Was just wondering what you guys though of like making a coil of say 12awg wire and sliding it down a pvc pipe and flow some water through it for a sort of water cooled low ohmic test resistor. I understand the temperature vs resistance stability will suck but I have a current meter that I can just test my PWM to see if it will handle near full rated current.

Only need to test to like 50,000 - 100,000 watts for a couple seconds.
I don't have the motor to try it with yet and don't really want to order it till I know I can drive it. I need to know if my driver chip can still make a decent square wave as the parallel mosfet current ramps up.

All I have now is a 250W 1ohm resistor I been using. Handles 1000watts for 5 seconds. I could water cool that and maybe do a quick 10,000 watt 1 second burst but I don't think that would be an adequate test.

SO is this idea crazy or no?

BTW I'm full of crazy ideas. Because I built a 3 pole variable timing electric motor just for fun:

 

Hello,


The idea is not so crazy, but water will conduct.
Oils are better for cooling.
In RF dummyloads there was transformer oil around the resistor.
In free air the resistor could handle 100 Watts.
Immersed in oil the resistor could handle 1000 Watts.

Greetings,
Bertus

 

That's not crazy; it's creative. Sometimes there is a fine line between the two.

The trick is to convince your spouse that you're dancing on the proper side of the line.

Thermal conduction will be your big obstacle. Get a great big ceramic wirewound resistor (or a bunch of 'em wired together) and suspend them in a decent sized water tank. The kind of resistors I'm talking about are wound on hollow ceramic tubes, as big around as your thumb, and are about a foot long. If you keep them with one end pointed upwards, when water heats (boils) it'll pull cooler water in the bottom.

Or, you might try using some stove or hot water heater elements from salvage. Lots of stuff you could use.

 

The trick is to convince your spouse that you're dancing on the proper side of the line.

Lol, you should see some of the looks my wife gives me when I show her some of my crazy things. Somewhere between puzzled and astonished with a side of, why did you do that when you can buy one?

Or, you might try using some stove or hot water heater elements from salvage. Lots of stuff you could use.

Omg, great idea! I have just enough time to use my wifes stove elements before she gets home

 

I think you missed the "from salvage" part. If you burn them up, you will be in a very dark, cold doghouse!

Better check the wattage rating of the elements before you try it.

 

As you've heard , not a crazy idea at all. You can use distilled water which doesn't conduct very well. I once used piano wire (steel) in distilled water to make a temporary high wattage resistor for test purposes. It didn't last very long, but it worked. However, my application was not at your desired power levels. This looks challenging, but has a good chance to work if you do it correctly (in my opinion). Mabey a long length of copper wire of the correct gauge will work. Keep in mind that this could have significant inductance which you can calculate. I'm not sure if that matters in your application.

Steve

 

I think you missed the "from salvage" part. If you burn them up, you will be in a very dark, cold doghouse!

Better check the wattage rating of the elements before you try it.

Hehe, well can't really blow them up anyway
They are 22.3ohms cold for the large burners. and they are designed to warm up. As they get hotter, the resistance goes up. So with my 120V supply I can't even get one of them to dissipate 1500 watts while submersed. I'd need lots of these in parallel. I'm starting to look around the house and noticing some things. Like 3000 watt wall heaters. 4000 watt water tank elements


Seems hard thinking about it. Maybe if I had some salvage heaters, I could cut the elements in like 4 chunks and wire those in parallel and submerse them to boost the current load.

 

Do you have a bunch of scrap angle iron? It has 10x the resistance of copper.
Stainless steel has about 53x the resistance of copper.
Nichrome wire has about 65x the resistance of copper.

 

Do you have a bunch of scrap angle iron? It has 10x the resistance of copper.
Stainless steel has about 53x the resistance of copper.
Nichrome wire has about 65x the resistance of copper.

Genius!

By george I think I do have some scrap iron around.

 

Ok, IRON has about 5.8x the resistance of copper. I was talking angle iron (made of steel) - because the kind you get nowadays is usually kind of thin.

 

Bailing wire is also made from mild steel, and is also kind of thin. It can also be kind of long. A five meter chunk of #14 should put you in the ballpark of 1 ohm. Or try a little over seven meters if your bailing wire is #12.

http://www.engineeringtoolbox.com/awg-wire-gauge-d_731.html
and
http://www.allmeasures.com/Formulae/static/formulae/electrical_resistivity/31.htm

 

I think his neighbor might be losing his chain-link fence soon...

 

I think his neighbor might be losing his chain-link fence soon...

Well I did want something the provided more privacy anyway

Been busy with a pile of University homework so I just now got around to increasing the test. Spend a few hrs modifying my bench supply and building a little panel interface with a current meter.

Got them up to 24v and 50 amps, only 1200 watts. Didn't even warm the plate they are mounted to (main heat sink not connected yet). Its funny the sound you can hear when you don't even have any motors connected. The frequency was somewhere in the 5k-10khz, almost sounded like it was the mosfets.
I'm gonna stab in the dark and assume this in normal for mosfets to sing at switching rates in the audible range. So if it is normal, sounds beautiful hearing them all work in parallel. If its not normal, well still sounds neat.

Can't really test my 120v yet as I don't feel safe attempting to bridge rectify my AC current and feeding that to the mosfets while I turn the pot on the low voltage circuit, as the ground would be shared. I don't happen to have any 1:1 transformers around either. I do have a 2x4 inch thick block of iron I could make a transformer out of. But chances are good that I don't have enough wire, will blow the fuse, or make lots of smoke.

I could maybe wire a few of my car batteries in series with my supply and maybe get 48v for another test. This would push me to around 10,000 watts with my current 0.25 ohm load.

 

Got them up to 24v and 50 amps, only 1200 watts. Didn't even warm the plate they are mounted to (main heat sink not connected yet). Its funny the sound you can hear when you don't even have any motors connected. The frequency was somewhere in the 5k-10khz, almost sounded like it was the mosfets.

Running the amount of power that you are, with essentially a square wave driving the MOSFET gates, you'll hear it if your drive frequency is less than around 20kHz. Even the wires will "sing".

I'm gonna stab in the dark and assume this in normal for mosfets to sing at switching rates in the audible range. So if it is normal, sounds beautiful hearing them all work in parallel. If its not normal, well still sounds neat.

Yep, it's normal. If you get your switching frequency above 20-25kHz, you won't hear it anymore. However, the faster you drive the thing, the more current is required from the driver circuit, and the more time the MOSFETS spend in the linear region. If it's not objectionable, I'd suggest leaving it at the lower speed.

Can't really test my 120v yet as I don't feel safe attempting to bridge rectify my AC current and feeding that to the mosfets while I turn the pot on the low voltage circuit, as the ground would be shared. I don't happen to have any 1:1 transformers around either. I do have a 2x4 inch thick block of iron I could make a transformer out of. But chances are good that I don't have enough wire, will blow the fuse, or make lots of smoke.

A transformer made from a block of iron will be terribly inefficient. Your eddy current losses will be horrific. Modern transformers are made from laminated stampings in "E" shapes from an iron alloy that is optimized for such use. All of the "E"s have the same porportions percentage-wise for minimal waste during stamping; they are scaled up or down depending upon the size of the transformer to be made. There are a number of sites on the Web for designing/winding your own transformers, but even going by all of the formulas, etc - the end result may still require "tweaking" for satisfactory results.

I could maybe wire a few of my car batteries in series with my supply and maybe get 48v for another test. This would push me to around 10,000 watts with my current 0.25 ohm load.

It's highly likely that you will either melt a terminal off of one or more batteries, or you may actually cause one or more batteries to explode due to the large release of hydrogen & oxygen that will occur under such load conditions. Lead has roughly 13x the resistance of copper, and the battery terminals and all of the plate interconnecting straps are made of lead. If one of these interconnecting straps fails during such a high load condition, it will make an arc which will ignite the HH & O mixture, causing a loud bang, bright flash, and your battery caps to enter low earth orbit.

I blew up a battery like that when I was a teenager; I'd connected up jumper cables backwards from Dad's Olds to a clunker I had. Blew the entire top off the battery, and I couldn't hear a thing for the next 10 minutes. Dad was NOT happy!

I suggest that you do not attempt to place such a heavy load on your batteries, unless you wish to risk a great deal of damage.

 

If it's not objectionable, I'd suggest leaving it at the lower speed.

I like hearing them, tells me the are working
But if I start running an electric motor, it may be too loud so I might increase the frequency up if my driver will handle it.

A transformer made from a block of iron will be terribly inefficient. Your eddy current losses will be horrific. Modern transformers are made from laminated stampings in "E" shapes from an iron alloy that is optimized for such use. All of the "E"s have the same porportions percentage-wise for minimal waste during stamping; they are scaled up or down depending upon the size of the transformer to be made. There are a number of sites on the Web for designing/winding your own transformers, but even going by all of the formulas, etc - the end result may still require "tweaking" for satisfactory results.

I know but it would be fun. The motor I made had laminated sheets to reduce eddy currents. I'll probably postpone the higher voltage test untill I finish the low voltage circuity and its power supply. I'm going to setup a zener diode and resistor to power the oscillator for driving a H-bridge mosfet at 400 or so HZ to drive a setup down transformer from the 120V DC supply. Then probably a voltage reg with filter caps. This should provide enough current for the gate drive circuit I have now.

I suggest that you do not attempt to place such a heavy load on your batteries, unless you wish to risk a great deal of damage.

The suggested current was 192 amps. This is well within lead acid batteries capability. My 66 Impala with high compression stroker motor uses over 300 amps while cranking for up to 10 seconds (when fuel emptied back to the tank).
Also, a bank of 10 parallel deep cycle batteries can supply 200-300 amps continuously for many minutes. This is the same as a single 12 volt being asked to deliver 200-300 amps since they are all in series anyway.

 

i've seen banding strap used as a load for testing car batteries. Maybe this would work for you in the right combination of lengths etc. My mistake i didnt see the second page and was replying to the first, talking about a test load.

 

Take a look at post #7 of this thread.

http://forum.allaboutcircuits.com/showthread.php?t=12372&highlight=load

You can buy water cooled resisitors, but as Bertus said, oil is better.

 

I will try some full load testing soon. I just finished up all the control circuity and power supplies including an current limiter circuit. Wanted to do this first so I could set a max current before dead shorting this thing through steel or otherwise .
Then I can slowly ramp up the tests.

 

You could also wire-wind around a thermally conductive tube and flow water through it. Then there is no problem with water conductivity. Is this AC or DC? With the bars of iron or whatever, eddy currents could be a problem.

 

Why not use copper water pipe? You could solder some lugs to the pipe and
it is real easy to run water through.

(* jcl *)