My first DRSSTC / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

My first DRSSTC

1 2 3 4 next last

Move Thread

LAN_403

Danielle

Sun Jun 24 2007, 11:45PM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

Ok, so I have been working on this for 2 months now and am almost done I only need to connect everything together and test it. The secondary is 4.25" 9" wound the driver board is a modified steav ward OCD DRSSTC with EVR modifications. I designed the half bridge though it is common. I would like to know what people think of this and if there is a way of testing the drive board without using it with the half bridge? I have tested the dubber and checked the voltages everywhere I could for the correct values on the driver board and everything checks out but I am worried of the smoke test.

Thanks,

Daniel.

Part Scavenger

Mon Jun 25 2007, 02:41AM

Registered Member #79 Joined: Thu Feb 09 2006, 11:35AM
Location: Arkansas
Posts: 673

Sure you can test the drive board without it being in the coil. Just build a 555 circuit to run at approximately the same Fo as your secondary, and feet that into the feedback transformer input. Or if you have a signal generator, it's pie. Then trace those waves through to the UCC's (Pin 2? I can't remember) with your O'scope. If you don't have a scope, uh, you need one, but I guess for the time being you could pass a low enough frequency through the feedback/interrupter circuits to rig up some LED testers...? I don't know how the UCC's would like it, but I'm betting with the GDT disconnected they wouldn't care.

Then to make sure your OCD is working properly, set the voltage at pin 2 of the LM311 to the voltage that corresponds to the datasheet current pulse rating of your IGBT's or even lower (now is a good time to do it, that way they won't fry when you forget to set this up later)

. Then attach your DC linear adjustable power supply to pin #3 on the LM311 and slowly bring it up, when you hit the voltage you set on Pin #2, the OCD LED should come on, and pin 5 on the 74HC14 should go low, and all output should cease. If you don't have one of those handy PSU's, a voltage divider circuit with a potentiomenter will do just fine. To test the UCC output, go ahead and test the output with the GDT disconnected between output of C5 and pin 6/7 of the UCC37322, if you want to see what the GDT form looks like, put the secondary phases in parallel and load them down with a burden resistor and measure across. The GDT waveform may not look right until you get it all hooked up together and you've got about 5-10V across the IGBT bus. Don't freak out about this one too much yet if it doesn't look right.

What do I think about it?
First off, you do good work-- you obviously spent time on it. Your driver board looks good. A couple things I noticed though, your filter caps look really small, and so does your toroid. I wouldn't worry about it at this point, but next H-Bridge revision needs to get that GE film cap closer to the IGBTs. The proximity of the GDT is not as important as that cap. The spacing near the screw terminals makes me a bit nervous too, but I am often just paranoid.

Good luck with first light!! Waitn' to see this thing fly!

Danielle

Mon Jun 25 2007, 03:59AM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

I will print your post and follow every step thank you Part Scavenger. although I do have access to a really nice scope and all other test equipment on Tuesdays so ill do it all then. Also I have another one of the GE film caps mounted directly to the +/- rails going into the IGBTs. and They will be mounted to separate opteron CPU heat sinks, or my old computers liquid cooling system.

Thaks again

Daniel

Danielle

Tue Jun 26 2007, 01:37AM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

Ok so I set it up and tested the OCD with my home made scope and the wave form is very strange. Is this what it should look like I thought this is what the output of the IGBTs was supposed to look like. (I didn't connect anything to the feedback so that may be the problem) but this is strange. or normal I don't know please help.

Edit

I added the PWM to the CT input and it looks like the previous wave form except it has high frequency modulations in it.

Part Scavenger

Tue Jun 26 2007, 02:40AM

Registered Member #79 Joined: Thu Feb 09 2006, 11:35AM
Location: Arkansas
Posts: 673

First off, it's going to be really hard, if not impossible to see an ~200Khz waveform clearly on a TV O'scope. I haven't ever used one, but it *might* be possible to see what's going on if you could pan the signal out a little more. Even so it's going to be REALLY hard to see and impossible to photograph unless you build a ramp generator and trace lock circuit. Right now, it's being screwy though.

WAVEFORMS:
Should look like this...

Feedback after running through 74HC14. (After pin 2)

Interrupter signal. Pin 3 of the UCC's also gets a signal similar to this.

UCC out (GDT primary current)

Gate drive signal on IGBT

Did you check out the 4HV wiki articles? What about the DRSSTC-FAQ on my site? It badly needs to be re-written, but it's got a ton of info.

Danielle

Wed Jun 27 2007, 12:44AM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

thanks for the help and I did fix it I will post a image of the wave form off the gates after the GDT soon and it looks really good. The wave form before the GDT looks exactly like the pics you have and I tested the LED and set it at 2.3V as I have a 1:100 CT and the current pules capability on my IGBTs are for 256A so over current is set for 230A as a safety margin. I will test the entire setup as soon as I finish my enclosure and figure out the cooling system. ETA is for Friday.

Danielle

Wed Jun 27 2007, 12:50AM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

Scope wave form after GDT 10x multiplyer so it is driven at 30V but I do think I should make the GDT from ferite to improve the wave form.
1182905578 632 FT1630 Scope

Part Scavenger

Wed Jun 27 2007, 01:52AM

Registered Member #79 Joined: Thu Feb 09 2006, 11:35AM
Location: Arkansas
Posts: 673

That actually doesn't look terrible. Try putting some more turns on the primary (which needs more turns on the secondary). Is the GDT loaded for this test? If it isn't, loading it might cut down on that little piece hanging out in the active region at the end of the trace...

Danielle

Wed Jun 27 2007, 04:30AM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

this was not loaded I will do a loaded test tomarrow and capture images of everything as they may be useful to the comunity and me the other wave forms are extreamly clean almost exactly like your samples.

ragnar

Wed Jun 27 2007, 09:20AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

The downwards-sloping nature of the positive square edges indicates the transformer is saturating -- more turns or a higher frequency could help. =)

1 2 3 4 next last

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