Posts from other threads seems to suggest that there's a possible issue w.r.t to SunSynk (possibly limited to just 8kw model) and certain lifepo4 battery makes where the charge voltage is to high effectively overcharging batteries. Instead of overloading those threads further perhaps we can track this here? ...

So far its been reported for Pylontech and Shot, I think both uses Seplos BMS, so this might be the common denominator.

I've also posted on the SunSynk forum:  https://www.sunsynk.org/forum/techical-support/inverter-battery-charge-voltage-too-high about this.

For pylontech refer to onwards:

 

For Shoto refer to onwards:

 

I dont think this is really an issue , If the cells stay under 3,65V they are not being overcharged. The most i see is usually about 3,55V. .

Edited March 12, 20232 yr by Nexuss

25 minutes ago, Nexuss said:

I dont think this is really an issue , If the cells stay under 3,65V they are not being overcharged. The most i see is usually about 3,55V. .

I seem to have it that your value is spot on and no overcharge. 

I wonder what is the source that anything under say 3.65V per cell is over charge? 

As a matter of interest, what is the BMS output (measured voltage and requested charge voltage) when fully charged? (Click on battery status and then BMS status to get this information.)

1 minute ago, JustinSchoeman said:

As a matter of interest, what is the BMS output (measured voltage and requested charge voltage) when fully charged? (Click on battery status and then BMS status to get this information.)

For pylons the charge voltage is always 53,2V and that never changes if you are using CAN comms atleast. The most the BMS voltage goes to is 53,5V

37 minutes ago, Scorp007 said:

I wonder what is the source that anything under say 3.65V per cell is over charge? 

In my case the batteries spec sheet says charge voltage should be 54-56.4V. Its a 16 cell battery so then max 3.525V per cell, right? Am I interpreting it wrong?

39 minutes ago, JustinSchoeman said:

As a matter of interest, what is the BMS output (measured voltage and requested charge voltage) when fully charged? (Click on battery status and then BMS status to get this information.)

Yet, SunSynk on the LiBMS page says charge voltage is 57.8V. This remains constant always throughout. Thus, for a 16 cell this equates to 3.6125V per cell vs the 3.525V max of the battery data sheet. The BMS voltage does reach this 57.8V at the end at which point the battery cuts out because the default over voltage protection on this battery is set at 57.75V. See attached screenshots. The last screenshot is of Battery Monitor after the total voltage protection kicked in, and at this point things has cooled down a bit already which is the reason why the total voltage is 57.4V (and not 57.75V) and the cells just below 3.6V each.

11 minutes ago, jgdt said:

Yet, SunSynk on the LiBMS page says charge voltage is 57.8V. 

This is a problem with your BMS. Parameters are not set correctly in the BMS and thus its asking for a charge voltage that is too high ,its not a problem from the inverters side.

16 minutes ago, Nexuss said:

This is a problem with your BMS. Parameters are not set correctly in the BMS and thus its asking for a charge voltage that is too high ,its not a problem from the inverters side.

Could be, but I respectfully disagree somewhat. The theory is SunSynk is reading/interpreting the wrong BMS parameter. I've checked all the BMS parameters and the only value close to SunSynk's 57.8V is the BMS overvoltage protection of V57.75 (round this up and you get 57.8V). After updating this over voltage protection parameter in the BMS the inverter followed suit and then uses that changed value as its charge value.

I've checked with about 4 different values and everytime the LiBMS page follows suit. So to me it seems that inverter is using the BMS' overvoltage value as the charge value., incorrectly. The 2nd screenshot is where I changed the BMS parameter from 57.75V to V56.9V, as a test, and it shows on the SS connect graphs as well under the 'V-charge-BMS' parameter.

I'm not saying the BMS is entirely innocent here, but it seems neither is the inverter. Its gonna be tough to get to the bottom of this. SunSynk and Shoto/SeplosBMS are probably both gonna say its issue on the other side. I don't know how to get into contact with Shoto/SeplosBMS support, so all option I have for now is to drive it from SunSynk support side.

PS: Cudos for the initial investigation and theory compilation goes to @jumper. I just tested the theory on my side.

Edited March 12, 20232 yr by jgdt

11 minutes ago, jgdt said:

After updating this over voltage protection parameter in the BMS the inverter followed suit and then uses that changed value as its charge value.

Seplos BMS is notorious for being (full of) shit lol ... I would just change that over voltage value to the max charging value of 56,4V then.

8 minutes ago, Nexuss said:

Seplos BMS is notorious for being (full of) shit lol ... I would just change that over voltage value to the max charging value of 56,4V then.

Yes that was my thinking as well to just set it to 56.4V on BMS. Thanks. It could be that SeplosBMS do not have a normal chare voltage parameter and then SunSynk uses this over voltage one in the absence of something sane? Would have been nice if the charge voltage could have been overridden on SunSynk inverter settings.

Are you using CAN or RS485?  CAN only has one output voltage (measured voltage) and one command voltage (requested charge voltage). So there can be no real confusion.

I have played a lot with the Sunsynk CAN link (developing my own BMS comms implementing the Pylon CAN protocol), and have found that it consistently errs on the other side.  I can never get it to charge to more than 0.5V BELOW the requested charge voltage.

6 minutes ago, JustinSchoeman said:

Are you using CAN or RS485?  CAN only has one output voltage (measured voltage) and one command voltage (requested charge voltage). So there can be no real confusion.

CAN from master battery to inverter. So are you saying on CAN there's just one simple "charge voltage request" message (issued by inverter) and upon receiving this the BMS just has one simple "charge voltage response" message to reply on and its replying on its own with the incorrect value? I.o.w there's no subjective interpretation of all the BMS params on inverter side via CAN?

43 minutes ago, jgdt said:

CAN from master battery to inverter. So are you saying on CAN there's just one simple "charge voltage request" message (issued by inverter) and upon receiving this the BMS just has one simple "charge voltage response" message to reply on and its replying on its own with the incorrect value? I.o.w there's no subjective interpretation of all the BMS params on inverter side via CAN?

No the Pylon canbus messages are of broadcast type, produced periodically by the BMS, and consumed periodically by inverter. 

1 hour ago, BritishRacingGreen said:

No the Pylon canbus messages are of broadcast type, produced periodically by the BMS, and consumed periodically by inverter. 

Correct - over CAN, the BMS broadcasts only 2 pack voltages - the measured pack voltage, and the requested charge voltage.  If you see a charge voltage of 57.8V on the SunSynk LiBMS page, then this is the value the BMS is broadcasting.

1 hour ago, JustinSchoeman said:

then this is the value the BMS is broadcasting.

Interesting, thanks for sharing this. So by this token it means issue is on SeplosBMS. Is there a way to sniff and see these CAN broadcast messages, for interest sake and verification purposes when approaching Seplos about this?

Maybe not even an issue.  Not sure of the Seplos control logic, but some BMSs do request a 'boost' charge when they need to balance the cells.

You can buy CAN sniffers, or build a simple one with an Arduino and MCP CAN board.

2 hours ago, JustinSchoeman said:

some BMSs do request a 'boost' charge when they need to balance the cells.

In my case the charge voltage remains 57.8V constant. Doesn't change at all, see attached screenshot from connect graph with the 'V-Charge-BMS' parameter in red (update rate here is every minute). According to the SeplosBMS documentation it doesn't seem like boost charge is applicable.

@ColinR4 Just wanted to say that i had a firmware update a few days ago and it seems they fixed the issue. My battery voltage does not overshoot at all now 53,2V absolute max not 53,5V like it used to be.

On 2023/03/19 at 3:48 PM, Nexuss said:

@ColinR4 Just wanted to say that i had a firmware update a few days ago and it seems they fixed the issue. My battery voltage does not overshoot at all now 53,2V absolute max not 53,5V like it used to be.

Can you confirm that your charging voltage is still 53.2V after your update?

 

2 hours ago, Jacques Ester said:

Can you confirm that your charging voltage is still 53.2V after your update?

 

No , on some days when the battery is fully charged early in the day it will tend to overshoot a bit .Its not constant and varies according to loads. I have seen this on a few sunsynks so its not something that can be fixed by the end user if you are using BMS comms.