Hello together, may someone can answere the question, is it pissible to connect more Controllerboards to one CanBus and what will then be shown on Victron side?
Don’t know if anybody has tried but all indications are that it cannot work. See this link Can I connect more than one BMS to the Venus GX? - Victron Community
The controllers would need to talk to each other and elect one as the ‘master’ to communicate directly with Victron on behalf of all others. This is how Pylontech and likely Victron batteries accomplish it.
maybe it can be possible in future if stuart integrate this in the firmware
A GITHUB issue has been opened for this…
Thanks very much for the hint, may it would be an better option to do it as the Pylontech do it. Use CAN as the connection for the Main Controller and connect the following by RS485 (as it is already on the board). Also probably it is not requested to connect several Shunt-modules, they could be easy addressable on Unit IDs.
It would be really great if you could think about such a solution, as i would say, this is the last feature which is missing to compete with the big BMS supplyers.
Never the less, you did a really great Job with the last huge update @stuart . Thanks very much.
/ Chris
The key point with this Chris, is that a single controller can handle up to 128 cells, so multiple cells in parallel can emulate a bank of pylontech batteries for example.
So there isn’t really a need for multiple controllers, unless you have a requirement for fail over or high availability.
Using 280ah cells for example, you could have 8x16 cells in parallel, for over 110kwh of storage, still with a single controller!
but how slow will the response time be, 5-6 seconds? not realy practical
Why do you need it quicker? What benefits does that bring?
well other bms are quick, response time under a second
but im doing well with 32 moduls and a roundtrip of 1,5sec
i am runnig offgrid a electric heater for my hot water
if the temperature is reached it switches off and the 4kw are going into battery then the voltage rise insanely quick.
you can imagine full battery with 56v then suddenly apply 4kw to it.
my sma sunny island need time to increase the hz for sunny boy to ramp down and as faster the bms tells them what to do, the better.
with 1,5seconds my voltage rises to 57,5v. i can just imagine what happens with 6sec response time.
but this is just for my case, for others the actual speed is fine i think.
This is where the CAN integration with the charger and the DIYBMS current shunt comes into play - the individual modules as just there to monitor single cell voltages and provide balance function.
The current shunt looks after the pack voltages and current sensing, proving useful data to the charger to limit its charging voltage/current.
In your example, a 56V battery would have its charging parameters limited by CANBUS, so it won’t dump 4kW into them.
Am likely too late to this discussion but wondering if anyone here has any insight regarding the related github issue #184 in terms of progress?
From the issue comments it seems some degree of coding/testing was done but there doesn’t seem to have been any related (public) fork of or contribution to diyBMSv4ESP32 code - not sure why.
There was a recent issue comment about “one more person testing on Victron but we need all the help we can get” and at least two respondents (including myself) but nothing heard since…
I’m working on building a modular ‘battery’ consisting of multiple (16S1P) packs, each with it’s own controller/shunt/switching etc but connected in parallel and therefore presenting as a single pack.
@stuart - my aim is to set up something like two or more of your powerBasket ‘packs’ to work together where all cells are individually monitored but where each pack can be managed independently, each with it’s own protection mechanisms etc.
Now I do understand what you meant by:
a single controller can handle up to 128 cells, so multiple cells in parallel can emulate a bank of pylontech batteries for example
…but it’s just not quite as modular in terms of easily being able to add/remove a single pack for whatever reason.
Judging from other discussions, I suspect there is some interest in having a modular pylontech-like capability but I guess it depends on how this project evolves though I’d like to contribute where I can.
-p
I have been testing this version of firmware for 4 months now. works like a charm.
I was one of those volunteers and have a test system up and running. I do not have a copy of the source to share: I was only a tester. I think it would be smart to update Stuart on the progress because he asked to be kept in the loop. I do not know if that has been done.
In other news, I found this today and thought it was an interesting approach to parallelize packs. It doesn’t address our specific use case, but interesting nonetheless. GitHub - limpkin/xt60_parallelizer: a custom board to monitor custom battery packs
Tim