Brandon avatar image
Brandon asked

Pylontech UP2500 100% charged but staying in absorbtion

Hey guys,

I have recently installed 2x Pylontech UP2500 batteries, Multi-Plus-II, Smart Solar MPPT 150/100 & Cerbo GX.

I am currently in the process of cell balancing the batteries. I am charging them through the Multi at the moment due to PV array not installed yet. I am balancing them individually.

I have set the current to 1A and have slowly been increasing the voltage up from 27.5v which has worked to get the battery to 100% charged. I should note that I have been changing the charge settings using Victron Connect with DVCC turned off on my Cerbo GX. The reason I am doing it like this is because that is the only way I can control the voltage due to my DVCC settings not having the "Maximum charge voltage" control option (Firmware is up to date). And to my understanding, If I turn DVCC on, the voltage adjustments I make in Victron Connect will have no effect and be overridden by the Pylontech BMS.

Anyway, once the battery had hit 100% charged, it stayed in absorption for another 20 minutes or so before giving me the high voltage alarm. Shouldn't the Multi have gone into float mode well before this?

Sorry for the ramble on but I have 3 main questions:

1. Shouldn't the Multi-Plus go into Float mode before the SoC reaches 100%? Or at least once it hits 100%?

2. It's my understanding that the best way to balance LiFePo4 batteries is to get them fully charged and them keep them fully charged for a little while to help balance them, then the high voltage alarms should go away in future charges. Is this right? If so how long are you supposed to leave them fully charged?

3. My first battery is 100% SoC. Can I call this battery charged for the sake of this exercise or should I be leaving it hooked up to the Multi and lowering that voltage again until it goes into float mode?

Thank you in advance for your help!

Multiplus-IIbattery chargingLithium BatteryBMSPylontech
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

3 Answers
Alexandra avatar image
Alexandra answered ·


The best way to do the first setup is to set up how the Pylontec Victron page recommended. Then leave the batteries to fully charge and sort themsleves out.

Multiplus wont go into float untill the batteries have fully charged, then they will show 99% or so since float is technically a lower soc/voltage than absorption.

It also ow wont detect float in anycase since you have throttled the charge amps under the tail current.

Bigger banks take longer to balance out your two if the are connected together will be about 3 days maybe depends on charge conditions and if you are cycling. I think the longest I have had a bank take is 7 days.

If your voltages of the the tow batteries are less than 0.2v apart. Plug then together and run the system in batteries charged mode with recommended settings. Leave it for the day. It should settle down.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

d3-energy avatar image
d3-energy answered ·

The first time I charged each pylontech battery with a lab power supply. The difference between the batteries (SOC) was too great.

I did need a voltage of 53V to reach 100% with all batteries. Now I use Victron's recommended settings for charging by the Multiplus.

The difference in voltage (at pylontech) between 92 - 100% is not that big in practice when charging.

You can also see the charging process visually presented in VRM. That gives you a good insight when CC/CV take place. You can compare it also with the Multiplus states.

My experience with balancing the cells is that pylontech manages this very well and you can rely on it.

If you are really interested in the individual cells of your batteries, you can read data using RS485. There is documentation for it on the internet.

If you set the Cerbo and Multiplus according to the recommended settings, you should in principle not get an alarm, because pylontech blocks charging at the end of the charging process (bms).

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Brandon avatar image Brandon commented ·

Thanks for the info. I did try switching on DVCC (and not adjusting any current or voltage at first) and allowing the Pylontech BMS to be in control of the charging process, but then I hit around the 95% mark and started to get the High voltage alarms.

I then followed the advice on the Victron/Pylontech setup guide and adjust the the voltage back (through DVCC settings) until the alarms went away. I charged the first battery and slowly but surely it came to 100% (took about 8 hours). Now with the second battery its been at 95% for 16 hours or more. I have the current limited to 1A and voltage limited to 27.1v. That's as high as I can get it without the alarm going off for now. Highest cell voltage is 3.56v, Lowest cell is 3.37v. The difference between highest and lowest cell voltage doesn't appear to have moved much at all and it concerns me that It hasn't even moved up 1% SoC.

I hav updated firmware in the Multi and the Cerbo.

0 Likes 0 ·
d3-energy avatar image d3-energy Brandon commented ·

I think there is a mismatch with the SOC that the pylontech (batt. 2) communicates to the cerbo.
That was the case with me in the beginning with some of my batteries.

The pylontech declined to provide a 100% SOC, although the voltage corresponded to a 100% SOC.
That was the reason for discharging again and then charging with a lab power supply to maximum cell voltage. From then on, the pylontech correctly gave a 100% SOC.
How the pylontech calculates the SOC I don't know, but it's not just based on the voltage.
When charging I see the SOC ramping up nicely and then it stops for a while and suddenly jumps to 100%. That last 5% SOC (in calculating) does not represent much energy in practice.

You probably set this up correctly (but I can't find it in your info), but for pylontech you need to set it in the following places:
- cerbo gx: DVCC (and ESS if assistant is present)
- VE.configure: charger and inverter tabs
The documentation could be a bit clearer for 24V batteries (correct limits).

I wouldn't worry too much about the cell balancing. You would have to discharge and charge a few times to be able to judge this.

0 Likes 0 ·
raymiller avatar image
raymiller answered ·

After a couple of attempts at charging new UP2500 (24V) Pylonech batteries my solution that worked (more than once) was to use a small power supply with Constant Current and Voltage capability <10W. This solution was derived by using data logging equipment, a variable lab power supply and the manufactures PC software Battery View to monitor the voltage of all the internal cells. Each battery was done separately.

The secret to success was patience and dialling the Constant Current back to around 150-200ma with the Constant Voltage set to 3.55V/cell (28.4V) for if any cell reaches 3.6V the Pylontech protection circuit is activated and everything stops. At 150-200mA (this current for a 48V battery will need to be reduced by about half) seems to be the current that the cell bypass circuit can handle without exceeding/reaching 3.6V allowing all the others cells to catch up, the process then becomes set and forget. Only checking to see that the battery terminal/power supply voltage has reached the 3.55V/cell level which then is in a fully balanced state and ready for installation. Hours to days may be needed.

The time can be shortened by increasing the current on the initial charging but will require manual intervention (dialling back the current to 150-200ma). When the high cell voltage of 3.6V (alarm state) is reached by the first cell a short delay of 5 minutes will be required while the 3.6V cell bypass circuit reduces its voltage below 3.6V to deactivate the protection circuit.

Lowering the voltage to 27V and current to 1A (as recommended by Pylontech) in my view was almost pointless as the cell which reaches the fully charged of 3.6V causes the whole charging to halt with the high voltage cell state. This process fails to achieve a fully charged battery with all cells balanced.

I should make a comment that the Pylonthech manual should alert the user of this cell balancing (design) problem and viable balancing proceedure at first installation as it seems to occur very frequently and causes many hours of wasted time and damages the brand.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Brandon avatar image Brandon commented ·

Hi Ray, thank you for the info. I don't exactly have access to all that equipment.

I can see how your method would work well but surely Pylontech doesn't expect the installer to implement a procedure this tedious with every new install. After all, if you are having to monitor and change all of this constanly to get the cells balanced, what on earth is the BMS doing?

I noticed you mentioned Pylontech's recommendation of 27V and 1A. Could you share a link to this resource? I have not been able to find anything from Pylontech on this topic accept for the Victron/Pylontech setup information, which isn't so helpful for this cell balancing stuff. My dealer doesn't seem to have much info either.

0 Likes 0 ·