Hi floating system voltage when BMS disconnects battery - what to do?

This has happened only a few time, but is disconcerting:
When my LiFePO4 battery is fully charged and no load on the system at all, the DALY (common terminal) BMS appears to disconnect the negative terminal of the battery from the vehicle negative/ground. As it probably should.
However, the MPPT charge controller then has no voltage reference any more and goes to an "imaginary" 24V expectation. That puts the system voltage (remember, no load) at the open circuit voltage of the solar panels (21.4V or thereabouts).
Some of my LED lights use so little current they individually don't seem to trigger the BMS to reconnect, since the solar panel cover their current draw easily. That means the lights then run at 21V - and they don't like it!
Worse still: my fridge has a voltage sensor to protect the compressor circuit. It senses 21V and considers that overvoltage - and won't let my fridge run! I can overcome it manually when I'm there by switching on a fan. The Voltage immediately drops below battery voltage and the BMS reconnects the battery negative. Then the MMP charger "remembers" that the battery is 4S (13.6V) and tunes to provide the right voltage and current.....
Should I put a small bypass resistor across the BMS, or maybe a diode or some other component? What other options are there?
The root cause seems to be that everything on the vehicle is grounded to the chassis and to each other - except the negative battery terminal - which is controlled by the BMS....
Sorry for the long-winded post.
Cheers,
Joe

What is the top charging voltage of the MPPT?
What are the battery and cell over volt settings on the DALY?
Does DALY give the error reason for the disconnect?

time2roll said:

What is the top charging voltage of the MPPT?
What are the battery and cell over volt settings on the DALY?
Does DALY give the error reason for the disconnect?

Click to expand...

Set to 14.4V - it's an auto sensing 12/24V MPPT charger. So it seems to ignore my setting when it can't "see" my battery.

I have no idea what the settings of the DALY are. It is not a "Smart" BMS. I can't communicate with it. Factory settings, I guess.

In an orthodox system the BMS should only trip under extraordinary circumstances.

Initially I think you could be over volt at the cell level. Unless you find the specs I would assume 3.650 as the max cell voltage. I would have my voltmeter on the cells as you approach 14.40 volts. Perfect match puts you to 3.600 and is so very close. If this is the issue consider dropping the MPPT to 13.8 to 14.2 zone.

time2roll said:

Initially I think you could be over volt at the cell level. Unless you find the specs I would assume 3.650 as the max cell voltage. I would have my voltmeter on the cells as you approach 14.40 volts. Perfect match puts you to 3.600 and is so very close. If this is the issue consider dropping the MPPT to 13.8 to 14.2 zone.

Click to expand...

Yep, battery voltage is 14.40V Dead on. The BMS hasn't "tripped", it just went to 0A in or out. Which effectively disconnects the battery from the ground.

JHovel said:

Yep, battery voltage is 14.40V Dead on. The BMS hasn't "tripped", it just went to 0A in or out. Which effectively disconnects the battery from the ground.

Click to expand...

I think you might be describing the charge fets "tripping".
The bms is common port but there are 2 current paths internally.
Sounds like its not a "smart" bms, confirm?

JHovel said:

Yep, battery voltage is 14.40V Dead on. The BMS hasn't "tripped", it just went to 0A in or out. Which effectively disconnects the battery from the ground.

Click to expand...

"0A out", are you saying your loads also get disconnected?

OK, so the specs for this "dumb" BMS are: 3.7V (x 4) overcharge disconnect, 3.6V (x4) charge reconnect - on paper.
In reality, I think it disconnects at 3.65V and reconnects at 3.55V (x4).
@smoothJoey: There are not normally any loads at all. When the fridge runs (4.8A) or the lights are on (1- 4.5A) or the waterpump is running (8-12A) or the hot water service is running (30A) everything is exactly as it should be and working really well!
The problem only occurs when the battery is fully charged and nothing is on. Then the "house voltage" shows solar panel open voltage (21V+ as I mentioned) and the fridge won't start due to "overvoltage" protection.
Somehow I have to give the system a continuous load when the battey is fully charged and there is solar power to keep the MPPT "busy".... I could experiment what the minimum load needs to be for the MPPT output to stay below 14.6V.... maybe?

One additional point/confession: All my loads are connected to the battery, not the MPPT 'load' terminals. Maybe that is an issue. Perhaps I put the fridge on the MPPT's 'load' terminal....

JHovel said:

OK, so the specs for this "dumb" BMS are: 3.7V (x 4) overcharge disconnect, 3.6V (x4) charge reconnect - on paper.
In reality, I think it disconnects at 3.65V and reconnects at 3.55V (x4).

Click to expand...

That is the thing about a dumb bms.
The only way to really know what its doing is to test it.
I recall the dumb Daly units had both their high and low triggers set out of spec.
Do you have a way to monitor the individual cell voltages while the battery is in service?

JHovel said:

@smoothJoey: There are not normally any loads at all. When the fridge runs (4.8A) or the lights are on (1- 4.5A) or the waterpump is running (8-12A) or the hot water service is running (30A) everything is exactly as it should be and working really well!

The problem only occurs when the battery is fully charged and nothing is on. Then the "house voltage" shows solar panel open voltage (21V+ as I mentioned) and the fridge won't start due to "overvoltage" protection.

Click to expand...

Do you have a shunt based battery monitor?

JHovel said:

Somehow I have to give the system a continuous load when the battey is fully charged and there is solar power to keep the MPPT "busy".... I could experiment what the minimum load needs to be for the MPPT output to stay below 14.6V.... maybe?

Click to expand...

I've not heard of an solar charge controller that needs to be kept "busy".

JHovel said:

One additional point/confession: All my loads are connected to the battery, not the MPPT 'load' terminals. Maybe that is an issue.

Click to expand...

That is generally considered a good thing.

JHovel said:

Set to 14.4V - it's an auto sensing 12/24V MPPT charger. So it seems to ignore my setting when it can't "see" my battery.

I have no idea what the settings of the DALY are. It is not a "Smart" BMS. I can't communicate with it. Factory settings, I guess.

Click to expand...

14.4 is 3.6V per cell. If any of your cells are runners, out of balance with the rest of the pack, you will get an over voltage disconnect. I would set to 13.8V on your SCC, and avoid the issue. You won’t get the pack up to 100%, but it will be plenty for day to day operation, and missing a few amp hours is better than not being able to run appliances due to BMS disconnect.

You are running a 12v nominal system.

Sounds like the issue is your MPPT losing it’s reference when the BMS cuts off further charging, now thinking it is a 24v system?

Many MPPTs have an auto system voltage sense which sounds like the problem to me.

Also, many MPPTs have a setting to override the auto sense and can be set to 12v nominal at all times, effectively turning off the auto sense.

Can you post a link to the manual of your MPPT or provide exact make and model?

JoeHam said:

You are running a 12v nominal system.

Sounds like the issue is your MPPT losing it’s reference when the BMS cuts off further charging, now thinking it is a 24v system?

Many MPPTs have an auto system voltage sense which sounds like the problem to me.

Also, many MPPTs have a setting to override the auto sense and can be set to 12v nominal at all times, effectively turning off the auto sense.

Can you post a link to the manual of your MPPT or provide exact make and model?

Click to expand...

That hits the nail on it's head!
I'll check the SCC manual to see if that option is there.
Mine looks like the 40A one here: https://www.ebay.com.au/itm/392824462054 I'll check the exact model tomorrow to confirm.

smoothJoey said:

Q1. Do you have a way to monitor the individual cell voltages while the battery is in service?

Q2. Do you have a shunt based battery monitor?

Q3. I've not heard of an solar charge controller that needs to be kept "busy".

Click to expand...

A1: Yes, I can access the battery cells easily and measure their voltages with my multi meter regularly. So far I have not seen any voltage differences of more than a few 10th of volts.

A2: Yes, I have a wireless shunt monitor connected all the time. It shows the voltage between battery positive and system ground/negative, as well as charge and discharge current and battery temperature (as well as calculated values from those 3 measurements).

A3: I used the wrong term: the charge controller needs a "reference" voltage to stay in 12V mode.

JHovel said:

A1: Yes, I can access the battery cells easily and measure their voltages with my multi meter regularly. So far I have not seen any voltage differences of more than a few 10th of volts.

A2: Yes, I have a wireless shunt monitor connected all the time. It shows the voltage between battery positive and system ground/negative, as well as charge and discharge current and battery temperature (as well as calculated values from those 3 measurements).

A3: I used the wrong term: the charge controller needs a "reference" voltage to stay in 12V mode.

Click to expand...

I hypothesize that your solar charge controller loses its voltage reference because the bms is disconnecting.
From your description I think the most likely cause for the bms to disconnect is cell over-voltage.
Pack over-voltage seems unlikely.
As others have already suggested configuring the charge voltage lower may stop the disconnect.
Is that worth a try?

JHovel said:

A3: I used the wrong term: the charge controller needs a "reference" voltage to stay in 12V mode.

Click to expand...

A supercap may fix your problem by filling in the gap before your bms reconnects.

Super-Supercaps

This is really quite fascinating. For a layman like me - not quite in terms of electronics, I understand capacitance and Farads - would there be a way to... equate, not quite convert, hundreds of Farads (hey, huge numbers, if you're used to micro and milli at best) to Ah? I mean (just curious...

diysolarforum.com

I have a similar over voltage problem on a multi pwm/ very old FLAs caused when little load coincide with high input shutting down the inverter at 15v momentarily but enough to cause the pc to reboot.
Fixed by the following- VHM-009 over voltage battery protectors switching on some 50w halogen globes above a fishtank @14.6v off @12.5v
- other over voltage sensors disconnecting the pv input to 2 pwms via NC relays on @14.7v and 14.9v
The above fixed the problem 99%. The addition of a supercap fixed the last 1%

When I eventually add LiPO I will probably just change the V settings and add more super caps.

I disconnected the battery for a few days and found that one of the cells is indeed over voltage.
I did a lot of reading in the meantime and learnt that my Daly BMS actually doesn't balance the battery. It just monitors and protects. Well, it does actually passively balance at "up to 30mA" which means it would take years to balance.....
So I've now ordered an active balancer that can balance between cells at up to 5A.
I had no idea they were available or even necessary for larger batteries.
I suspect that might fix it permanently.
I'll manually balance the cells now and update this topic as I learn more.

Active balancer alone could still take a few weeks or more depending on capacity and time spent out of balance.
If possible I would turn the charge voltage down below the point of disconnect.

[SIZE=4]JHovel[/SIZE] ,I understand what trouble you are meeting.​

When the battery full charge to 14.4V, your battery goes to protection automatically and does output any DC .We suggest you to set your charger controller less than 14V , Do not trigger the battery protection system.

Will you use a DC fridge?

Alice2021 said:

[SIZE=4]JHovel[/SIZE] ,I understand what trouble you are meeting.​

When the battery full charge to 14.4V, your battery goes to protection automatically and does output any DC .We suggest you to set your charger controller less than 14V , Do not trigger the battery protection system.

Will you use a DC fridge?

Click to expand...

I have reduced the charge control to 14.2V on the basis of the suggestions here.
And yes, I am using a DC fridge.