Load resistors

[liquibyte]

If I have 30V @ 3A running through a 10 ohm resistor, what would be considered the minimum wattage of the resistor to be safe and not burn it up?  I know that you're going to say 90W but that would get super hot super quick.  Mineral oil can perhaps?

[Legion]

200W - 300W

[Galaxyrise]

I think you'll need more thermal management than just a resistor power rating.  You have 90W of heat to move off that resistor, whether the resistor is rated for 100W or 1000W.  If you look at the power derating curve on a resistor's datasheet, you'll see the rated power drops as temperature rises past a certain amount, like 70C.  So if you can't cool your resistor at all, even a 1000W resistor may go up in smoke.

[liquibyte]

That's what I thought.  I ended up making my own out of an old stove burner.  Gets up to around 270F but it works, I'm getting 3A out of it.  I actually only cut one down to 10 ohms, I'm going to make a second one tomorrow and load both sides and see how things go.  I figure a stove eye is rated for high temps and I have no desire to plop down good $$$ for high wattage resistors and the accompanying heatsinks and fans to try and load this thing.

[Rudane]

You could always do what the power guys do, put a fan on the load. It seems weird, but I've seen industrial fans mounted to the outside of power transformers for airflow reasons.

[Phaedrus]

I built an 86 watt load recently for power meter accuracy check / calibration. Three 500 ohm resistors with 120V going through them.

To cool it I mounted them on a solid copper heatsink meant for server CPUs. I initially tried to solder the resistor casings to the heatsink, but it didn't work. I would have screwed them down, but I didn't have a good drill bit for making the holes. So I applied thermal grease, wrapped the resistors down with wire, which I soldered to the screw mounts on the corners. I attached the mains wires, applied heatshrink and electrical tape liberally for safety, then added an 80mm, 2800RPM fan (powered off a 12V boost regulator from USB).

I don't have a picture of the finished product on hand (and it's a bit of a mess to be honest, though it does give an exact 166.65 ohms across a wide temperature range), but I do have a picture of the heatsink:

Despite the heatsink and fan, the resistors can still reach 70C after just 10-15 minutes of running and it continues to increase logarithmically the longer you run it.

[liquibyte]

I think there's a link on here somewhere to a site where a guy used wire in a bucket as a load which got me thinking about how to load this PS down.  We recently got a new stove so I kept the old eyes as a backup and was thinking that basically they're just big assed power resistors so why not try them as a load.

The resistance wire in the center doesn't take solder well but I managed to get it to stick "good enough" for my purposes.  I've actually got another three of these wired in parallel and it gets to 11 ohms but I really wanted to load this thing at a full 3A so I figured I'd take the fourth and cut it up and see if I could get it to come out to 10.  It took me 2 tries to get the right length for the cut on the eye to get my 10 ohms but it worked out in the end to right at 10 ohms.  Heats up like a champ but I've got it sitting on an old metal case that has nothing in it so the heat's not much of an issue.

This has me now thinking about getting a spool of resistor wire and try my hand at a truly precise version of this wrapped on some sort of ceramic rod and maybe mounted on a stand.  I figure if I can do it this way, I'd be able to get any resistance I wanted.

[sleemanj]

I think there's a link on here somewhere to a site where a guy used wire in a bucket as a load

That was mikeselectricstuff

[liquibyte]

That was mikeselectricstuff

Yeppers, that's the one.  Good inspiration there, I'd never have thought of trying with a stove eye otherwise.  I don't know why watching that took my mind to those but it did.  I still think I'm going to give nichrome a try and see what I can come up with.  Should make for some interesting experiments anyway.

I know the radio guys have probably known about this kind of thing for years, but I found this interesting as well http://www.k4eaa.com/dummy.html

[SeanB]

Just remember the temperature coefficient of that stove wire means at 70C your resistance will be higher than at room temperature, and the hotter it gets the higher the resistance goes. On the connecting front the best is to weld the wire, or to crimp with nickel crimp sleeves for the stove element wire.

[liquibyte]

Just remember the temperature coefficient of that stove wire means at 70C your resistance will be higher than at room temperature, and the hotter it gets the higher the resistance goes. On the connecting front the best is to weld the wire, or to crimp with nickel crimp sleeves for the stove element wire.

I want to get this right so I looked up how to calculate this and ran across this calculator.  According to this a temperature rise from 25C to 150C will result in a rise of 0.48 ohms resistance in nichrome from 9.6 ohms to 10.08 ohms with a rise of 125C.

0.0004 * 125 * 9.6 = 0.48

So I should start with a lower resistance as it will increase almost a half an ohm as it hits around 150C or perhaps aggressively cool it to maintain a lower temp at the rated?

This is an area I've never gotten into before.  Makes me wonder about the 0.47 ohm 16W current sense resistors I've got in the case.  I've checked their temp as the PS gets loaded down and they rise up to around 96-97C when fully loaded at 30V and 3A.  I don't have them actively cooled yet but planned on it using an LM35 setup once I find one that's agreeable to my sensibilities i.e. simple and doesn't require an avr or pic.

Edit: board's formatting my greek letters to garbage and I can't figure out how to make it show up right, see the link for the equation.