Buidling benchtop AC/DC powersupply

[JCobra14]

Well basically what i want to do is build a bench powersupply, with a variable AC/DC output, isolation transformer (for AC), Current and Voltage meters, all of this being self contained in some kind of case.

i've already built a AC powersupply pretty similiar to this, but i want to add a DC supply, i'd like to have shortcircuit protection too, preferably without using fuses.

I basically have 2 ideas going for building the DC powersupply, the first one is to use a seperate AC variac, and just rectify and filter the output to get a variable DC voltage.

The second, which i've already built a powersupply around, would use a stepdown transformer, and convert the output to DC. Would use a DC~DC regulator of some king to drop the voltage.

I've also thought about using the AC Variac, and then a silicon-based regulator of some sort to "Fine Tune" the output.

I'm really looking for advice on the short circuit protection part of this, i'd like to avoid fuses, simply for convenience sake.

 

[blahblah99]

How much current are you going to be drawing from this? If it is less than 1A, there are tons of good schematics on google.

 

[JCobra14]

practically i'm not going to be drawing more than 500 mA, but i'd like to have the capacity for say 5 A or higher,

I'll check out the stuff on google )

 

[CTho9305]

would a thermal fuse work? it should stop conducting at some temp, then when it cools down, start again? thats a guess of what a thermal fuse might do. I dont actually know. I just saw some at radio shack

anyway, any reason a computer power supply wouldn't work? just grab an old (AT) one and regulate any lines as necessary.

 

[JCobra14]

Well, i'd like something along the lines of 0-150 volt range, which would be a pretty hard mod of a computer powersupply,
i think i just want to use an AC variac, it'd be the simplest thing, and have a much higher output,

the thermal fuse sounds like a great idea, i've never heard of it before,

maybe i could do something around a hot/cold resistor as well, which is much like the thermal fuse you described

 

[StandardCell]

<< Well basically what i want to do is build a bench powersupply, with a variable AC/DC output, isolation transformer (for AC), Current and Voltage meters, all of this being self contained in some kind of case. <SNIP> I'm really looking for advice on the short circuit protection part of this, i'd like to avoid fuses, simply for convenience sake. >>



Well, I'll give you my two cents:
1. My dad was a bench tech for the last 40+ years, and from what I've seen from power supplies he's built, he uses pre-designed circuits from electronics magazines. I know it's partly a matter of pride (and at the time for me it was too), but building a variable PS is really tricky and you should stick with a tried-and-true formula. Not sticking to it will get you what happened to him in the old days - namely, blowing repair units to kingdom come because of poor circuitry. Popular Electronics or something else that you can get somewhere should have this kind of stuff.
2. If you want to build your own ANYWAY, why don't you try the alt.electronics.* Usenet Newsgroups and ask there?

Cheers...

 

[Demon-Xanth]

Look on the LM317 datasheets, iirc they actually have a schematic for a variable power supply w/ current limiting. (featuring the LM317 regulator)

 

[Mark R]

For the DC supply, look at some of the monolithic voltage regulators that are available. Most are rated at fairly low currents, but some are available with ratings of 5A (with adequate heatsinks), or they can be 'boosted' by adding an external transistor. Most of these feature thermal protection, short circuit and over current protection - in general they are more robust than any circuit that you could realistically build.

Something like the LM1084 would seem ideal for this.

 

[Superdoopercooper]

<< For the DC supply, look at some of the monolithic voltage regulators that are available. Most are rated at fairly low currents, but some are available with ratings of 5A (with adequate heatsinks), or they can be 'boosted' by adding an external transistor. Most of these feature thermal protection, short circuit and over current protection - in general they are more robust than any circuit that you could realistically build. Something like the LM1084 would seem ideal for this. >>



Ditto what Mark said. There really is no rocket science in an AC-to-DC converter. You need a full-wave bridge rectifier, an good regulator, and a couple of capacitors (one fairly large, and probably a couple of smaller ones - ~0.1 - 1.0 uF). Maybe a potentiometer or some pression voltage dividing op-amp to change the output voltage.

PM me if you need me to find you a design. Look at Maxim, TI/Burr Brown, Linear Technologies (probably the best one) websites.

 

[sitka]

Maybe a solution is
- for high amp DC I don't think you can't build cheaper than the Pyramid ps 36kx 12>15V 35AMP. Then kit additional controls for more specific output. Plop the Variac on top for AC.
Lots of options, little committment.

 

[JCobra14]

~mark, the IC you linked to looks promising,

i've built several supplies around the LM317 IC, so it looks like it pretty much the same, just heftier' current output

 

[jamarno]

A simple way to get any desired current limit is adding a second LM317 in series with another one, the second LM317 connected as a current regulator. www.national.com has information and schematics.

You can't rely upon fuses, themal or otherwise, to protect solid state equipment, including power supplies, because they're not nearly fast enough in most cases.