6v sag too much?

If your cutoff start is set to 3.0V and the cutoff end to 2.8V then the VESC will start to reduce power linear ones it reaches 3.0V.

So if you give 100% throttle and your motor max is set to 60A it will be behave like this. 100% throttle at 3.0V single cell voltage = 60A 100% throttle at 2.95V single cell voltage = 45A 100% throttle at 2.9V single cell voltage = 30A 100% throttle at 2.85 single cell voltage = 15A 100% throttle at 2.8V single cell voltage = 0A

And the VESC is extremely fast to redice the power. So when your cutoff end is at 2.8V it will very unlikely be possible to go below that. And 2.8V a cell is still relatively conservative so that you don’t need to worry when you feel that the VESC reduces the power. You can even drain until the board barely moves and still be on the safe side. But it is of course important that you use other cutoff start and end values for LiPos (3.5, 3.3).


This is exceedingly helpful, thank you.

I’ll send you some beer money for all the work you’ve put into this, and your continuing support here.


I did notice that a vesc will still run the motor at reduced speed beyond cutoff end, but this is unloaded. It’sd probably fail to drive anything loaded.

I’m around 200lb, I get 20-23wh on my commutes (very spirited). This is on 12s2p A123 with a variety of stuff. Carvon v2.5 dual, banggood 5065 single, jlabs 6355 dual. Amazingly consistently around 20wh/mile.

What’s everyone’s weight @Jinra @Titoxd10001 @jmasta?

Are the lipo guys relatively light? Or does lipo really delivery more efficient wh/mile? A123 have very good voltage sag characteristics, but not as good as lipo, I think.

And I’ve found it’s really hard to squeeze out the last few % out of the pack. The voltage sags crazy when the pack is near depleted. But A123 kind of have weird behavior.

1 Like

The problem with this theory is that the vesc doesn’t monitor single cells, it only monitors the battery as a whole. So trusting the vesc to fully protect your battery is false security because if your cells become too unbalanced, you could wind up over discharging cells before the vesc backs off. A BMS on the other hand, can provide full protection by monitoring individual cells or cell groups.

I have the Metr app and module. Here is a record of some riding I did today with some long gradual hills and some short steep hills. Numbers recorded from just 1 of 2 Vescs. Motors are 6374 200kv gearing is 15/40 with 90mm wheels

1 Like

Because of that I recommend to use 2.8V as cutoff end. Gives you some wiggle room. But the cell drift on li-ion cells is very little. At LiPo cells it is much worse.

Watt hours per mile is the overall efficiency of the entire system (and driver). I’d argue battery chemistry is somewhat irrelevant to this value. It’s simply how much power did you draw, for how long, and far did you travel? The VESC only knows how much power it consumed, not where it got it from

Wow, after reading this thread I’m concerned about my next build with 10s4p VTC5. I didn’t know that the voltage sag is so extreme with Li-Ion :fearful:. On my mountainboard with top mounted voltage meter and 10s1p 10Ah Turnigy Graphene Lipo the worst voltage sag I’ve ever seen was about 0.6V :grin:

1 Like

Same here, started from Li Po’s - I’m very happy with them and I wanted to move into glorified Li-ons. After reading a lot I’m considering staying with Li Pos for longer :slight_smile:


Actually I think different chemistries waste differing amouts of energy under heavy loads. I think its related to internal resistance and voltage sag.

Personally it seems very apparent to me when my pack is nearing empty. I bet I could pull out the last 20 wh at .5C loads, but can’t do it going uphill on my board.

Your pulling very little amps. If my math is right your at 12.87 wh/mi.

I pull 23 wh/mi average on flat land. I think how much power you consume has a lot to do with rider weight, riding style and board setup. I’m over 200lbs and hit 30mph on my normal route. I estimate if I go slow I can get 15 wh/mi but once I go fast it can double the wh consumption.

So if your lipo pack is 185wh/12.87=14.37mi range. Which is close to the range you said

If I were to ride my board with your battery I would in theory get 185/23=8.04mi range

Granted with a 18650 pack you can’t get the full wh because of voltage sag.

Agreed, going 30 mph will create a lot of wind resistance that can greatly reduce mileage even more so than weight. I’m currently 195 lbs which isn’t far from your weight. However, I did not have much wind resistance riding at moderate speeds on a relatively calm day.

Under normal conditions the cell drift may be minor but in case something goes wrong like welds failing and separating cells in a cell group which could cause that cell group to discharge much faster than other groups. There have been reports of weld failures just recently. This is just one reason why it is so much better to use a BMS and not bypass it during discharge so that the BMS can’t shut the system down if such a situation occurs.

Vesc as primary voltage control BMS as emergency backup control

There is a much higher chance the BMS fails than that the battery fails. And when you have it included in the discharge cycle it can shut the system down during braking. And there were already many people who had issues with their BMS who did exactly that. Also people who used the BMS you sell.

So if you ask me i think they should only be used for charging. But that’s personal preference.

1 Like

First allow me to make clear that I have no affiliation with Bestech and I am not in the business of selling bms’s. The few extras that I have sold, I sold only to help others and I sold them at my cost + shipping so my promoting of them is not motivated by monetary gain.

As far as them failing, I can only give my account which is that I have been using them for over a year now with no problems. Often times I have rode my brakes down steep hills with a full battery and the bms did not shut down but rather took control of the regen charging not allowing it to exceed the overcharge setting of the bms and once I stopped braking, the bms proceeded to trim the cells back down as it does during the balancing process.

Maybe I missed some posts but I have not heard of this particular bms failing lately although I have heard of the cheap bms’s from eBay failing.

Ah I think I see it now. Speed as you say. Your top speed / avg speed are 26kph,12kph. In our backwards miles units that is 16mph,7.5mph.

I’m running 29mph top speed 13-14mph avg. Sounds like totoxd10001 is in the same ballpark.

I should try a slower ride to see if the number shift dramatically. But if feels safer in front of the cab then behind it…

Adding to the BMS conversation, my beef is that I don’t know what they’re doing. I’d like some kind of error indicator and a way to access individual cell voltage. I suppose we have 2 open source BMS projects going for that now.

With no BMS I’m forced to check regularly, but I feel like I’d need to do that anyway. I live in fear like lowguido and his distaste for solid state switches.

Yes going fast reduces range by a lot. It’s kind of funny that some claim you can get max efficiency by going close to top speed

so at full charge (49v roughly) max sag is about 4v, usually less than 3v. which makes sense, higher voltage, less amp draw.

so…i’ll just keep the pack charged up, this should mitigate the ridiculous sag issue.

a good bms can. A bad BMS is worse than no BMS. I try and charge to 41 volts using samsung 30q’s checking voltages manually from time to time. seems fine. edit - the bestech BMS shipped here is about 1/4 the costs of the cells. that’s a big factor why I don’t use them