The existing one I have, 2 models back, looks like this outside the case:
I do appreciate:
Going to follow this project closely 
So far this new interface of @Louisvdw is working like a “hand in a nice glove”.
Titbit: Re. the BMS vs BMV SOC. A friend on WhatsApp yesterday told me he has the same issue with a brand name lithium bank. He also used his BMV to show them that their SOC is, a lot more than mine, off.
He’s waiting on some updates.
I think I will keep the SmartShunt in place as it is integrated with everything else, and more reliable ito the actual SOC, which I use a lot.
How can i get in touch with you. Great work
Best would be to send a PM in this forum.
And the latest model arrived in record time!
Manufactured 2021/01/05, arrived in Cpt at my door at 8am on 2021/01/22.
From China via India --> Dubai --> Jhb --> Cpt using FedEx.
Model: JBD-AP20S003-L20S-200A-B
And it came in a smaller metal box than the previous model, so it is TTT proof!
This one: https://www.alibaba.com/product-detail/Sales-promotion-smart-7S-24V-10s_1600127268513.html
I asked for the RS485 port to be added … comes with a USB cable:
… and now I can use @Louisvdw software to connect the Venus/VenusRpi with the RS845 port whilst I can view the actual data on the BMS using the Bluetooth Xiaoxiang software on my phone!
How cool is that!? 
To test, I wired it for 24v as this one can also do 24v or 48v, up to 20 cells.
Also bought a 12v BMS, will test that one just now.
Total cost (when the $ was low in Dec): $156.53 with Fast Delivery.
= ±R 2370.21
SA Rates and taxes: ±R575
= ±R3000 for 2 x BMS (48v and 12v) with extra ports added on both, delivered.
I thought I’d put in a picture of what a well-balanced battery looks like. This is the “developer end” of the system. This battery is charged beyond full. But note how close the highest and lowest cell is. As noted before, this battery freaked out at 54.5V when it was new.
That is a well balanced battery indeed!
I’ve tried doing that balance charging with a function in my driver as well, but the voltage differences between what the BMS reads and what the inverter supply using that same value was to big to make this work well.
At some stage I need to dust of that code again and see if I can make it work.
Some feedback on my 2nd Life cells.
12v 100ah Bank 1: I got the cells to have 0.010 - 0.011v difference whilst charging.
Once the system was full, no current into the batteries, slowly the one cell ran away again.
So I discharge them down to 30% SOC and back up again at 20amps whilst using “The Bulb” (now with 2 bulbs and 4 filaments).
Took a bit of tap dancing but in the end, now, the 2nd time, they ALL behave:
12v 100ah Bank 2: Drained them to 3.18v, then with “The Bulb”, took them to 2.8v each.
Then connected them all in parallel and left them for 36h … and counting.
Will charge them up at 20amps too … and see the results.
Worse case: IF I have to use “The Bulb” again, I will go back to 2.8v per cell and recharge them back up at 1-2 amps as Plonkster suggested.
Must say, learning a ton about this lithium thing. 
The latest build V0.3 of the driver has been released.
With the new version the driver will
Found my way here from the Victron community, then github etc following links to this topic. I am yet to get my Venus GX device and have not yet put my battery together. I have a bunch of the 280Ah cells and am currently looking for a BMS to put it all together. Am wanting to use a Quattro to charge the 48V battery pack in a reasonable time. I have a Solaredge PV inverter sitting on the wall waiting to be commissioned and inspected. I should be producing electricity from photons soon.
Glad to have found this information covering Louis van der Walt’s developments.
Just my intro post. New to Home solar. I have been playing around with 12V/18V solar panels and cheap PWM chargers to support my camping setup. This mostly to keep my portable fridge going to keep things cool when away for a few days when away from the grid. Earlier setups without the solar meant dicing with flat batteries in the vehicle. I am now moving to home solar and want to have my own powerwall. Have gone down the Chinese cells path and will compliment my Solaredge system with Victron components. I am attracted to Victron Energy because of their open source approach and the ability to tap into a great community.
Cheers from the other “Down Under” Australia.
Welcome GRO, I am glad to see more and more members joining from abroad…
Solaredge has enough Sunspec support to work with a Victron system, but unfortunately power limiting is not supported (because they don’t implement the “immediate controls” model). My main gripe with SolarEdge – I’m sort-of the defacto maintainer of the component that talks to these inverters – is that they only allow one concurrent connection, and in their latest inverters I’ve had at least one case where this connection also isn’t all that stable.
Waded through a really long network dump trying to figure it out, eventually noticed that the inverter stops responding, then some seconds later you see an ARP request from the inverter (meaning the control board has lost its ARP cache, ie it likely rebooted). In any case, patched the problem… but caused me enough swearing at inverters that only allow one concurrent connection… (cause obviously I will only ever want to access it from one place… NOT).
Well, they are not the only ones at least. SMA is guilty too 
Getting off topic here so I’ll keep it brief.
What you have to say is not encouraging. I already have my concerns about their tech support here as I have been waiting a week to have a fault resolved. My installer is pulling his hair out.
Yes this perhaps needs a thread of its own. But a lot if this is down to preference, loyalties, and what not too. On a technical level I like their product. Using the boost stage of the inverter to also be an module-level MPPT… genius. But it comes at a price. It means putting the boost stage on the roof in a very hot environment. It means the “optimisers” need to be excellent quality, or even a sub 1% failure rate adds up to at least a whole number of failures in many installations. Also, coming from an open source background, I have a strong preference for making things that work together rather than striking out on your own. And when SolarEdge said they want to leave the Sunspec alliance some years ago, for me that was just another reason that tipped the scales against them. They haven’t actually done it, but the support in the inverters have remained as minimal as it always was, basically only the mandatory model 1 (common), and model 101/103 (that has the voltage and power values). And only a single concurrent connection allowed.
But they do work well, and where properly installed (apparently string lengths must be considered) they do the advertised job.
I’ll take this to private messages as soon as I work out how
Hi Louis,
I am busy building a small battery and quickly put your software on my RPi (2B) to check it out - very nice work.
One thing I noticed is that VRM portal reports an incorrect temperature even though the correct temperature is seen on the DBUS. Any ideas why that would be ?
Also, is this the right place to report this?
Ingo
Hi Ingo
I suspect this is not an issue with the battery driver, but how VRM works. The VRM Dashboard only updates for a few minutes and then you need to refresh it before it starts updating again.
This might be what you are seeing.
I suggest you rather switch on the Battery Temperature Sensor widget under the Advance tab.
This is the data that the driver populates and it will keep updating the graph while that page has focus, unlike the dashboard.If the widget is correct and you have refreshed the dashboard and it is still showing the incorrect values you will have to contact Victron about that.
I hope this helps
You are 100% correct, the widget is correct but the dashboard finds a strange value. I will give the Victron forum a try.
Edit: I posted the question on the Modifications forum and after I posted I realised that just after adding the Temperature widget on the Advanced Tab the Dashboard updated to the correct value also. No idea where the 21 degrees came from but stranger things have happened…
If you look at the graph over a longer period (say 2 days) can you see the 21 somewhere in the history?
Also in your GX device what does your temperature sensor setting show under DVCC?
