Simple, cheap and effective "load" for a battery tester.

[swbluto]

One of the major problems with typical loads that we use for battery tests is that it generates a lot of heat and this can seriously be dangerous when you have red-hot coils inches away from wood or other flammable substances. So, what I did is that I extracted some nichrome wire from my heater and then submerged it under water and attached to it by adjustable alligator clips(Adjustable resistance!) - I measured the current using a home-made amp meter(Good for upto 100 amps or so; it's actually just my e-scooter's current control circuitry. 8) ) and it worked pretty well. The wire didn't get red hot since the water serves as a very effective coolant and it has a high heat capacity so it'd take a long time to heat(The more water, the longer it'll take to heat to a high temperature) and the wire was pretty thin so it extracted the heat pretty quickly. One thing about my testing, though, is that apparently the resistance was increasing as the wire heated as the current decreased from 15 amps to 13.5 amps after 5 seconds - So, I suspect it really isn't nichrome. EDIT: Actually, I found out that the temperature of the BMS's mosfets very quickly rise to a very high temperature so I suspect their ON-resistance was increasing during discharge.

Anyways, this is just a quick, simple and cheap method of building an adjustable current tester that is much less dangerous than in-the-air nichrome - the only "more expensive" thing that one might need to source is a DC ammeter to measure the currents you intend to draw.

 

[BatteryMooch]

Nicely done! Thanks for posting your results.
Hmm...I'm thinking that if the wire is coiled small enough, I could heat the water for my coffee and test cells at the same time!

 

[monster]

you could use oil which is no conductive, but if water works...

 

[dogman dan]

There used to be milk boilers like that in the baby section of stores. I guess everybody has a microwave now.

 

[Ypedal]

http://ypedal.com/Lbd.htm

:wink:

 

[swbluto]

Ahem. Clean water is not conductive(Or not that conductive - my local tap water has a resistance of around 100kOhm about 1 cm apart - The load is going to have a current that's 1000+ times higher, so it's rather trivial.). Furthermore, oil has a low heat capacity so it heats up really quickly and it's much more expensive than good old tap water.

And, lightbulbs are good and everything, but they have a finite lifetime, they heat up the testing environment relatively quickly(assuming it's indoors), it creates a lot of light(Could be a good or bad thing but I've heard someone complain about it on here - something to do with sleeping), requires a trip to the store, quite a few parts and bulbs to get reasonable currents and some assembly and the resistance, hence, current isn't as adjustable(Or, rather, it's adjustable along given intervals which might be good enough if you make sure to buy a few "current tweaking" bulbs to fill in the gaps between the main work horses.) or comparable levels of adjustability takes a little more work.

But, I'll admit, this method does have its cons. You can't change the resistance value during the middle of the test - that could be rather dangerous! For safety, I'd disconnect the battery, wait a few seconds, and then change it. But for constant resistance and "approximately constant current" loads, it works pretty well. Also, I don't know what kind of affect water has on nichrome or the alligator clips in the long term(Dry them immediately after each use would be my suggestion), but a worst case scenario projection would be it destroys them within 1000 or testing cycles. Replacing the alligator clip or changing the nichrome wire? That isn't so hard.

 

[swbluto]

Just to make a suggestion, if anyone does this, you probably don't want to make your alligator clips too close to each to reduce the resistance. It might be easier to just bunch multiple wires together so they parallel together to reduce the resistance - I say this because my load short-circuited when I placed the alligator clips about an inch from each other and underwater fireworks ensued.

 

[ZapPat]

I just tried the "nichrome in water" load, and think this is a very good solution for high power tests. One major advantage is that the reistance doesn't go up as the wire heats, which is an annoyance when using the wire in air. And the water multiplies by many times the continous power that can be dissipated by the same wire in air. It's also suprising to see how slowly the water heats up...

I would hope that the Nickel and chromium content of the wire will help it resist corrosion. Low salts content of the water would also help much for avoiding corrosion. I bet being red-hot in the air is worse for it yet than being cool in water, and these things are made to be red-hot so they must be pretty resistant to oxydation.

Now I just have to finish fiddling with the PID current loop algorithm in my PIC to turn this resistive load into a constant current load...

 

[swbluto]

I haven't checked your progress in the other thread, but are you planning on doing current control through switching means? I'd imagine that help to make sure that, well... nothing gets hot. :wink: (Which does sound counter-intuitive with all our experiences with "high power" and heat.)

Oh, it's best to place the wire near the bottom of the water, as the warm water naturally rises to the top. It's kind of fun, too, when I was discharging 15 amps * 50 volts = 750 watts into the water and I could see the water around the wire kind of "squiggling" and it was rising upwards. At about 1500 watts or so, I even saw bubbles instantaneously forming where it contacted with the water! But I was only directing the current along a short 4 inch strand of nichrome wire so the heat/distance ratio was pretty high. With enough wire, you shouldn't see bubbles all though that might be fun to watch!

At wikipedia, http://en.wikipedia.org/wiki/Thermal_co ... ity_values , it appears water's thermal conductivity of .6 is 24 times higher than air's thermal conductivity of .025. Like fan cooling, I can imagine you'd get better greater heat dissipation if the water was moving(Like having a water churner of some sort?), so I can imagine you can put some crazy loads on nichrome wire pretty cheaply.

 

[liveforphysics]

I like this idea a lot, and I just realized something.

You can get a pretty darn accurate idea of the quanity of energy discharged from the cell if you perform this test in a well insulated water vessel, like a large plastic thermos, and correctly measure the quanity of water to begin with, measure the starting temp and the ending temp, find the deltaT (temp difference), and use the specific heat value of water (1 calorie/gram °C = 4.186 joule/gram °C), to calculate the energy that was added to the system. As long as it's well insulated, and the water doesn't reach boiling point, you could accurately measure the quanity of energy discharged from the cell.

You could make a compact and super cheap 100-10,000+w cell energy tester in this way.

 

[ZapPat]

I found something much better than nichrome (heating element wire) to use as a cheap and powerfull load. It's easy to find, very cheap, flexible and rusts much less than nichrome in water!

And it came from the dollar store! One very long roll of multi-purpose wire for 1$. It has a resistance of 0.25ohms per meter, and looks to be made from some tin alloy or similar. I first thought it was aluminum, but then I tried soldering and it took the solder no problem!

Anyways, I took a piece of wood and put a few screws in it, and then wrapped a 4 meter length of wire around the assembly. I doubled up the wire as to get about a 0.25ohm load. I used some 14 gauge silicone (or sil icone as the Chinese wire is marked - to avoid copyrights maybe?!) that is soldered to each end of the load wire.

View attachment DollarStore resistive load-on-a-stick.jpg

So I can now put my "load on a stick" in a bucket of water, and I have a 1000W+ load to play with! Make sure you immerse even the soldered silicone wire ends in the water or they'll be your weak point. Right now it's my 14gauge wires that are the weak point, since they become pretty warm at the 30amps I was testing this thing at so far.

swbluto: Yes I am using a switcher and my 48V20Ah PingV2 for these tests. The switcher is actually my prototype controller, of which I'm only using one phase for the moment (and cheapo ir3307's for now). I was using 80kHz as a switching frequency, but will test soon using 20kHz to see how much less heat will be generated. I've calculated that the switching losses are actually just over double the conduction losses right now at 80kHz, so it should run quite a bit cooler at 20kHz (but more ripple). BTW, I also have a 330uH coil at the controller's output (before the wire load), but no caps yet. I am aiming to turn this into a three phase rig soon which would let me test at pretty high currents with no additionnal ripple. What I find really cool with this is that a brushless controller can potentially become many tools with the right firmware, and sometimes a bit of additionnal hardware (output filter, resitive load, etc...). I've so far thought of: Brushed motor controller (high efficiency three phase), welder, battery/cell discharge tester. :lol:

 

[BatteryMooch]

Great news ZapPat! Thanks for the update. I'm looking forward to hearing more about the controller as it evolves.