Silicon Chip Jan/Feb 1992 - 40V 8A power supply

[joelby]

Hi all,

Does anyone have a copy of Silicon Chip from Jan & Feb 1992, or specifically, the article on the 40V 8A power supply? I have such a power supply, but not the schematic and other details, and mine failed last night.

I went in to the State Library, but they only have copies of the magazine dating back to 1993. I photocopied the revised version of the power supply from the 1998 issue and might consider upgrading mine if anyone can remember back this far and thinks it's a good idea It shares the same case and the main toroid, so it wouldn't be too expensive to build.

I suspect the power MOSFET has blown - things that cause me to suspect this is that giant headline on the revised power supply article stating that the new one is short-circuit proof, and a comment saying that the power MOSFET was prone to destruction in the original one. I'll try replacing that if I can get in to see what it was.

[amspire]

You can buy the Jan 1992 issue from the Silicon chip website, along with a photocopy of the feb 1992 article.

Also you can buy the April and May 1998 issues that had the updated 40V 8A power supply.

[joelby]

Touché!

I didn't really want to spend $24 on two articles to fix what's probably a very simple problem, but if replacing the transistor doesn't help, I might have to

[amspire]

I don't know the circuit, but if the controller IC is directly connected to the MOSFET, it may have blown too when the MOSFET shorted. May be worth buying a spare when you get the replacement MOSFET.

But it should be easy to fix. If any other power devices failed, they are probably shorted too.

Richard.

[joelby]

I tested the power MOSFET using my DMM on diode mode (charge up the gate, test drain-source, then discharge gate-source with fingers) and it seemed okay. I'm not sure if it was mounted on the board properly, though, and while I was fiddling with that, both legs fell off the MUR1515 diode. Unfortunately it's not really clear if this was the original failure, since it's been happily sitting on the bench for a year since I last shifted it!

Having spares of all the chips seems like a good idea, though! It looks like it doesn't use a dedicated controller IC, but rather "ua 339PC 8811" (a quad comparator?), 74C14 Schmitt triggers, and an LM324.

I like the opto-isolator arrangement a lot - now that's isolation!

[PeteInTexas]

I think Dave has a stash from when he was a boy according to one of his videos.  You might be able to persuade him to crawl up in the attic and retrieve it.

[joelby]

Yes! He should haul them down and set up a Silicon Chip and EA lending library in the new lab.

[PeteInTexas]

Yes! He should haul them down and set up a Silicon Chip and EA lending library in the new lab.

I wonder if he would be in violation of SOPA if he simply talks about the article you refer to in one of his videos?

[amspire]

Having spares of all the chips seems like a good idea, though! It looks like it doesn't use a dedicated controller IC, but rather "ua 339PC 8811" (a quad comparator?), 74C14 Schmitt triggers, and an LM324.

Excellent. The uA339 is the same as the LM339. They are all garden products and are probably all available at Jaycar stores, the usual distributors, or even just on ebay. You could probably get 10 of each delivered for under $10.

Back in 1992, there wasn't a great range of fast switching diodes like the MUR1515, particularly in Australia. Now there are heaps of alternatives, so you could probably find a tougher alternative if you wanted. Perhaps a schottky diode. Replacing diodes can be tricky though. You can ask about a replacement if you are not sure.

I like the opto-isolator arrangement a lot - now that's isolation!

That is a bit of fiber optic cable, is it? It looks like a 30KV isolation solution on a board with a few hundred volts isolation. I can only assume they did it to eliminate capacitive coupling between the sides, rather then just isolation.

Richard

[joelby]

Yes, I'll order replacements of everything from element14, and can get to work on it at 7am when the courier bashes on the door. I'll just stick the to existing diode, since it still exists (actually, I substituted the MUR1520G), and the closest alternative I could find to the BUK436-100B FET with only two minutes of research, the IRFP150N.

In the revised edition, they replaced the optical fibre with.. well, no optical isolation at all. Maybe chip optocouplers were still expensive in 1998?

[amspire]

In the revised edition, they replaced the optical fibre with.. well, no optical isolation at all. Maybe chip optocouplers were still expensive in 1998?

Not surprised. It doesn't look like there is any HV in that area al all, so it looks like a very expensive way to minimize capacitive coupling.

Probably the designer just wanted to use some fibre-optics.

[joelby]

You're absolutely right - 240 VAC goes straight into two transformers - a small 12V one and the 35V toroid. The revised design ditches the former and generates 12V with a 7812. No high voltages are present on the PCB.

[EEVblog]

I think Dave has a stash from when he was a boy according to one of his videos.  You might be able to persuade him to crawl up in the attic and retrieve it.

I only have some Silicon Chips. They are at the lab now, so can't check. But doubt I have that issue.
I collected EA's, and had them all from 1963 or so. But sold most of them :-( probably still got 10 years worth in the roof though.

Dave.

[joelby]

Good news, everyone! I replaced the MOSFET and diode and the power supply now works again. An easy fix! I'll hold off on upgrading it to the 1998 MY spec for now

Thanks for your help, Richard.

[amspire]

Great.

It looks like a useful power supply. 40V at 8A, super easy to service and it only uses really common parts.

I saw that one recently was sold on Grays Online Auctions for over $200, and with Gray's commission and delivery charges, that would have totaled over $250.

[joelby]

Yeah, I think the Grays Online one turned up while I was Googling for more information and my mind was suitably boggled. I paid $100 for mine on eBay two years ago, which I thought was pretty good value for an 8A supply.

The analogue ammeter is quite nice, but I do wish it had a digital voltmeter and perhaps a fixed 5 V setting rather than 13.8 V. It looks like that can be changed with a trimmer, though - something to do while the lid's off! Now for a quick trip to Jaycar to get some thermal grease.

[M. András]

a shematic for this supply would be good