15V zeners in there.

Shaun

Mon Nov 12 2007, 07:04PM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

15V zeners huh...I only have 12V and 33V right now. Vgs max for IRFP260 according to datasheet is +/- 20V; I wonder if I can get away with a 33V zener?

If this bridge fails, when I order more IRFP260s I'll also get some proper zeners and TVS diodes.

BTW, I insulated the primary from the secondary with 4 layers of thickish plastic and several layer of duct tape. I'm making a new bridge now, we'll see if that along with powering the logic from 120VAC only can make this work.

Shaun

Mon Nov 12 2007, 08:36PM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

Well, there go my last IRFP260s...*sigh*

Even after all those precautions, it still failed. At the time of the failure, the probe was connected to pin 1 of the inverter and I was detecting a strange 800kHz rounded square wave (looked similar to a capacitor charge/discharge wave). I could draw short, 1-2mm arcs to a grounded screwdriver. The fluorescent bulb was lighting, but only around the primary, not so much the entire secondary.

This time I blew a 12A fuse in the variac, and the FETs are shorted all pins. BTW, I had 33V zeners on the gates.

Here's what I think happened: For one reason or another, the feedback was picking up an 800kHz source (my sec Fres w/no topload should be just over 300kHz). This was being fed into the bridge, which was switching out of tune with the secondary, causing small arcs and massive overcurrent.

I'm pretty sure its an antenna problem, we'll see in a week or so when I get more silicon.

If anyone has any suggestions, I'm open for them.

Mates

Mon Nov 12 2007, 09:50PM

Registered Member #1025 Joined: Sun Sept 23 2007, 07:53PM
Location: Czech Rep.
Posts: 566

Shaun wrote ...

BTW, I insulated the primary from the secondary with 4 layers of thickish plastic and several layer of duct tape. I'm making a new bridge now, we'll see if that along with powering the logic from 120VAC only can make this work.

Do not forget to leave gap of air between...Once the primary touch the isolation which is in contact with the secondary any kind of isolation becomes conductive...

Shaun

Mon Nov 12 2007, 09:56PM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

"Once the primary touch the isolation which is in contact with the secondary any kind of isolation becomes conductive"

I'm pretty sure an air gap is not needed. I've seen a huge number of SSTCs with the primary wound right on secondary insulation. Yes, racing sparks can occur if HV punctures this insulation, but that should not happen because the insulation is fairly thick and there's not much voltage on the primary.

Dr. Dark Current

Tue Nov 13 2007, 01:21PM

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

When you get your new FETs and connect them up, please post waveform measured directly at the gate of one FET with just logic/gate drive power. It is possible that your GDT is not ideal and is causing the FETs to go into linear region/false switching.

I don't know how big your antenna is but make it at least ~10cm long.

Have you tried switching over the primary polarity? It is possible that the 800khz you saw is just some switching noise being fed back.

If you want to look at the frequency, don't put your probe directly to the antenna, this is loading it down. Put it on the output of the gate drivers instead (or at least on their input).

Edit- part of reply deleted

Hope this helps

Zum Beispiel

Tue Nov 13 2007, 06:08PM

Registered Member #514 Joined: Sun Feb 11 2007, 12:27AM
Location: Somewhere in Pirkanmaa, Finland
Posts: 295

Shaun wrote ...

Well, there go my last IRFP260s...*sigh*

Even after all those precautions, it still failed.

Yeah, SSTCs are tricky that way

I burned 4 sets of irfp450 (total 8 deaths) on my Mark II coil before getting it to work properly. Haven't had problems since.

Suggestions:

-Use more primary turns (10 should be good), when you finally get it working, try reducing them one at a time. Also, wind the primary on a separate former. This probably isn't causing problems now, but can in the future when you get it working.

-Don't bump up the variac to full when testing. Only use 30-50V. This should make it oscillate. Even 10V should be enough.

-Check that your gate drive is good with no bridge power.

-Get those 15V zeners and wire them like in Steve's coils. Once again, this isn't probably what is causing your problems, but you need those to protect the silicon from spikes. 33V is too much. Way too much.

-Make sure you have proper grounding. A copper pipe driven into the ground works wonders.

-Try swapping the primary connections around. Do this over and over if necessary.

-Don't give up

previous 1 2 3 4 next last

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