Load test

[Norway]

Hi. As I have a problem powering up EggRider display on my BBSHD, I tested my battery using a light bulb as load. 56,5V on a 75W light bulb. According to my BMS the battery drew 13w. Is it ok? Or too low? Can’t figure out why it doesn’t turn on.

[AZguy]

75W@120V gives resistance of 192Ω

13W@56.5V gives resistance of 245Ω

Resistance in bulb filaments goes up with temperature but the error here is in the low 20%'s so while not jiving completely it is in the realm of "close enough" based on the measurement techniques - for one thing, I suspect most BMS's aren't accurate...

[Norway]

Thank you for your reply . As I live in Norway we have a system with 230v. With your setup that mean the bulb has ca. 700Ω. Guess it means I have a problem with my battery. Any thoughts on what it can be?

[JPLabs]

Can't calculate current the same for AC and DC, though, right? The resistance calculation for AC is not simple constant current, it's sinusoidal, so gotta do some extra math. It's the hard way. Then you have induction effects, too, right?

Need to use accurate bulb resistance to make sense of the battery load test.

Just measure the DC resistance of the bulb with your meter, and use that value to check the battery current delivery! Keep your math and measurements in the DC realm, which is nice and simple.

[Norway]

I measured the resistance in the light bulb and it was only 49Ω. That means I should have more watts?

[Tommycat]

Resistance in bulb filaments goes up with temperature

That being said, why not try to power the display and system with your charger as a power source to verify correct wiring/operation. Most are adequately protected from shorts, but I wouldn't discourage you from installing an in-line fuse. (just in case) And should provide enough power for the display, and perhaps turn over the motor slowly if desired.

[Norway]

I’ll have to try that when I’m back from business trip

[AZguy]

Can't calculate current the same for AC and DC, though, right? The resistance calculation for AC is not simple constant current, it's sinusoidal, so gotta do some extra math. It's the hard way. Then you have induction effects, too, right?

Need to use accurate bulb resistance to make sense of the battery load test.

For purely resistive single-phase loads DCV = ACVrms... more phases it goes out the window

Just measure the DC resistance of the bulb with your meter, and use that value to check the battery current delivery! Keep your math and measurements in the DC realm, which is nice and simple.

The resistance a bulb shows when lit will be *much* higher than when "cold"... the filaments are silly hot when lit - over a couple of thousand °K and so the resistance is similarly higher

That the bulb is 240V makes a lot more sense with it having about 770Ω hot (for 75W) - about 3x the much cooler 56.5V temperature resistance (~245Ω), and to me at least , that makes the numbers jive close enough that I'd bless the results!

Certainly makes a lot more sens than the calculations at 120V - funny how missing information can screw that up!

[JPLabs]

For purely resistive single-phase loads DCV = ACVrms... more phases it goes out the window

The resistance a bulb shows when lit will be *much* higher than when "cold"... the filaments are silly hot when lit - over a couple of thousand °K and so the resistance is similarly higher

That the bulb is 240V makes a lot more sense with it having about 770Ω hot (for 75W) - about 3x the much cooler 56.5V temperature resistance (~245Ω), and to me at least , that makes the numbers jive close enough that I'd bless the results!

Certainly makes a lot more sens than the calculations at 120V - funny how missing information can screw that up!

I was hoping you'd check in, thanks for the info, for all of us, as always.

[paxtana]

Hey, you guys ever seen a brake resistor burn up from the sort of voltages an ebike battery would put it through?

The 1000w brake resistor I usually use for load testing made a small popping sound few days ago,, then some smoke started coming out of the exit points for the fiberglass insulated wiring on what is an otherwise completely sealed aluminum block. My colleague said it was fine since it was still working but I'm not so sure.

[AZguy]

I don't know the specifics of the resistor you have - if you have data throw it my way... even just measuring the resistance with a DVM would help

It's likely what they call a wirewound resistor and is basically a big heating element and made to get pretty darn hot... if they get too hot, unless it's so way over the top usually something aside from the element overheats and melts... I've even seen them get hot enough to melt the solder joints and then once everything cools down they work again although my trust in them sort of decreases =]

Does it have a fan?

[paxtana]

It is this one, advertised as 4 ohm. No fan.

When I measure it with a multimeter it reads 4.2.

If resistance would change on a light bulb when hot, is that the case with something like a brake resistor as well?

[AZguy]

It is this one, advertised as 4 ohm. No fan.

When I measure it with a multimeter it reads 4.2.

If resistance would change on a light bulb when hot, is that the case with something like a brake resistor as well?

Yes but the bulb filament runs about 2500K where even if your resistor is running at 100°C (~373K) it's still a long way from a bulb filament... The resistance change is close to the ratio of the temperature in Kelvin (K):


Let's say the resistance at 25°C (~298K) is R₂₅

For the load running at 373K you'd expect the R_LOAD(hot) ≈ 373 / 298 × R_25C ≈ 1.25 × R₂₅ so about a 25% increase over the room temp load

For the filament running at 2500K R_BULB(hot) ≈ 2500 / 298 × R₂₅ ≈ 8.39 × R₂₅ so >8× over room temp

Very different beasts for sure



As for your load and it's power handling:


4Ω @ 50V would give us P = 50V² / 4Ω ≈ 625W (~12.5A)

However at the top of a 52V (14s) pack 58.8V² / 4Ω ≈ 864W (~14.7A)

And at the top of a 20s pack 84V² / 4Ω ≈ 1764W (~21A)

Point being that the power going into the load goes up by the square of the voltage so you can get to the point of dumping a lot of power into it and even a 52V pack near the top would make it very hot


A couple of suggestions... the resistance seems very reasonable for a hard discharge... For 48V or 52V packs it's pushing it a bit on power but then again these are robust devices that can take some abuse although I'd mount a small fan to the unit and blow air across the little fins on the top - that will help more than you might think

Better yet (and if you're going to do 20s packs for sure) I'd put four of them in a 2p2s arrangement - this will keep the resistance at 4Ω but quadruple the power handling... for 13s/14s packs it would only get warm and with 20s it would get hot but not even close to too hot

Let me know how it goes

[paxtana]

Thanks for the insight, I will give the fan a try :)