10kWh LifePO4 DIY Battery Addition..... Here is my write up.

You could test by charging from the grid?

So am I correct in saying that for Solis the default consistent charge rate is in the range of 65 - 70 amps as it derates after 10/15 (who knows at this stage) number of minutes at 100amps, but some have managed to get it bumped up to run at 100 amps (depending on the firmware version)?

Or have I dreamt that last bit!?

:) that or get shopping for a sunsynk and pass the solis along!

Anyone used a Deye hybrid?

Based in the UK with a UK CEO and support team. Good guys. They have a strong presence in South Africa too. I'd buy a Sunsynk over anything else now.

Anyone interested in this?

Not sure why this has happened, max current set to 95amps as only using 25mm2 cables, have new 35mm2 cables to go in

Quite likely the bolts weren't tightened enough.

Don't ask me how I know 😁

Wasn't too bad actually but years ago I had two pieces of cable connected to each other with a nut and bolt and covered with electrical tape. By chance I grabbed the connection at some point and nearly burnt my hand. Loose connection. Problem was gone when I tightened the bolt.

@Scoopsire It's not the Solis inverter that is derating to 10amps, it's the Seplos BMS.

There are 2 settings in the BMS that control it:

1. Take the initiative current limiting charging
2. Passive current limit charging

From my investigations so far to figure out what is happening, is that if you charge at 100amps, one of your cells might peak at 3.65v before the rest of the them so it might trigger one of the overpressure warnings, etc. I believe what is supposed to happen, is that the BMS try to balance that cell with the other cells while still charging so that all cells charge up to 3.65v. From what I have seen is that the Seplos BMS, hits that limit on one of the cells and then changes the SOC to 100% and won't accept any more charge.

My current solution to this is so limit charging to 30amps in the Solis which means it will take longer to get to 100% SOC

Only the Seplos BMS can't balance cells 😂

It only burns off high cells into heat, losing you your electricity. An active balancing BMS on the other hand charges a capacitor from your high cell and discharges the capacitor into a low cell

I completed my 16S4P battery configuration for a total of 800ah. I also replaced my Seplos with a JK BMS. Since doing that, I noticed that the battery does charge to 100% but I typically never see more than 54.5v (it never goes to the absorption voltage of 56.0v). The cells themselves usually are around 3.45v at 100%.

Are others seeing this, any pointers on potential configuration changes?

When I set up the JK I completed a full charge/discharge at 50A 2500w. Given our current weather, the battery typically doesn't go below 70% during the night. I'm sure that will change next month once Heatpump and more load occur.

Just finished testing a set of 16 x 305Ah Envision LifePo4 cells from NKON. Details below for reference.

Positive Comments:

• Very well packaged, came in 4 boxes of 4.
• All cells exceeded nominal capacity in tests.
• Overall happy with the price and purchase so far.

Other Comments:

• Including bus-bars/nuts, shipping & VAT from NL, the order total was €1918.13 (or 119.88 each). Looks like they only charged NL VAT.
• All barcodes/numbers were scratched out and unreadable. NKON should have mentioned that.
• Advertised as Grade A- cells (A minus). Clearly not! (Fogstar advertise them as Grade B)
• I did query the grading from NKON before purchase and they replied that they are Grade B but their supplier has them as A-....but never changed the description on their site.
• It appears to me that NKON offer little or no value add & just re-ship the boxes as received from their supplier in China. (Maybe they test a few at the start???)
• They are now out of stock but may still be available from Fogstar in the UK.
• Some of the cells have some minor bulging on the wide sides typically in line with the terminals (Doesn't appear to be too bad).
• Took just over 1 month to complete the testing. (See tested Column)

Comments on Cell Tests & Spreadsheet: (See Images)

• used ZKETECH EBC-A40L for testing. This can nominally charge/discharge at 40A. However after digging into the docs, it is only rated for 35A charging & 40A discharging. It looks like it de-rates to low 30 amps as it heats up. Otherwise it did the job - no complaint.
• Initially I considered testing overnight but quickly changed to 1 day charge test and the next day discharge test.
• The test cycles can been seen in the image. Each cell tested individually. Essentially, discharge to 2.5V and immediately charge to 3.65V. After a rest period (overnight for me) discharge again to 2.5V and recharge to 30->50% capacity.
• The good news is that the capacities exceeded nominal under these test conditions.
• The bad news is that they were delivered at 6% SoC, when the datasheet specified 30->50% SoC for storage/shipping. On the other hand it does look like a capacity check was carried out before shipping from China as each box of 4 seemed to be grouped by measured capacity (tested in Mar-23, delivered in Jul-23). See Box Rating Column in image.
• The average charge was 311.32Ah and the average discharge capacity was 309.69Ah.
• The weight, Voltage & IR columns were measured on Delivery.

Next Steps:

• Waiting on metal case (See image) from Alibaba which will be a while via Boat. Using PayPal, it cost US$295 including shipping to door & DDP(all taxes). Also includes all the bits and pieces required except BMS/Display. Includes 2 terminals only (not 4 in image).
• I am using my own JK BMS (16S2A). They will punch out the holes for whatever BMS/Display you want. Template is usually available from BMS manufacturer.
• Delivery is quoted as 40 working days (8 weeks), But will be much longer for me due to initial shipping delay & Typhoons in China.
• I will post a link to it once it arrives and is OK.

Dyness come in 15S and in 16S. You can build a DIY battery and use it in parallel with your Dyness, but both have to have the same number of cells in series (and be the same chemistry).

Only one of the two packs then can tell the inverter what to do. Usually you'd pick the higher capacity one. If that is your DIY pack, then its BMS will control the inverter and the Dyness will just piggy back on the settings (protecting itself too)

I have done this with a very large DIY setup and a 5kWh server rack pre made pack in parallel myself, worked fine