How to build a $20 47A power supply

[bruce750i]

Thanks,
That's a good question. I'll stall now for someone else to chime in....

[BrianG]

IIRC, the switch attached to those small wires simply grounds them to turn it on (similar to how ATX PC supplies work with the green wire). Since you floated the grounds on each supply, I'm surprised it even works at all.

What I would do is use a double-pole single-throw (DPST) switch and have the small wires on each supply hooked to each pole. That will allow you to turn both supplies on at the same time, yet only use one switch and still have them isolated. A DPST switch is basically two SPST switches in one package.

However, a DPST switch may be hard to find, so just use a DPDT switch and just don't use two of the contacts: http://www.radioshack.com/product/in...ductId=2062499

[mistercrash]

I'm going to have to try that. Thanks Brian.

[_paralyzed_]

I haven't been in this thread in a while. I now feel wholly inadequate about the size and performance of my power supply The cases are cool

Mistercrash- BG gave you an example of a dpdt switch, but scroll down and read the reviews, that particular one appears to be crap. Hopefully you can source a better one

edit: 57 pages of them on fee-bay http://www.ebay.com/sch/i.html?_from...All-Categories

[BrianG]

Originally Posted by _paralyzed_

...BG gave you an example of a dpdt switch, but scroll down and read the reviews, that particular one appears to be crap. Hopefully you can source a better one

Well, I can say that I've personally used that switch on a number of projects and they all still work. Maybe the reviewers don't understand how to use them, or are using them in applications where more current is drawn than the switch can handle? Maybe there was a bad batch? Maybe it's a combination of user stupidity and normal number of DOA units?

In this case, the switch is passing very little current so it should be fine. But at the very least, it was an example of a DPDT switch.

[mistercrash]

I saw the reviews and I didn't pay much attention to them because I know that BG would not suggest I buy a piece of crap. Harold how could you insinuate such a thing about Brian! Shame on you. Now go build yourself a proper power supply.

[BrianG]

Now watch and the switch you get breaks proving Harold right. I would have to hang my head in shame.

If that happens, at least it's not difficult to replace it. RadioShack carries a number of DPDT switches. It comes down to how much room you have for the switch. Some are quite large, but if you have room for it, using a larger version won't hurt.

And yes, eBay also has a large selection (and probably cheaper too), but even though I don't especially like RS (their stuff is quite pricey), sometimes it's nice to be able to get something right away locally.

[_paralyzed_]

I was just parroting the reviews, I know BG knows his stuff.

I got a server power supply on the way!

Thanks to everyone in this thread I decided to start working on a power supply for myself only I'm going about it a bit differently. I have access to old used up copiers and thought since they are setup to run 24v I would give it a go to see how things worked out.

Here is a picture of the power supply that I am using. The question I have to all of you is there any way to figure out how many amps or what the wattage of this power supply actually is? If it isn't sufficient I'll most likely go the server PS route. Right now I have a nice stable 24.5 volts on this supply and will be mounting it in a custom aluminum box if I can find a way to figure out what amp's or wattage I am getting out of it.



Any ideas??

[BrianG]

Google the PS model number from a sticker or something? The printer itself should have the power requirements printed on it, and then you can make a rough calculation based on that.

The power requirements are all AC specs, this particular has a requirement of 15amps at 120v. We are a dealer of the copiers it came out of and I have the schematics and part numbers to the power supply but there are no specifications published about it.

I guess what I was hoping is that there might be a way to identify it by the components or if there was a tool that I could use to get those ratings I could go that route.

[BrianG]

I assume all the red wires are 24v and the blues are ground? If so, then loosely judging by the number (8?) and gauge of the wires, I'd say each wire is good for ~2A, which is around 16A, maybe a bit more.

What you could do is monitor the output with a voltmeter while steadily loading down the supply with resistors or something until it shuts down from over-current. Then a little math to figure out the shut-down current, and then subtract ~10% as the "working" current.

Originally Posted by BrianG

I assume all the red wires are 24v and the blues are ground? If so, then loosely judging by the number (8?) and gauge of the wires, I'd say each wire is good for ~2A, which is around 16A, maybe a bit more.

There are 11 red 24v wires ranging from 18-22gauge wire. I thought I saw a video that had a charger showing the operating amperage? Maybe that was the amperage that the user set the charger to draw.

Thanks for the idea on the resistors I'm just afraid of "frying" the supply doing that. Is that a possibility?

[BrianG]

Yeah, some chargers show input voltage/current, but they don't care if the supply is struggling as long as the voltage and current is there. And yes, increasingly loading down a supply until it shuts down can conceivably cause harm, but any decently engineered supply should shut down well before that happens.

Looking at your loosely judged "estimate" with 11 24v wires I should be able to safely assume 24v at 20amps I'd believe. That should be more then enough power to run this charger: Icharger 306B

Correct?