Connect the controller using a usb cable.
Open a serial terminal to the controller, reboot it and press space bar when it powers up.
You can then select a new WiFi hotspot
1 Like
Hello all,
Not sure if this is the correct place to post this, so apologies if Iāve got it wrong!
Iām looking at using diyBMS with an 18S LiFePO4 setup, itās 4P totalling to about 840Ah off the top of my head. Itās going onto a Victron setup with a Multiplus 10000VA inverter, smart solar MPPT and Cerbo GX.
My question is, should i use just 1 module/slave board for all the 4P cells, or divide them in two or four and have multiple module/slave boards, for the paralleled cells.
Secoundly, I assume diyBMS is happy integrating with victron Cerbo? And if so, should i add a separate contactor for the batteries or is this not necessary normally?
Thanks!
Why 18S? Typically people use 16S and thatās what the diybms all in one monitor board is for - monitoring 16 cells in 1 module.
Because the batteries are in modules of 6S that canāt easily be divided up as theyāre welded tabs. Iāve only just arrived at this installation but Iām rather worried itās all too high voltage for the victron systems. I suppose that would put an absorption voltage of about 65V. Is that too high for victron stuff, Iāve not worked with Victron before. Itās a 250/100 MPPT.
This is far from the only 18s Victron system out there. As long as your solar array yields higher voltage than the batteries, youāre all good.
The MPPTās data sheet says that its āequalization voltageā is 64.8V by default in a 48V system, which is perfect when you have 18S of LiFePo4 (3.6V per cell).
Ok thanks, iāll give it a go seeing as itās what we have. Back to the original question, would it be recommended to run parallel module boards or just one per parallel bank?
Secondly on Victron integration, is diyBMS generally rather happy with this? Any need to run a contactor inline with the cells or not really
Thanks!
Hi, Everyone,
Thanks to all of you and Stuart especially, i realise the project of 32 cell of lifepo4 with the V4.5 modules. (I started to assemble the devices when the all in one was not available)
Iāve did few days only of testing and seems to work good.
I Connected the Victron cerboGX to the diyBMS controller via CAN.
But i have problem with the āDynamic Charging Algorithmā.
If activated, during charging the charging stop for some short, see the Victron Smart Shunt monitoring current attach.
Seems the BMS sometimes send the value 0 to the Maximum current allowed for 1seconds.
There is no CAN problem see attach.
There is no reset of the bms.
what about my parameters in diybms, are they ok ?
I donāt have diybms shunt, only Victron smart shunt.
Let me know what information i can give you more to help solve the problem ?
Thanks for help
Mickael
What did you enter as global settings for the modules?
I currently only have the Victron shunt and am not experiencing this behaviour.
In the Victron VRM portal, you can also view the CVL and CCL values under āAdvancedā. You could take a screenshot there.
Hi, Thanks for reply
Here is my settings for module 1 but all the same for other module except calibration that i did.
CVL and CCL work fine and are updated good.
But when i activate ādynamic charge algorithmā sometime CCL drop to 0 for 1second and the inverter instantly stop charging, and start again. We can hear a āclackā noise in the Multiplus at this moment
I connected the diybms to Home assistant via MQTT and can see the value CCL drop to 1sec.
This problem happends since the first start of the installation.
To find the source of the problem i was thinking :
Why during normal charging (battery voltage at about 3,3V ) diybms controller could send CCL = 0 to the cerbo ?
Also is it normal that the display of the tile āCharge modeā is always written āStandardā ?
Thanks
Best regards
You have not enabled dynamic charging.
Here are screenshots of my settings. I have 18 cells, for your information.
āCharge Modeā will not change for you, this only works if you use the DIYBMS shunt. Therefore, āAbsorptionā and āFloatā also have no effect.
Yes it is on purpose because i activate for testing only as iām afraid my inverter will not like this if i leave for long time.
relay made noise when current drop so fast.
Thanks, i took my settings here from āstuartā
Iām not a programmer so i donāt know where to search but i did testing again and connected putty on serial port for logging but i didnāt understand what to look for.
I just noted problem happends around line 523784
and on line 577013, i uncheck ādynamic chargingā and no more charging problem.
diybms log_2025-12-12.txt (91.8 KB)
I have seen reports of this with another inverter over can bus.
I added lots of logging to the controller code, but never found the root cause. From what I could see, the controller never sent the zero current setting.
Iām interested in your mqtt logging though, is that just logging from diybms? Perhaps we can use that to find the issue.
Do you have any rules that are triggering (temperature etc)?
Thanks,
On the diyBMS controller ESP side, i connected to serial usb port on diybms but i didnāt found the problem on the log diybms log_2025-12-12.txt
But maybe i donāt know what to search for.
On the Victron Cerbo GX side, connected MQTT to home assistant, i found this graph where we can see this current dropping.
What do you mean by ā From what I could see, the controller never sent the zero current setting ā ?
On my picture graph we can see āBattery Maximum Charge Current Allowedā dropping to 0 many times
If you saw this on the CAN bus it means this value come from the ESP.
Maybe a variable in the calculation that could end up to send overt the CAN bus āBattery Maximum Charge Current Allowedā = 0 sometimes ?
Let me know what loggin i can do, problem happends instantly as i check dynamic charge algorithm.
No no rules
But i donāt have current shunt installed, maybe something to do with this ?
Only victron smartshunt connected on Cerbo GX.
There is a check in the code which runs before any data is sent on the can is - it basically checks the BMS is in a safe position/configuration including the rules.
If this is fails itās checks it will default to āstopā charging - Iām wondering if this is what you are seeing, something isnāt happy!
Although you donāt use the rules with the relays, the logic in them is used for this check, so make sure they are configured with meaningful values.
i have an issue - after 1 and a half year of working perfect my ESP, LevelShifter and CAN-Tranciever burned suddenly - First i assumed overvoltage of 5V supply - but my dc/dc has good output voltage.
Now i have to replace this parts and after that i will check if there are other defects.
Are there any issues known on Controller v4.5?
No known/reported issues like you reported - could have been a surge on the 5V supply, any chance something touched the board/shorted out?
Hi @mktime
The code for the Victron integration is all here.. diyBMSv4ESP32/ESPController/src/victron_canbus.cpp at 210a268982c8ab72ccea4992d3abab10c4be162f Ā· stuartpittaway/diyBMSv4ESP32 Ā· GitHub
The code runs very frequently and starts with a default of zero max. charge current as an assumed safe position.
The code also runs several checks to see if the various rules and settings/cell voltages etc are all in a known safe range, and if so, the max. charge current is set in the CANBUS message (351).
The checks are in the code IsChargeAllowed and IsDischargeAllowed.
So, if you can use the latest code Iāve just deployed on GITHUB branch - please download the artifacts from here (you will need to login to github)ā¦ Update graph library Ā· stuartpittaway/diyBMSv4ESP32@71eddfc Ā· GitHub
Monitor the serial output of the controller, we are looking for errors along the lines ofā¦
I (287463) diybms-rules: Set warning 7:ChargePrevented
I (287467) diybms-rules: Set warning 8:DischargePrevented
I (287473) diybms: Active errors=1
The code will also output a warning if the set charge current is ZERO
PS: I also noticed that Iād left a comment against Victron and the 351 message - if Victron doesnāt receive this message at least every 3 seconds Victron will stop charging automatically - its a failsafe to ensure if the BMS dies, the battery stops charging. DIYBMS should be sending this message approx. every second.
You are right.. tca is also damaged so each 5v part ist damaged. Must be supply failure
Howdy all, joining in as I found DIYBMS through some rabbit-hole diving. Iāve been trying to find a good, affordable solution for doing battery management on an electric vehicle conversion Iām planning, and the open source nature of this project + the fact you can have it manufactured and sent to your door makes it theoretically appealing. So far it has seemed to me like the V4.9 would potentially be near-perfect to daisy chain from module to module, with the cell taps populated with each of an EV modulesā cells. Typically nice, small cells are used (as is the case with some Volvo hybrid modules Iām looking at) so the passive balancer top hat should work fantastically. Beyond that, itās got temperature readouts, great capability for grabbing accurate voltage, and the controller speaks canbus, which should mean it can work just fine with a ZombieVerter VCU.
My main concerns are:
- The cell tap connector and all connectors on the board seem to lack locking tabs. Obviously thereās a few ways to lock the connector in place, but itās something that requires consideration regardless.
- That I may have some blind spots and missed something else entirely.