question

ivo avatar image
ivo asked

ESS not usable for hot batteries? (FIAMM-SoNick)

We have an installation with many hot batteries connected in parallel.
We are experiencing the following problem with ESS "optimized (with battery life)":
The system cannot hold its dynamic SOC low-limit, it drops way below it.

Could it be that ESS has a problem with the self-discharge current of the hot batteries?

The batteries slowly self-discharge through a heater to keep themselves warm. This would explain, why we are able to run ESS successfully on other installations with fewer batteries. Too many batteries and the self-discharge current would become bigger than ESS' low charge current?

In particular, from the wiki:

6: SoC has been below SoC limit for more than 24 hours. Charging with battery with 5amps

5 amps would definitely not suffice to keep more than 3 batteries warm.

Is our assessment of the situation correct?
If yes, is there a way to adjust the #6 state's charging current?

Sincerely

Ivo Graber

ESS
2 |3000

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

5 Answers
mihair avatar image
mihair answered ·

Hi @ivo


Where is the DC shunt installed, if there is any, or your SOC metering device?

In order to have an accurate SOC you should also meter the current consumed by the battery heating system.


Also, you should activate "Has DC system" and provide a constant current power supply on DC bus equal with your heating system consumption


Kind regards,

Mihai

2 |3000

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

ivo avatar image
ivo answered ·

Hi @MihaiR

thanks for your prompt reply.

> Where is the DC shunt installed, if there is any, or your SOC metering device?

There is no external shunt. The BMS does all the measurements (SOC, current, voltage etc.) and provides them to Venus. It also controls the heating. There is no external control for that.

I have activated "Has DC system" but cannot see any effect. If I understand the manual correctly, what this does is add an additional "box" to visually indicate the difference between the DC current measured by Venus (VE.bus) and the one measured by the BMS. This would indeed be handy to quickly check the power consumed by the heaters. However I cannot find said box.
In any case, this presumably wouldn't solve our problem.

> provide a constant current power supply on DC bus equal with your heating system consumption


This is not practical.
The heating is variable, it depends on ambient temperature and on battery usage patterns, so it would be problematic to use a constant-power charger.

Since we already have a controllable charger (a Quattro), it guess it should be involved in a solution to our problem with ESS.

This -of course- if the problem we experience is indeed caused by self-discharge.
Can you, or somebody else, confirm or reject that hypothesis?


regards

Ivo


3 comments
2 |3000

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

mihair avatar image mihair commented ·

You have a power theft into the system. I can confirm your hypothesis.

How can be solved?

1. Either Victron modify the slow charge current in order to compensate with the power taken by the heater. This will not be ok for other battery chemistry, ex. Pb based

or

2. Add a constant current supply (CCS) to DC bus. The current must be at least equal with maximum heater current consumption. When the heater will take less than maximum current supplied by the CCS an the battery is full then the Multi will inject the power into the grid

0 Likes 0 ·
ivo avatar image ivo mihair commented ·

Thanks for the confirmation.

Since this problem can be solved in software, I think it should be. More hardware means more costs, more power losses and more stuff that can break. So we are looking for a solution based on option 1.

There must be a parameter to adjust the slow charge current somewhere in the Victron source. Something like that cannot possibly be hard-coded, right?
I have sifted through the code of dbus_systemcalc.py and batterylife.py but failed to find it. Does anybody know?
Should I ask on GitHub instead?

Sincerely

Ivo

0 Likes 0 ·
mihair avatar image mihair ivo commented ·
0 Likes 0 ·
mihair avatar image
mihair answered ·

Hint for building a suitable CCS for Victron ESS:

Use a high quality bench power supply capable to limit both current and voltage and supply the output to a Bluesolar MPPT controller connected to DC bus. Set the bench power supply current limit accordingly with the max heater consumption and voltage below max MPPT input voltage

Obviously, you must connect the MPPT to Venus device :-)

2 |3000

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) avatar image
mvader (Victron Energy) answered ·

It looks like indeed; the BatteryLife option is not suitable with that battery type.

To then conclude that all of ESS doesn’t work is premature.

Sustain will do it; it works on voltage. See ESS manual for how sustain works.


To answer the more software developer question;

The 5A parameter is probably in the hub4control source code; which is closed source, and C or Cpp. I’ll send you an email about it; Ivo.

2 |3000

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

ivo avatar image
ivo answered ·

Hi Mathijs,

> Sustain will do it; it works on voltage. See ESS manual for how sustain works.

Thanks, I will look into this and report back.

> To then conclude that all of ESS doesn’t work is premature.

Well, I admit the title chosen is click-baity.
But it got your attention, right :)
cf. Betteridge's law.

Sincerely
Ivo

2 |3000

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