Help with PSU (Temp control fan, load bank, & PWM circuit)

[ronv]

I wouldn't test the bad one for sure. I think the 170 volts might have been a real number. That is about the voltage of the PSU from the AC line (Peak voltage).

 

[jocanon]

wow, that's high...on a side note, maybe I should invest in a higher quality volt meter. I have two $5 voltmeters and they are giving me two different readings

 

[ronv]

None of them will capture the quick changes. What would be nice is a storage scope.
Doubt if you will find a 26 volt TVS at radio shack, but maybe 2 15 volters.

 

[jocanon]

2 15 volts in series? I think a storage scope is in the $150 to $200 range if I am not mistaken.

 

[ronv]

Yes 2 in series.
The fact that the TVS blows says the voltage was way out of spec.

Yea even the little hand held scopes are 150-200. Sometimes you can find a good one on ebay but you have to be careful.
I bought my las one on ebay, but it was a local seller so I could go play with it.

 

[jocanon]

I am thinking if this is going to be an issue with these PSUs, then at some point I should invest in a good storage scope to test each one before I sell it, especially since I plan on guarunteeing them for a year. The way I am looking at it is there are a couple reasons why someone would pay me to make them a PSU instead of doing the mod themselves, 1. they don't want to take the time to learn how (big one), 2. they trust that if they buy it from me it has been tested and works (and is guarunteed in the event it fails).

So, is my thinking right, that really I need an oscilloscope to properly test the PSUs?

 

[ronv]

I don't think you have to go that far, but if they are overated and tend to konk out a large stock of TVS's might be a good idea.
One of the little voltmeters or your good on might give an indication of a bad one if the voltage goes screwey before it goes.

 

[jocanon]

I don't think you have to go that far, but if they are overated and tend to konk out a large stock of TVS's might be a good idea..

lol, yes a very large stock!

 

[jocanon]

Maybe I should solder on some sort of plug that I can snap them into so I don't have to solder each one...well, I'll wait, if I am replacing these like every day, then maybe

 

[()blivion]

OK, watched the video. What's probably happening is the current traveling through the wires is blowing your TVS out when the system shuts down. This means the TVS can't handle the energy pulse being supplied to it, which isn't good.

Edit: Or the PSU is bunk, can't forget this.

I don't know for sure if this will work but you can try maybe putting two or more TVS's in parallel with each other. This *should* let them do some energy sharing when that big pulse of energy hits them. However, it's possible the exact trip thresholds of the TVS's are to far apart (do to parts tolerance). If so then only one will trip, take all the energy onto it's self, heat up, and explode like the one is doing.

Option 2 is to get bigger TVS's, if you can find any.

Option 3 is a full crowbar circuit of some kind, such as using the circuit we have and just slamming the MOSFET's fully on.

Finally, could roll without a TVS at all, which will most likely take out a MOSFET instead of the TVS's next time the surge happens.

-()b

Edit: Looks like I was a little late to the party... LOL.

 

[jocanon]

I blew a TVS once when I was just running it at 47 amps continuously without tripping the over current protection.

 

[jocanon]

Yep, bad PSU. I wired up another isolated one and the voltage is solid, actually it dropped from 25v at no current to 24.8 at 47 amps, it was constant 24.8 though, not bouncing around at all. I think a small .2 voltage drop is normal, right?

 

[ronv]

I would say .2 over 50 amps is real good.
Glad it was just a bad one.
Now you know why you built the tester!

 

[jocanon]

you better believe it!

 

[jocanon]

My only question now is, do you think tripping the over current cause the PSU to go bad, or maybe it was bad to begin with and I just didn't know.

 

[ronv]

It's really hard to say. Often safety circuits aren't as well designed and tested as the basic unit.

 

[jocanon]

I think just to be on the safe side, I will not trip the over current protection on the units I sell. I will test them up to 50 amps continuous.

 

[ronv]

Yea maybe a good idea.
Do the fans run at full speed when you run the current up.
Did you put them on the heat sink?

 

[jocanon]

yes, and yes - or at least pretty close to full speed...I haven't fully tested it yet since I didn't run it at full speed for very long, but after a minute or so, the fan speed was getting up there. If it doesn't go full speed, I just need to move the thermistor closer to the current sense resistors.

 

[ronv]

I was worried about the PSU fans.