Random cutoff

[Harold]

Hello Dacian.

I have some random cutoff power on my SBMS120 : It happened at least 3 time this morning.
I noticed it as my computer suddenly switch off and rebooted. Same for my internet router.
It seems to only happen on the SBMS output, but not on the external load output (or cutoff is too short to be detected by the inverter on external output).

I've noticed this could happen when at the end of charging battery.
So, I was looking at the monitor 3/5 page of the SBMS120, and was quite lucky that the phenomenon had occurred just at that moment.

Here is what I've seen :
Everything was working ok.
Only CFET, EOC and DFET were on, batteiy level was showing 100%
Then it suddenly change to CFET off, OV on, and battery level to 99%. Just at that instant, the computer and internet router rebooted !

I've then taken a video where you can see how  Error_Status value changed

dec 20480 / bin 101000000000000

dec 28672 / bin 111000000000000

dec 08321 / bin 010000010000001   cutoff has happened at that moment ! you can notice from the video where the computer stopped playing the sound in the background!

dec 24577 / bin 110000000000001

I think that this already occurred before but I wasn't close enough to catch the phenomenon.

Here is the link to the video :

https://www.dropbox.com/s/wdgvfzuw3cklpdy/IMG_2907.MOV?dl=0

In advanced settings,
Over Voltage = 2.55V
End of Charge = 2.53V

I don't think they've been modified.

I was only drawing about 3A continuous on the SBMS120, and the bread cooker on the inverter intermitently 28A on external load.

[Dacian Todea]

Harold,

8321 contains DSC but that is likely because it was trying to reconnect. Guess Short circuit current is set at the default 576A

Maybe there was an error that cleared much faster so was not catched in the video so also check the cell balance max dV and increase that to something large say 3999mV (of course this will require a power cycle to become active) then see if this ever happens but I can not think on anything else.

I assume that in particular circumstances when charging is stopped (should only happen at end of charge) a load is connected at exact same time and so there is a super fast transition probably showing some cell voltage way outside the current limit for max dV

I'm 80% sure this is the problem and to confirm change the cell balance max dV to 3999mV as that should prevent this sort of problem from occurring.

[Harold]

Hi Dacian.

I've changed the cell balance max dV to  3999mV. (max possible value is 4800mV)

But same cutoff happens at end of charge !

Average load was about 3A.

In parameters :

Over Voltage =3.55V

Over Voltage Delay =1s

Over Voltage Recovery = 3.40V

End of Charge = 3.53V

I'm a little confused about "Over Voltage" and "End of Charge ".

Why "Over Voltage" would be reach if  "End of Charge " has triggered ? 

It seems that SBMS120 is still charging when "End of Charge" is reached. So, it then reach the "Over Voltage" limit that  temporarily cutoff the DFET at same time CFET is switch off.

Is that normal ?

I have SBM120 version hard : 0.3d, soft : 5.0d

I've upgraded the firmware to 5.0d to get the Robert's software features. But I think I noticed that strange cutoff before.

[Dacian Todea]

End of charge is just a flag that takes no actions the Over voltage is where the charge fet will be turned OFF so if any cell is 3.55V for more than 1 second. The discharge FET should not be affected is likely due to other flag like cell fail due to large cell delta when turning OFF charging.

Have you power cycled the SBMS120 after you set the max cell balance to 3999mV ?  If you did not fully powered OFF the SBMS120 by removing the 10pin connector then the new setting is not in effect.

[Harold]

Yes, a complete power cycle has been done by removing  the 10 pins connector.  And I've checked it again, it still says 3999mV.

The cuttoff happens right at the moment the OV flag is turned ON. I'm absolutely sure of that.

And what  the "Cell Balance Max dV " has to do there ?

I mean, there's also a "Cell Balance Min dV "  (set at 11mV). 

How do these parameters behave ?

If the "Cell Balance Max dV " would cutoff the DFET if limited value would be reached, would the "Cell Balance Min dV " also cutoff DFET if any deltaV is below 11mV between cells ? Il think this is a non sense.

What are these limit for ? How do they behave ?

And what put you on that track ?

Have you noticed the strange behaviour of #2 Cell at 47' of the video ? It goes from 3.435V to 3.510V for no apparent reason just before the cutoff. 

[Dacian Todea]

The max cell balance voltage if that limit is exceeded even for a few ms (not noticeable on display) a error will occur and both charge and discharge fet will be turned OFF. The cell min dV has no influence on charger or discharge fet that affects only when cell balancing is enabled.

But if this did not fixed the problem then there is another error that triggers this maybe overcurrent limit but that should not be the case if you have the default values.

Yes that is normal what happens at second 47 in the video as cell balancing was just enabled on cell 1 and cell 3 but not on cell 2 as that was the lowest cell voltage and as cell balancing current is enabled the cell voltage measurements are not real due to voltage drop on cell monitoring wires that is why cell balancing is enabled for a few seconds then disabled for another few seconds (likely 3 seconds on and 3 seconds OFF if you have the latest firmware).

You where in the first monitoring screen so not sure what error triggered the fault as the Stat only updates once a second and some flags clear much faster than that but if you were in monitoring page 3 when that happened is likely I could see the error flag and know what created the problem.

Looking closely at the video I see at one point 8321 that means Discharge short circuit so likely that disabled the DFET but not sure why that will be the case if you have the default settings for discharge short circuit current and discharge short circuit delay.

[Harold]

Despite having the last firmware :

Cell Balance ON = 6s

Cell Balance OFF = 2s

Are you sure upgrading to last firmware would modify any parameters ?

I don't remember having to re-enter parameters after upgrading to make my SBMS120 working with new firmware.

Very strangely, when checking the Over Current Charge, Over Current Discharge and ShortCircuitCurrent, they would show the default value, but when pressing the OK button, they would appeared to be 18, 36 and 144A, I think the last values they were programmed before the upgrade.

I've changed them to the max values possible.

From Over Current Charge to Short Circuit Delay, I now have these parameters respectively :144A, 250ms, 144A, 60ms, 576A, 100us.

We'll see if that makes any difference.

I wonder how it would be possible to trigger the ShortCircuitCurrent at 144A when average load was less than 2A, and no other loads to be connected at that moment ?

[Dacian Todea]

6 and 2 seconds is fine for the SBMS120. Absolutely unrelated to your problem.

It is very possible to get to 144A when charging is turned OFF as battery will need to take over the Load depending on what type of load you have connected (even if not powered ON if it has large capacitors).

[Harold]

The only big capacitor bank are on the inverter but that's external load !

Otherwise, I have mainly 3W to 10W DC/DC buck or boost converter to power each items (maybe 7 or 8 units).

Anyway, we'll see with the new settings at the max 576A.