[RE-wrenches] LFP Inter-Battery Communication

Sat May 17 09:15:07 PDT 2025

I watched the video and this makes sense to me. Can we get some input from other manufacturers on the same condition or are the battery management systems accomplishing this already? And does it happen in the same fashion as this bleed off of the higher charged cells?
An explanation would be wonderful manufacturers please and thank you.
Dana Orzel - dana at solarwork.com - 208.721.7003

On May 17, 2025, at 9:52 AM, Maverick Brown via RE-wrenches <re-wrenches at lists.re-wrenches.org> wrote:

 Jay,

Over time I noticed that with lower voltage (say 54.4 [3.4vpc]) and little to no absorb was causing a wider gap in the SOC of individual batteries. I found that the solution was to charge to a higher voltage and to absorb longer that 5 minutes, etc.

I could observe this spread of SOC on various Lithium battery brands and I started editing the settings to get back to a more Lead Acid style approach. It works.

I was typing an even longer explanation… This video says it all:

<https://youtu.be/METijpjpKWE>
<hqdefault.jpg>
FAQ: Charging Cycle: what is absorption for, and is it needed? l Battle Born Batteries<https://youtu.be/METijpjpKWE>
youtu.be<https://youtu.be/METijpjpKWE>

I chose 3.45V per cell for all LiFePO4 batteries (so 55.2 for 16s batteries) and so far, all customers, all brands are fairly tight on the SOC range per battery. I have to be onsite to measure that, of course, but data looks good.

I set the Absorb time long enough to give the individual battery entities time to internally balance and catch up on SOC. The higher SOC batteries will naturally taper their charge current. But, 55.2 is not high enough to trip a Charge Relay in any battery.

As a side note, Fortress sells the Guardian and that lets you see the SOC of individual batteries from afar and then you an do something about SOC remotely. I wish I could hack the Guardian for other brands, I would install it everywhere. I wonder if Solar Assistant can see individual CAN connected batteries??

All Y’all, Let me know what you think and if you have any feedback.

Thank you,

Maverick

Maverick Brown
Off-Grid Solar Commander since 2006
Maverick Solar Enterprises, Inc.
 • Solar Commander Remote Power
 • SunFlow Systems Cathodic Protection
maverick at mavericksolar.com
512-460-9825

On May 17, 2025, at 10:21 AM, Jay <jay.peltz at gmail.com> wrote:

Hi maverick

Can you explain the logic of 55v for 3 hrs?

Thx
Jay

On May 17, 2025, at 7:50 AM, Maverick Brown via RE-wrenches <re-wrenches at lists.re-wrenches.org> wrote:

Jason,

I think it is valuable to have SOC information available if you also have an InsightHome device that can show the SOC data for reference etc. For legacy inverters, voltage still makes the decisions.

I don’t know of any battery that communicates together to balance between individual battery entities. That still leaves the necessity to charge via voltage high enough to get all batteries to high and similar SOC.

The E-BOX says via Pylontech protocol that its battery voltage range is 46.5V to 56.8V, so I think 48V LBCO is fine and certainly 44 is too low.

There is a “.8” version of battery firmware and a “.16” version. I can send you the .16 version that I have (SPBMS16SRP2205V1.5.18.C16). I use HyperTerminal to install the software and it is the absolute easiest battery to view the details using the Pytes USB to Serial cable (other cable brands could also work). HyperTerminal was provided free by Pytes. I can send you that as well. A lot of this might still be on their  website.

I lieu of having a Pytes Hub to stack the existing bank as multiple groups of 10 batteries, you could upgrade each battery to .16 firmware and make two communicating stacks of 10 that don’t communicate together. If you have InsightHome, you can wire the CAN port of the left 10 to Insight and the RS485 port of the right stack to Inight and make “two battery banks” show up in Insight. On the Pytes, the CAN RJ45 Blue pair is CAN and the Orange Pair is RS485 (from the RS485 port). If you have InsightFacity you can use both CAN ports. CAN devices show up automatically and RS485 device need to be setup via InsightLocal / Setup / Device Detection / Detect Device / RS-485: 1 to 10.

On the Charge Controllers, I would set Absorb to 55.2 for 3 hours.

I even started using EQ at 56.0 on some systems.

Good luck today!

Maverick Brown
Off-Grid Solar Commander since 2006
Maverick Solar Enterprises, Inc.
 • Solar Commander Remote Power
 • SunFlow Systems Cathodic Protection
maverick at mavericksolar.com
512-460-9825

On May 16, 2025, at 6:22 PM, Jason Szumlanski via RE-wrenches <re-wrenches at lists.re-wrenches.org> wrote:

The advantages of closed loop communication between LFP batteries and inverter systems are pretty obvious. But when you are operating in an open loop because you are working with a legacy inverter that does not support battery communication, how important is it for batteries to be wired together?

I started contemplating this today when I ran into a Pytes stack of 20 E-Box 48100R (5kWh) batteries. Whoever installed them wired all 20 in a single communication daisy chain. Come to find out, this version of the battery only supports 8 or 16 in a communications network. The bank is connected to a couple of XW+ inverters, so it is operating based on voltage. The owner must have had to use an unconventional method to turn the batteries on, because typically you would only press the switch on the master battery. In this case, that doesn't work because the communication chain is broken at some point.

The reason for my visit was the battery is sitting at 43 volts because the generator didn't start and the low battery disconnect was set to 44 volts, well below the 49 volts recommended by Pytes. Side note:  I could not set the low battery cut out higher than 48V with these inverters.

I was there to get these jump started with a Chargeverter. I rigged it up and got things going, but various batteries were alarming out. Ultimately I just disconnected the communication cables on all the batteries and they all started charging from the Chargeverter which was connected to the common bus. Awesome!

So back to the original question... Are the batteries really doing anything in terms of balancing when they are connected in a communications daisy chain when they are not enclosed loop communications with an inverter? How important is the communication anyway? And in this case, assuming that the maximum batteries in a communication chain is eight, I would probably be better off setting up four separate communications with four master batteries in groups of five. Again, I'm not sure how important that is in terms of balancing. What if I had different quantities per communication group? They're not talking to each other anyway!

Not to muddy up this thread, but I will say that the Chargeverter saved the day. Within 20 minutes, I was able to get the inverters turned on and the SCP to help me identify what went wrong. Then I turned on the inverters and got them charging the batteries simultaneously with the Chargeverter and solar. It really just needed a jump start! I had rigged it up in a way that I could easily do this and then safely disconnect the Chargeverter while the rest of the system continued to charge the batteries. It was my first experience with this little gem, and I'm happy to have it in my toolbox now.

Tomorrow I am going back to restart the charging process and hopefully get the batteries up to 100%. I also plan to check the firmware, which I am almost positive is mismatched in this stack of batteries. All of my off-grid systems are a boat ride away, so I'm going to pretend it's a weekend and I'm not working. Haha.

Jason Szumlanski
Florida Solar Design Group
_______________________________________________
List sponsored by Redwood Alliance

Pay optional member dues here: http://re-wrenches.org

List Address: RE-wrenches at lists.re-wrenches.org

Change listserver email address & settings:
http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

There are two list archives for searching. When one doesn't work, try the other:
https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

List rules & etiquette:
http://www.re-wrenches.org/etiquette.htm

Check out or update participant bios:
http://www.members.re-wrenches.org

_______________________________________________
List sponsored by Redwood Alliance

Pay optional member dues here: http://re-wrenches.org

List Address: RE-wrenches at lists.re-wrenches.org

Change listserver email address & settings:
http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

There are two list archives for searching. When one doesn't work, try the other:
https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

List rules & etiquette:
http://www.re-wrenches.org/etiquette.htm

Check out or update participant bios:
http://www.members.re-wrenches.org

_______________________________________________
List sponsored by Redwood Alliance

Pay optional member dues here: http://re-wrenches.org

List Address: RE-wrenches at lists.re-wrenches.org

Change listserver email address & settings:
http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

There are two list archives for searching. When one doesn't work, try the other:
https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

List rules & etiquette:
http://www.re-wrenches.org/etiquette.htm

Check out or update participant bios:
http://www.members.re-wrenches.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20250517/2e38c749/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hqdefault.jpg
Type: image/jpeg
Size: 11344 bytes
Desc: hqdefault.jpg
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20250517/2e38c749/attachment.jpg>