Idiot's guide to a self build battery

[Caesium]

Am I missing a step 1.1) to balance the EVEs first? I think Joe and Caesium you both had an active balancer and balanced the cells first right? I need to check, but I think Deriy balanced and checked the cells before dispatch, so hopefully voltages should be super close still.

I'd expect them to all be very close to 3.285v out of the box if they were anything like mine.

If you're adding an active balancer then there's not much need to do anything manually here - it will take care of it for you.

If you're not adding an active balancer then I'd suggest at the very least watching each cell voltage carefully on the first charge to make sure there's no runaways. (ie one of them shooting up towards 3.65v early; the sign of an unbalanced cell).

[Joeboy]

Am I missing a step 1.1) to balance the EVEs first? I think Joe and Caesium you both had an active balancer and balanced the cells first right? I need to check, but I think Deriy balanced and checked the cells before dispatch, so hopefully voltages should be super close still.

I'd expect them to all be very close to 3.285v out of the box if they were anything like mine.

If you're adding an active balancer then there's not much need to do anything manually here - it will take care of it for you.

If you're not adding an active balancer then I'd suggest at the very least watching each cell voltage carefully on the first charge to make sure there's no runaways. (ie one of them shooting up towards 3.65v early; the sign of an unbalanced cell).

Caesium, how did u do with your other bms? Are you still running original 3rd party bms? When I left Scotland all battery charges were within a hundredth V of each other and that mostly to the basic nature of the measurement device.

[Caesium]

Caesium, how did u do with your other bms? Are you still running original 3rd party bms? When I left Scotland all battery charges were within a hundredth V of each other and that mostly to the basic nature of the measurement device.

They're definitely worth having - I was a bit worried that there was nothing protecting the batteries. The BMS will cut off the current if any single cell gets too high or too low. Although they should stay balanced with the active balancer attached, it's still a nice safeguard to have. The inverter is set to about 55v at the moment, so about 3.43v per cell.

I'm completely without communications to the inverter at the moment, but I could bodge something up to get the JK BMS talking to the inverter eventually. But I don't think there's much need, I can get SOC and various other data directly from the BMS now. There's bluetooth to talk to it locally via a phone app, and also I've hooked up an ESP32 to its serial port for longer term data logging. Getting some nice pretty graphs

I only ever see voltage imbalances at the very high end of charging and this is quite common, not really too worried about it. It's only about 50mV difference and even that will probably drop over a few successive charge cycles as the active balancer nibbles away at the differences.

Overall very pleased with the entire system. The only thing I hanker for now is a bit more charge/discharge capacity (eyeing up a 10kVA Victron inverter/charger, but the price makes my eyes water a bit..)

[Kenny000666]

Ok so here’s mine, I put mine under the work bench, it’s close enough to the inverter and the pylon and put insulation around the sides and the front when I finish. Figured the concrete floor would serve as a heat sink if it ever gets too hot, but I doubt it would be necessary.

Very well packaged, thanks Deriy (Colin)







JK BMS (0.6a active balancing, up to 20s, 100amps charge discharge, brought from AliExpress for something like £50 - the supplier seems to have accidentally sent me 2 )



Levelled out the floor



Stacked the 15s and surrounded the pack with wooden boards and PIR. Cut up the Padding/insulation inside the box to protect the spare EVE cell.


All cells checked, my multimeter shows as between 3.303v - 3.304v, albeit slightly higher than the test values Deriy included (they had the cells at around 3.268v).

So tomorrow, to hack up one of the bus bars to join the 2nd row with the 1st row. Make an Eve cell sided enclosure to store the BMS and to act as a filler to complete the 2nd row. Once finished, I may screw in some 3x2s over the top to prevent any expansion upwards (and as a preparation to a 2nd stack if I ever go that way, or just storage)

Connect up all the cells to the JK BMS, and give it a whirl. I chose the welded stubs and had the bolts and nuts provided. In hindsight probably would have been better to have welded bolts, means one less set of tasks.

Side note: as I have a spare Eve, I’m going to do some reading on nowtys thread to try and understand why making a 16s pack to go in parallel to the Pylon isn’t going to work.

[Caesium]

Nice start Rather than hack up the busbar, could you not put a sheet or two of ply between the two rows to raise the second level of batteries, so the original busbar fits? That will probably give you better conduction/less resistance than having to make anything DIY.

You can't put a 16S DIY battery in parallel with a Pylon as basically the Pylons are 15 cells inside. Not quite as big as these, but the voltages are the same, and putting a 15S battery in parallel with a 16S battery will result in some quite spectacular current flow (as the voltages will try to equalise), with predicable fireworks results.

If you're anything like me though, that spare will come in handy if you decide to go full-on 16S and sell the Pylons, and go completely DIY

[Tinbum]

You can't put a 16S DIY battery in parallel with a Pylon as basically the Pylons are 15 cells inside. Not quite as big as these, but the voltages are the same, and putting a 15S battery in parallel with a 16S battery will result in some quite spectacular current flow (as the voltages will try to equalise), with predicable fireworks results.

Don't think that's the idea, it's to put the extra in series with the Pylontec as in Nowtys thread.

[nowty]

My battery system is well weird, I have 40 kWh of 14S LMCO in parallel with 20kWh of 16S LFP.

14S LMCO is similar voltage to 16S LFP.

As the Pylontechs are 15S it was advantageous to add a 1S DIY cell in series to the Pylontechs to make a 16S LFP battery to parallel up to the 14S LMC battery.

It’s not something I wanted from the outset but I had one Offgrid system and one Ongrid system of different battery chemistries. I was able to combine the two so everything is now Ongrid and it all works now fit and forget.

https://camelot-forum.co.uk/phpBB3/view ... 954#p12954

[Joeboy]

Hey guys,
Happy new year!
Batteries I ordered in November arrived slightly earlier (1 week) than I had expected! 50 days from order (44 days from dispatch) to my house. So time to begin my project of building my enclosure and connecting up.

Question for those who hooked up to pylontechs.
My plan was to
1) disconnect the pylons from the inverter
2) connect the EVEs + JK BMS in parallel to the pylons
3) do the standard procedure of allowing the 2 packs to balance
4) once balanced close the disconnector and have almost 20kwh of storage

Am I missing a step 1.1) to balance the EVEs first? I think Joe and Caesium you both had an active balancer and balanced the cells first right? I need to check, but I think Deriy balanced and checked the cells before dispatch, so hopefully voltages should be super close still.

Sorry Kenny, missed this. I did indeed buy a charging unit but didn't use it. I set all the batteries in a line, hooked them together in parallel and equalised them. Took a note of the voltage after 20 minutes or so then adjusted (drained) the pylontech stack to match the voltage on the day and made the final.connection. Worked out well.

I can't see any pictures out here but it reads is if you are well.on your way to victory. I'll enjoy seeing them when I get back, congratulations!

[Kenny000666]

Day 2
Would have finished the job today if i didn’t run out of crimps, just 3 short. Ah well, will hook it up tomorrow.

All wired up to the BMS and turned it on. Looks good.

The most annoying part was keeping the BMS cables tidy. Next time, start with a good cable management system before hooking up. For now it’s the best I could do with keeping them as neat as I could.

Hacked up some of the spare bus bars (they just copper bars coated in chrome) so it is relatively easy to drill.



This is the BMS screen, and everything looks good, I think. At least no errors are being thrown by the BMS, and it’s actively keeping the voltages as close as possible. For £50 I do recommend getting a BMS as it’s pretty much plug and play.



So tomorrow, finish off preparing the wires for connecting to Inverter and Pylon (after buying another pack of crimps)

Instead of building a filler pack, I just put the spare Eve into the 2nd row. Maybe one day I do what Caesium did, and get rid of the Pylon. Had read what nowty did about adding another DIY pack to the pylons to make it a 16s, but for now, I’m not at the level to digest it, so will just KISS it for now

Question for you Caesium, what info did you use to connect the JK BMS to an ESP32? Did you have to buy a special version of the BMS to do that? I’m thinking once things have been hooked up, I’d like to connect the BMS up to my raspberry pi and read the BMS info into HomeAssistant.

[ALAN/ALAN D]

Hello Kenny

What type of B.M.S. is it.

Where is it sold.

Regards Alan.