No, I didn’t test XLITE-24 with Metr. However, I suppose this has to do with the firmware version rather than with XLITE-12 vs XLITE-24.
1 Like
I don’t know how metr implemented this, but it might actually be related to using the newest VESC protocol for CAN bus by default instead of the old diebiems. This can be changed with the ebms app over USB, but you will loose VESC communication…
Maybe the easiest way to get around your issue would be to use VESC BMS protocol on metr if it is implemented. I guess it might eventually get supported if not already. XLITE by default is now sending the same CAN bus message as the VESC BMS does. (except the hardware name in the CAN message first packets)
2 Likes
@ENNOID Gonna order an XLITE-20-V2. Gonna order the balancing and temp cables too. Do the Thermistors come in a two pack or should I set quantity 2? Also, why’s the shipping $35? You not in the US?
I’m actually in Canada, but the shipping cost is flat rate 35$ for all around the world.
I want to improve that eventually with the Paypal button options, but haven’t been able so far. It would be much easier to set up I guess if I was using a real e-commerce platform instead of simple paypal button.
The flat rate is overpriced for US, break-even for Europe & Asia and quite cheap for Australia.
Please set the thermistor QTY to 1 and I usually ship 2 or more.
2 Likes
I’m probably in the minority here…but I don’t mind the shipping cost, because I get the package in like 2 days.
2 Likes
So I went ahead and preordered the new XLITE-20-V2! 
One idea I have for the buzzer is to have an “idle timeout alert” feature. i.e. if VESC is idle after 10 minutes and the BMS is not connected to a charger, have the buzzer trigger.
What do you all think? 
Buzzer could be programmed to do many things, but it is right now by default only enabled for fault indication. It will emit short beeps after a fault according to the associated fault code number.It can also disabled…
There are parameters on both VESC and XLITE for “not used timeout” and “min current threshold” for going to power down state. So I don’t think it is necessary to add a buzzer alarm for this.
Althought, several VESC cannot turn themselve OFF, but some can do it. I think it will become more and more common in the future. For instance, my next E-VESC revision will support momentary push button and will be able to shut itself down completely when left unused. Some might want to keep the ADC3 pin available instead of having a push button, but I’m not sure who is actually using it and for what.
Cool, I’ll take a look at the “not used timeout” and “min current threshold”.
From what I’ve been told not a lot of VESC support he HW feature to turn themselves OFF. I do hope it will become more common in the future. Momentary push button support in your E-VESC would be ideal. ADC1/ADC2 are used for left/right footpads for self balancing applications, but I don’t think ADC3 is used for anything.
Have you thought about targeting your E-VESC revision to fit in the flowglider box: little focer V3 WTFoc MK2 by flowglider - Thingiverse The main two complaints for the littleFOCer is no momentary push button support and no 12v/5v 3A rail for lightbars.
You currently have the market cornered right now for a smart charge only BMS that fits for this application, so was just wondering if you considered targeting this market with your E-VESC design for a complete solution.
Either way, I look forward to receiving my XLITE-20-V2 and you’ll probably see a PR or two from me down the line in the future.
1 Like
For folks who are looking to install a 24s XLITE (I assume at the moment, Onewheel builder people), I tossed together this video today on setting up the 24s XLITE v1.
Granted, this is a larger 20s6p, but the process is more or less the same. I hope it helps.
8 Likes
Well done Mario.
Maybe a few things I’d like to add:
-To reboot the BMS after going in power down, you can simply go in the command line terminal and type “reboot” and click on send command. The BMS will do a full reboot. This is useful and will avoid disconnecting the USB or charger to initiate a full reboot. V2 will have push button long press for that as well
-VESC tool V5.02 and above works even better than EBMS tool for configuring XLITE. The way it is done via VESC tool is completely different. Config options enum are stored in the BMS itself and read by VESC tool directly. The compatibility is then always perfect and you don’t need to care about firmware and VESC tool version compatibility. Fewer config options are available and you cannot really mess things around with VESC tool. So there is no “store to flash” needed and a simple write is needed for both writing and storing
-You can fine tune current readings for charger and also tune charger and pack voltage on XLITE
4 Likes
Question: looking to change the cycle time for balancing. Currently the bms (xlite) will charge for a second or less then switch back to balancing then repeat. Is there a setting to increase this to 3-5seconds? I was looking through the app and can’t find anything
It does not really works like that.
BMS will enable the charge mosfets until the max. cell voltage reaches the cell soft overvoltage limit.
It will then wait for the charge timeout to get cleared to retry enabling the mosfet.
1 Like
Ok so if I understand correctly I must have my soft over voltage limit value too close to the highest value that initiates a charge state causing it to cycle too quickly between the charge and balance state?
1 Like
Have you ran this ESC in any boards? Thoughts?
There is no such thing as
highest value that initiates a charge state
Maybe you should tell me what’s wrong with your charging process, It would be a lot easier for me to tell you what may happen.
Going into balance state is just fine. It will keep charging. Most problems occur if you are trying to charge too quick when the cells are getting close or above 4V and can’t take properly the incoming current. The cell voltages go through the roof every times the charging process is enabled.
I know some people wants to charge in 30 minutes, but this is very tricky to do safely with a generic CC/CV charger unless you stay below 80% SOC.
While I don’t think it is a viable alternative for e-skate application, safe and fast charging (current throttling for end of charging cycle via CAN bus communication) is right now only fully implemented for TC chargers like this one: https://sd94f85fb6682eee7.jimcontent.com/download/version/0/module/15141527225/name/TC%20Charger%201%2C8%20kW.pdf
2 Likes
What im seeing is the cmismatch value during balancing fluctuating so rapidly that it never seems to complete the balancing state when plugged into a charger. Though i am a novice so having so much more information then ever before with this interface as well as trying to learn as i go, i am likely looking in the wrong place for what could be a non problem :-). That’s why I’m here lol just paranoid because I’ve had some bad dumb BMS luck previously that has gotten expensive. I want to make sure, now that i have more control over exactly what is happening, i don’t mess up the settings.
I think I figured out what might be causing the perpetual balancing. In the terminal config it shows: cellbalancediffthreshold of 0.010
.01 might be finer matching then my cells can achieve? Should i set the value higher? .02 or .03?
Thanks for the lesson, i need all the help I can get.
Balancing can take a very long time. Several hours initially. 0.01V difference is normally fine.
I think it took a bit more than two days for one of my 29Ah pack that had a P-group at 3.6V and the other P-groups that were at 4.1V
The mismatch can fluctuate like that during charging, but will eventually succeed staying below 0.01V when the charging process stops.
1 Like
Thanks, I’ll just let it cook a bit longer.
Unrelated question: testbms connection testing has fail results on 5 of the groups, what bms state should it be in to test for best results? Is this concerning or what am I looking at exactly in this test output?
“testbms” command in the terminal should succeed when the BMS is running idle (Not charging, not discharging)
If not, check your balance connections. This could also explain part of the fluctuations you see.
Show me your command results.
1 Like
