Author Options:

Using a Benchtop PSU to heat wires? Answered

I modified a 300W ATX power supply that I got for $1 the other day to use as a benchtop power source, much the same way the people who have posted Instructables for this have done.  I want to do some experimenting heating up wires, but I don't want to harm the PSU.  I believe it has built-in short circuit protection, but I want to be safe anyways.  I think that if I put in a high-wattage resistor in parallel with the heating wire, that might save the PSU.  If I did this, what value resistor (ballpark range) would I need so that power would flow through the wire until it heated up too much (thus increasing resistance across it), causing the power to go through the resistor instead?


Uhm... There are a number of errors here, starting with the assumption that the wire's resistance increases with heat (it's more likely to go the other direction), followed by the assumption that this would cause power to flow through the resistor instead (rather than "as well"). Two resistances in parallel always yields a resistance that is lower than either of them. The formula is (R1*R2)/(R1+R2).

Huh...I always had learned that heat increases resistance (which in turn produces more heat). As for the change in current flow, I was under the impression that electricity always takes the path of least resistance, although I understand what you are saying.

not always. some devices have a negative coefficient. Wire, however, does I believe play by the rules", although the effect is mild unless significantly overloaded.

Correction: Most conductors and common resistors do, indeed, have a positive temperature coefficient. But opposite of what has been suggested in this thread, this means that as temp increases, resistance increases. But these guys are right about your idea. Your suggested scheme with the parallel wire would not protect the PSU from overload. It would do the opposite, actually.

Hmm...That *is what I was trying to suggest, perhaps not worded correctly.

Ahhh. I see what you mean, now. :)

eh...I am not a master of words at times and can be confusingly obtuse on occasion. no bad. We're on the same page.

Just to make sure you're thinking the correct names: Higher heat = LOWER resistance = MORE current = MORE heat. A 1 meg ohm resistor will be barely noticable for heat because it has such a high resistance - a .1 ohm resistor will explode because it allows too much current. A short circuit 'usually' (always) gets hot enough to let the magic smoke out because the resistance is so low. It's easy to confuse the power disspated by a resistor with the actual resistance of the ...resistor.

Safety (protecting you and the psu)
Seconded seandogues idea of a suitable slo-blow fuse in SERIES with the hot wire. Use ohms law to calculate the required fuse. Measure the ohm resistance of a length of your hot-wire, and use ohm's law to calculate the needed amperage at a given (known, probably 12) voltage.

example:  measure your wire is 25 ohms.

12 = I * 25
I = 12/25
I = 0.48amps.  a 12v .5amp slow blow (maybe .6 or .7 amp) fuse would probably be better, blow less often.

Your heat-wire (probably nichrome?) itself will regulate the current output by its relatively stable resistance as the temperature changes. 

a filtered PWM circuit can be made to change the average current to attain different temperatures.  There are lots of schematics and readings on this website.

I would simply use a current limiting circuit to regulate the maximum output, and follow it (actually put these next items before) with a slo-blo fuse or circuit-breaker

Adjustable current regulation circuits can be made very easily using little more than a cheap LM317and a few extra components. Circuits are available via the LM317 datasheet or literally hundred of websites via a quick google search on the term "current regulator" or "current regulator lm317"

I say adjustable to clue you to the idea that adjusting your output to suit a particular hotwire might be a good thing, yes?