Converting Computer Power Supply to Variable Supply

depending on the power supply, you can trick it by a variable resistor in the reference feedback circuit - but if you try this with one of the newer supplies it will be much harder. look for an older generation computer, or if possible a linear not switchmode power supply.
Or, alternately, use a variable resistor on the output

I have a variable power supply to supply my soldering iron. Your hotwire saw is electrically like a soldering iron, right? Can you buy a power supply for soldering irons (inexpensive and plentiful).
i

It would help if you posted approximately what voltage and current your hotwire requires.
A straight PWM circuit would be simple and quite inexpensive, 5 to 10 bux worth of parts if bought new. This could handle up to 50 amps with no problem.
Depending on current and voltage, a "buck converter" is only slightly more complex and has the property of magnifying current to the extent that load voltage is less than supply voltage. It acts essentially like a variable DC transformer. The trick there would be finding (or making) an inductor that can hack the load current without saturating.

Nope. But I used to cut foam using a basic arc welder as the power supply. I would attach the welding leads to the ends of the cutter wire. I would start at the lowest setting on the front of the welder and increase it until I got the right "feel" for the cutting process. Too high and the foam would melt into a blob. Too low and the foam wouldn't "cut" smoothly or easily.
Gary

Don, THANKS! I know there was something out there that worked like this, just couldn't remember the term.
I need a power supply for that old 60Volt motor I posted about and can't find a good 60ish volt transformer to work with.

Marc,
Even though a "bucker" will certianly work.. IMHO a good ole Variac and some diodes from hell are they way to go... I have an old computer tape drive spindle motor mounted on my BP.. Some people think it's rated at 60VDC, some think 90VDC (seriously I've see EXACTLY the same motor rated at both , I mean same fricken part number)... I have it on a Varaic.. FWIW one of the neat things about these old computer tape drive motors is that hey were meant for continious hard duty and were meant to be air cooled (they used the output of the vaccume system for the tape drive to blow into the motor)... and yeah I managed to get the first stage vaccume assy (basiccaly just a fan in a shroud) from the tape drive and I air cool it... I plow away on the BP all day doing SERIOUS metal removal and the motor is cool as a cucumber...
Checkout this url for some pictures of the Variac and Diodes from hell.. :-) and although it shows an import lathe (my OLD lathe) I have 3 of these motors and ..duh .. stuck one on the BP... You'll also see a smaller version that I used to power a gearmotor as a power feed on my old import mill/drill..... and yeah these are kinda old.. since I now have a BP and a Clausing :-)....
--.- Dave

I am not sure why the OP would need diodes for his hot wire application. Just a variac should do it.
i

Iggy,
this is not for the OP... this is for Marc who has a 60VDC motor.. Folow the thread back :-)
--.- Dave

The PC supply will need a "bias load" on it to put out power. With a PWM on the +5 output, it will shut down with no load on the +12. A 250 watt AT (not ATX) power supply will ive 30 Amps at 5 volts (150 watts) or 11 amps at 12 volts (132 watts).
If you can find a 350 or 400 watt AT supply your output is almost double. AT supply has a remote switch, while ATX is switched by the mother board (more difficult to control) You COULD use a saltwater rheostat on the output, but a PWM would be a lot better.

I'm sure you can do it, but I don't see ANY reason to use DC. All you need is line isolation and safety precautions if you have much over 60 volts or so of either flavor. As others have said, it would help if you told us how much power you need; voltage range, current range. And maybe what the cold resistance resistance of the installed cutting wire is.
Pete Stanaitis ------------------------------
greggspen wrote:

I don't know. I have a 20 lbs e-core that I could wind as an inductor, a PIC (which I have a bunch of) could provide the PWM for a FET and a big diode later I have a supply.
Regulating the PWM should move the voltage around. I have a fair number of electronics parts, this seems like a cheaper way of going for now.

I think the issue is what kind of 'computer power supply' is it. It could be a large linear frame, large Switcher capable of hundreds of amps, A general linear that is highly filtered ? The number of letters in 'computer power supply' 2^19 variations or 2**19 variations.
If a switcher - you can't use a variable transformer. Output regulation only. If an Analog - then an input autotransformer 'Variac' (trade name) is most useful.
Remember to draw out your circuit - know the hot resistance of your wire so you can calculate the output regulator if it is a switcher. High current means a high wattage tap. I suggest knowing the most and the least resistance and steps - experiment in other words with the needed HIGH POWER resistors.
A switcher or an analog supply can be done this way - :
Once you find a good operating set - then make a set of two or three resistors that you can tap into. In other words, switches won't work unless you happen to a very high current switch.
Something like this :
+ ----/\/\/\-O-/\/\/\/-O-/\/\/\-O O~~~~~~~~~~~~~~~~------- -
The tilde or " ~ " equates to be the existing Nichrome wire or what you use. The " /\/\/\ " are resistors - maybe 100 watt 5 ohm or maybe more. The " O " are points where you bolt your heater wire (or a wire attached) to the various points of the resistor tap.
The + and - are the supply points.
Remember work the numbers with the hot resistance value (voltage divided by current) of your heater wire.
Martin
Martin H. Eastburn @ home at Lions' Lair with our computer lionslair at consolidated dot net TSRA, Life; NRA LOH & Endowment Member, Golden Eagle, Patriot"s Medal. NRA Second Amendment Task Force Charter Founder IHMSA and NRA Metallic Silhouette maker & member.
greggspen wrote:

Maybe it's been suggested but a suitable 240-120 transformer could be used for 120-60. Or just a 240/120 primary winding if the current rating is good enough. A variable auto-transformer, such as a Variac, could also be a possibility if you have one. Also, two 24V and one 12V transformer with the secondaries in series.
Don Young

According to Martin H. Eastburn :
[ ... ]
But you also need to be capable of supplying enough current to handle the cold resistance, which will draw more current.
Enjoy, DoN.

12:10 pm, D> I've thought of it, the issue is that the draws going to hit about 1.2KVA at the max. Thats a big transformer. I think I can do a buck converter with mainly parts on hand. Maybe have to order a FET and driver, but thats going to be like 3 bucks vs over a hundred.

I've got more PIC's than Carter has Liver pill's... Including a complete PWM rig with current foldback and LCD display I did for a client to control high power CO2 lasers...
I STILL hung the BFV on Diodes from hell.. :-)
What ever spins yer crank....
--.- Dave

Well, it'll be a nice learning experience creating the buck converter. If its not sufficient other routes will also be educational.

Dave, see Karl Townsend's post about AM radio noise. Could you look at the PIC source code for the sensor and determine the duration of the "read initiate" pulse to the sensor? The source code can be found at I'd like to determine the viability of putting an AM broadcastband RF trap in the sensor line.
Don W0LAP

Go to your local cable tv company - they use 60 v power supplies on the telephone poles to power the cable plant. Probably get a used one for nothing. Ken.