Are esk8 calculators wrong?

I’m a bit confused, i’m looking to find out whether i should get a 36, 38 or 40T wheel pulley with a 15T motor pulley.

I’ve used this calculator: http://calc.esk8.it/#{“batt-type-lipo”:0,“batt-cells”:10,“motor-kv”:190,“system-efficiency”:85,“motor-pulley-teeth”:15,“wheel-pulley-teeth”:38,“wheel-size”:97}|

But i’ve noticed it calcuates the rpm with Nominal Voltage * motorKV

The nominal voltage of a li-ion cell being 3.6 V

But, a fully charged cell is 4.2 V which would mean my topspeed on a fully charged battery is a lot higher than what this calculator tells me right?

36V 37 km/h topspeed(weighted)

43V 44 km/h topspeed(weighted) (closest i could get to 42V)

I feel like the calculator is misleading, if i can actually go 7 km/h faster than what it tells me, given that my battery is fully charged?

I do know that voltage sac is not taken into account, but i dont know by how much, and my point should still stand.

It uses nominal as a good indicator. Of course you will go faster in full charge but I wouldn’t be worried. Nominal voltage will be closest to your average voltage throughout the cycle. Not to mention voltage sag

1 Like

All my boards go slightly slower than the calcuator predicts even at full charge.

I wonder why big boy? Hahahah lol JK I couldn’t resist I’m an ass I know

4 Likes

Check out discharge graphs for pretty much any cell. You’ll see that with even smaller current draw, the cell instantly sags a fair bit lower than 4.2v/cell. Nominal is a better measure of speed even at full charge.

1 Like

It also doesn’t take into account that the maximum duty cycle for the VESC is 0.95.

2 Likes

They’re not wrong. They are a simple guess. And they are a pretty good one at that.

The efficiency parameter is just a fudge factor. We use these kinds of scaling factors in engineering all the time. Is it 100% accurate? Absolutely not. But it works

When I was designing my board, I derived my own calculator similar to what you see online. However I mistakenly used full 4.2V voltage and my calculations were way off. Using nominal voltage with a 80% efficiency provides a much better estimate. You can go ahead and use 4.2V, but you will be wrong, just like I was :smile:

1 Like

U got me there hahaha

3 Likes