ESS heavily discharging battery when SOC has reached 100%

Question

question

mihair asked Oct 23, '18

ESS heavily discharging battery when SOC has reached 100%

Hi Victron team,

I have configured an ESS parallel system based on the following components: CCGX running on Raspberry PI (v2.20), MultiPlus 48/5000, BlueSolar MPPT 150/35, PV inverter on AC IN and CAN REC-BMS controlling charging/discharging of 400ah 16s Sinopoly LFP battery.

When the battery reaches 100% SOC the Multi switches to inverter at full power and starts discharging the battery into the loads/grid. Also, even the MPPT is ESS/CAN controlled it continues to supply current into battery at 100% SOC.

Please see the attached pictures. In pictures number 1 and 2 you can see the battery at 100% SOC as is reported by the BMS and the charging parameters. The max charging voltage is correctly lowered and displayed at 52.8V. At any SOC below full this parameter is 55.2V

In picture number 3 you can see the MPPT parameters. Why is the charge voltage 60.8V? This value shouldn't be controlled by the BMS?

In picture number 4 you can see how Multi discharge the battery right after this reaches 100% SOC

Finally, in picture number 5 you can see how MPPT continues to charge the battery in battery hysteresis period (max charge voltage still at 55.2V)

Thanks,

Mihai

ESS

ess-problem.jpg (181.4 KiB)

2 |3000

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

mvader (Victron Energy) answered · Oct 23, 2018 at 09:32 PM

Hi @MihaiR,

> When the battery reaches 100% SOC the Multi switches to inverter at full power and starts discharging the battery into the loads/grid.

In case the commanded voltage by the BMS is dropped a lot when the battery reaches 100%, ie below the open circuit voltage, then indeed it might be that the Multi discharges the battery. It will however stop the discharge once the voltage has reached the lowered charge-voltage setpoint.

> In picture number 3 you can see the MPPT parameters. Why is the charge voltage 60.8V? This value shouldn't be controlled by the BMS?

Try updating the firmware in the Multi, as well as running the latest ESS Assistant. Does that resolve it?

Matthijs

2 |3000

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

AvB answered · Oct 25, 2018 at 08:19 PM

Hello MihaiR, Actually Rec BMS is already well supported. I have three systems running for a year already in Senegal with REC BMS with 14S NMC cells. I do recognise your system behavior, but don't know my settings by heart and cant reach REC BMS via internet. Like already pointed out by Matthijs the error is in the BMS settings. When it reaches 100% it should not suddenly decrease the charge voltage value. As in your system it suddenly drops, the MPPT stops charging and the Quattro will quickly 'help' reaching the desired voltage communicated from REC BMS and thus it's only option is to quickly help discharge the battery to get to that voltage... Than the Rec bms increases the voltage again after its configured hysteresis and Mppt starts charging again and Quattro stops discharging.

There are dozens of other REC BMS systems working with Victron around the world, it really works great, but you need to configure the right settings. Communicate about it with REC BMS their BMS is not working well with your cell voltage settings. Regards, Anco

2 |3000

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

AvB answered · Nov 02, 2018 at 06:18 PM

Hi MihaiR,

First point still charging isn't really an issue in my opinion. That is within the acceptable "searching" area for all connected devices. If you really want even lower top voltages on the battery for longetivity than arrange that in BMS.

About the MPPT stopped? That is indeed not good. ESS should have taken MPPT energy on DC bus and export it right away through inverter into grid if there is still room to export... Check your settings and gridcode if it is activated for exporting.

Regards,

Anco

2 |3000

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

Modifications Space

Node Red Space

Answer 1 · 2018-10-24T06:58:08Z

mihair answered · Oct 24, 2018 at 06:58 AM

Hi Matthijs

All the Victron products are at the latest software version, including ESS assistant

"In case the commanded voltage by the BMS is dropped a lot when the battery reaches 100%, ie below the open circuit voltage, then indeed it might be that the Multi discharges the battery. It will however stop the discharge once the voltage has reached the lowered charge-voltage setpoint. "

Does the charger settings in ve configure still apply for combination ESS & CAN-BMS ?

Strange thing, if DVCC is disabled the Multi dose not discharge at 100% SOC but the MPPT is on his one charging curve, not applying what BMS sent through CAN gateway (CCGX) to VE bus

Please help

Is it possible to access the system remotely?

Thanks,

Mihai

1

2 |3000

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

mvader (Victron Energy) ♦♦ commented · Oct 24, 2018 at 10:40 PM

Hi Mihai,

Does the charger settings in ve configure still apply for combination ESS & CAN-BMS ?

The voltages do not, they come from the BMS. So the place to fix it, is in the BMS.

Please help

Is it possible to access the system remotely?

Yes its possible to access the system remotely, but sorry we can't help with this. A lot of time goes into making the system work well with any type of battery, and we have a long queue of batteries to work on with regards to this. Please contact Rec-BMS; they are very knowledgeable about this.

I hope its sorted soon, success, Matthijs

0 ·

Answer 2 · 2018-10-24T07:06:11Z

mihair answered · Oct 24, 2018 at 07:06 AM

Please see the picture below depicting Multi configuration

ve-conf.jpg (194.0 KiB)

ve-conf.png (90.3 KiB)

2 |3000

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

Answer 3 · 2018-10-25T07:06:17Z

mihair answered · Oct 25, 2018 at 07:06 AM

Hi Matthijs and Victron team,

I did a investigation on what messages are actually received by CCGX from BMS by running candump and this is what I found:

A. There are no dropped packets showing at the output of #ifconfig can0 on RX

B. 0x351, 0x355, 0x356, 0x35A, 0x35B, 0x35E, 0x370 sentences are correctly received.

I also looked at 0x351 and correctly found charge/discharge voltage and max charge current. All those values are identical with the one displayed on various places by the CCGX.

Would you please tell me if there are other messages expected by the CCGX besides the ones stated above at B?

Please point me in the right direction for further investigations. Maybe some processes didn't correctly process informations received from BMS

Thank you so much,

Mihai

@mvader @Victron_Answers

3

2 |3000

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

mvader (Victron Energy) ♦♦ commented · Oct 25, 2018 at 07:16 AM

I dont think messages are missing. I don’t understand why you think they are?

0 ·

mihair mvader (Victron Energy) ♦♦ commented · Oct 25, 2018 at 07:32 AM

Because you point me on investigating the BMS.

If BMS is ok, then i don't know how to further investigate this issue

I must confess that i'm worried by incorrectly charge/discharge the $10k LFP battery

0 ·

mvader (Victron Energy) ♦♦ mihair commented · Oct 25, 2018 at 10:36 AM

I think the 52.8 volt charge voltage setpoint, which is transmitted to our system by the BMS, is too low. But that’s just a guess.

Someone, with access to the BMS parameters, experience with Victron, and knowledge about the voltage/SOC behavior of the cells you use, needs to sit down and test and try what the right number is.

I understand your predicament, but I’m sorry that I can’t help further. We don’t offically support those cells/bms.

However, I do know that rec bms is very knowledgeble. Please contact them for support, not us.

0 ·

Answer 4 · 2018-10-25T11:00:15Z

mvader (Victron Energy) answered · Oct 25, 2018 at 11:00 AM

Ps. If someone from the community here can help you, not Victron staff, then ofcourse by all means.

The main goal for this site is to have community help community side by side with Victron staff monitoring & helping where needed.

2 |3000

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

Answer 5 · 2018-10-25T17:26:01Z

mihair answered · Oct 25, 2018 at 05:26 PM

"I think the 52.8 volt charge voltage setpoint, which is transmitted to our system by the BMS, is too low. But that’s just a guess."

Today I disabled DVCC (with the assumed risk of having uncontrolled charging current from BlueSolar MPPT) then reseted the ve bus and waited for main AC coupled Solarmax PV inverter to fully charge the battery. When the SOC has reached 100% everything went well meaning that Multiplus has been instructed by the REC-BMS to stop charging by modifying the voltage from 55,2V (3.45V/cell) to 52,8V (3.3V/cell). At that very moment the only current flowing into battery was from BlueSolar MPPT charger.

Few minutes after I enabled again DVCC (without modifying default settings). Instantly, Multi started to discharge the battery at full power and current from MPPT went to almost zero.

When the battery has reached 95% SOC ( 5% SOC is programed hysteresis into BMS software) the BMS instructed the Multy to stop discharging by modifying the charging voltage from 52,8V to 55,2V

Question, please: What is the correct stop charging voltage value for making DVCC working correctly and without triggering discharging at 100% SOC?

Thank you so much for your support!

ps. If this is going to work you can include REC-BMS into your supported batteries. This is a very well engineered product, allowing the owner to customise every important parameter.

@mvader

2 |3000

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

Answer 6 · 2018-10-26T07:10:39Z

mihair answered · Oct 26, 2018 at 07:10 AM

@AvB

Thank you so much for your comment.

I contacted REC and they responded very quickly with possible resolution. I will test it over the weekend and come back here with results in order to help others in the same situation

2 |3000

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

Answer 7 · 2018-10-29T10:10:21Z

mihair answered · Oct 29, 2018 at 10:10 AM

Hi,

Problem solved. Setting the end of charge voltage hysteresis at 0,05V solved the problem.

There is a small current dip at the end of charge but it won't last more than 10 seconds (picture attached)

Thank you so much for your support!

Mihai

rec-bms-ok.jpg (154.8 KiB)

3

2 |3000

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

boekel ♦ commented · Oct 29, 2018 at 10:25 PM

Hi Mihai,

Please note that with the balance voltage settings in that picture, the battery never really get's balanced.

Balancing starts now when average cell voltage is 3,39 volts - and battery is being charged, but due to charging stopping when a cell gets to 3,4V this is never reached.

Change to 3,3V to be effective.

The REC balancing is a bit weird, it only balances when charging (start voltage), or when above x volts it balances all the way till back at x volts (end voltage)

Due to LFP SOC curve, there is not much time when it can be balanced during charging..

1 ·

mihair boekel ♦ commented · Oct 30, 2018 at 07:43 AM

Hi Boekel,

Thank you for your concern. I'm aware about the fact that BMS, as it is configured now, doesn't balance my pack. In fact, I don't really want automatically top balancing because this process will keep the battery for very log time in upper knee of the charging curve where you actually slowly destroy the cells . I did it initially manually and I will do it again once per year.

Kind regards,

Mihai

0 ·

mvader (Victron Energy) ♦♦ commented · Oct 29, 2018 at 10:24 AM

great! thank you for letting all here know.

0 ·

Answer 8 · 2018-10-31T12:35:59Z

mihair answered · Oct 31, 2018 at 12:35 PM

Good day,

Observing closely the ESS system behavior I stumbled over the following situation which I do not consider it as optimal for battery longevity.

After the REC-BMS send the end of charge signal through CAN bus to CCGX by lowering the Max Charge Voltage and Charge Current Limit to 53,6V respectively 1,1A, Multi and MPPT devices still continues to overcharge the battery as you may see on the pictures below. Also, sometime the MPPT output is lowered at 5W even there is plenty of sun with no PV shading. The export setting in ESS is enabled

Is this a normal ESS system behavior?

Thanks,

Mihai

@mvader (Victron Energy Staff)

still-charging.jpg (71.5 KiB)

mppt-stopped.jpg (97.8 KiB)

2 |3000

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

question