Diagnosing charge issue in first cold week

[ezwryder]

I’m trying to understand if I have a problem or if my system is operating as expected. In brief, I have two SOK 206 AH heated batteries in parallel charged by a Victron 100/30 solar charge controller. The system is installed in a remote cabin where temps this week have fallen below 32 F.

Today I noticed that the charging process was cut short. This morning the charge level was at 70%. On a normal sunny day it would take about 6 hours of full sun to generate the 125 AH needed to get them to 100%. However, it’s cloudy and the panels couldn’t have been putting out more than a few AH. After 2.5 hours the pack voltage went to the cutoff at 14.6 and the charger dropped into absorption then float mode.

I did not have a way to test the heaters when I put the system together, but IF they were working, I would not have expected the charge cycle to end prematurely. If the BMSes put the batteries in protect mode, would that fit what I’ve described?

I’m 1,000 miles away from the system so the only info I have is what I get from my Victron components. I can make changes to the solar charge controller if that seems warranted, but I am unsure what else I can do.

I would appreciate any thoughts/opinions on what might be happening. Thank you.

 

[MisterSandals]

Today I noticed that the charging process was cut short.

What info do you have to conclude charging was cut short?

 

[ezwryder]

To get from 70% to full, the system would need to put about 125 AH in the pack. The array produced less than 20 AH today which is about average for a rainy day. So I would have expected to see the pack at 73-75% at the end of the day. Instead, the battery voltage drove to its cutoff in just a couple of hours and the system floated for the rest of the day. I don't know if that answers your question, it's the best I know how to explain it.

 

[chrisski]

Determining SOC by voltage can be misleading because of the flat curve, but you’ve used this system for a while.

What you described with going to 14.6 and then going to absorb and float, makes me think you only used the 20 - 30 ah that got put into the batteries. If you had a heater fail, I doubt the charging voltage would get to 14.6 and still have 100 more ah to go, which is about 25% of the battery total.

How sure are you that you actually used 122 ah? You probably have some data of what an average day is + or - something. If 122 ah is typical, you have a well designed system to keep up with that much every day and be 1000 miles away.

 

[ezwryder]

Determining SOC by voltage can be misleading because of the flat curve, but you’ve used this system for a while.

What you described with going to 14.6 and then going to absorb and float, makes me think you only used the 20 - 30 ah that got put into the batteries. If you had a heater fail, I doubt the charging voltage would get to 14.6 and still have 100 more ah to go, which is about 25% of the battery total.

How sure are you that you actually used 122 ah? You probably have some data of what an average day is + or - something. If 122 ah is typical, you have a well designed system to keep up with that much every day and be 1000 miles away.

The AH in/out readings come off of the Victron shunt. The system averages about 80 AH consumption per day but the camera loads are timed to only come on during daylight hours, so solar output often supplies their power requirement. On cloudy/rainy/snowy days it's typical for the battery pack to drop 10% per day, and its been cloudy for about 3 days. So I am confident the 70% reading, corresponding with about 125 AH consumed, was accurate.

This is the first week the system has encountered sustained cold temperatures so I don't know what is typical and what is not. Prior to this week, I knew that if the shunt reported 125 AH had been consumed it would take 6 hours of full sun to recoup that. And I know that today, it was cloudy, the panels generated little output, and yet the system says "No mas." I would like to understand better what drove the battery pack to the cut off point when it seems to me they could have taken a lot more charge.

 

[chrisski]

The AH in/out readings come off of the Victron shunt. The system averages about 80 AH consumption per day but the camera loads are timed to only come on during daylight hours, so solar output often supplies their power requirement. On cloudy/rainy/snowy days it's typical for the battery pack to drop 10% per day, and its been cloudy for about 3 days. So I am confident the 70% reading, corresponding with about 125 AH consumed, was accurate.

This is the first week the system has encountered sustained cold temperatures so I don't know what is typical and what is not. Prior to this week, I knew that if the shunt reported 125 AH had been consumed it would take 6 hours of full sun to recoup that. And I know that today, it was cloudy, the panels generated little output, and yet the system says "No mas." I would like to understand better what drove the battery pack to the cut off point when it seems to me they could have taken a lot more charge.

I don’t know why it is out of whack.

I know my Victron Shunt is real accurate, but can drift slowly, but nothing like you’re seeing. I saw mine away from reading 100% and 0 ah remaining to 98% and 7 ah remaining at full SOC until I got my settings dialed in.

This is how my shunt is set, and is I consider this dialed in for my system.

I think you have equipment problems or a friendly neighbor charged your batteries.

 

[MisterSandals]

This is how my shunt is set, and is I consider this dialed in for my system.

I was just working with someone yesterday and their threshold was 2A, i suspect it is the default value which seemed high to me when i saw it.

How did you arrive at .1A and what are the downsides of using a low threshold? I cannot imagine any downsides, but willing to learn.

 

[chrisski]

I was just working with someone yesterday and their threshold was 2A, i suspect it is the default value which seemed high to me when i saw it.

How did you arrive at .1A and what are the downsides of using a low threshold? I cannot imagine any downsides, but willing to learn.

When I installed my lithium batteries, I was getting what I described above as the voltage went up and current stopped, so I adjusted some settings, the current threshold was one of them. Between that current threshold, tail current, charge efficiency factor and charged voltage, I now have an accurate reading I have not had to reset on dozens of uses versus when it was slipping 1% per day.

Do not know which setting or which combination of settings fixed my problem.

————

For those not aware charge current threshold defined by Victron is

“Current threshold. When the current measured falls below the “Current threshold” value it will be considered zero. The “Current threshold” is used to cancel out very small currents that can negatively affect the long-term state of charge readout in noisy environments.”

 

[ezwryder]

I don’t know why it is out of whack.

I know my Victron Shunt is real accurate, but can drift slowly, but nothing like you’re seeing. I saw mine away from reading 100% and 0 ah remaining to 98% and 7 ah remaining at full SOC until I got my settings dialed in.

This is how my shunt is set, and is I consider this dialed in for my system.

I think you have equipment problems or a friendly neighbor charged your batteries.

I think the shunt is okay. I think it is doing what it's supposed to do which is to reset to 100% when I sees the charged voltage hit the preset.

What I don't understand is why the voltage drove so high so quickly. That is what caused the solar charge controller to back off and I'm trying to understand if that's something the battery BMSes caused to happen or if there is another issue.

I guess if it doesn't resolve in a couple of days I will just have to make a trip back there. "Honey, darn the luck. I have to go back to the cabin to flip a switch. See ya in a couple of weeks."

Thinking of it that way, I can definitely see an upside to this issue.

 

[chrisski]

Only thing I see is the lithium battery should have a 1 for a Puekert exponent since it’s not lead acid and have different total kWh ratings based at a high versus low load.

That would not cause what you’re seeing.

 

[ezwryder]

After a couple of days of cloudy cold weather, the area where the cabin is located had a sunny day. Still cold, but sunny enough to warm up the "winter room" of the cabin to 60F. In the late afternoon near the end of the charging day, suddenly the batteries started charging. Since the room warmed up and stayed warm for several hours, this seemed to mean to me that the cold temp cut off of the BMS was working but the batteries had to rely on ambient room temperature to get back up to a safe charging temp, so the heating pads were either not working or ineffective.

This morning the cabin temp was 25F and at first full light, the same pattern began to repeat with the battery pack jumping up to max and the solar charger going to float. But then, about 30 minutes later, I noticed the system had switched back to bulk and was charging at full output. That seems to suggest that both the cold temp cutoff and the heaters are working as they should. That's great, but I just don't quite understand why that didn't happen with the first period of cold weather and it still leaves me a little unsure about how the system will perform over the rest of the winter.

 

[ezwryder]

both the cold temp cutoff and the heaters are working as they should. That's great, but I just don't quite understand why that didn't happen with the first period of cold weather and it still leaves me a little unsure about how the system will perform over the rest of the winter.

I think I have figured this out. When it turned really cold, it also was very cloudy for the first few days, all day. When it's cloudy/rainy, my panels only generate about 40 watts/hr. My system overhead (router, antenna, Victron components) is about 30 watts. In short, that did not leave enough power to operate the heaters. On about day 4, when it was still cold, but sunny, the system began to charge.

 

[chrisski]

I think I have figured this out. When it turned really cold, it also was very cloudy for the first few days, all day. When it's cloudy/rainy, my panels only generate about 40 watts/hr. My system overhead (router, antenna, Victron components) is about 30 watts. In short, that did not leave enough power to operate the heaters. On about day 4, when it was still cold, but sunny, the system began to charge.

Sounds good. Do you feel like going to the cabin to fix this??

I use excuses like this to test my RV for a few days. I drive 34 minutes one way to boondock and use about 10 gallons of diesel for my tests. You have 1000 miles to go.

 

[ezwryder]

Sounds good. Do you feel like going to the cabin to fix this??

I use excuses like this to test my RV for a few days. I drive 34 minutes one way to boondock and use about 10 gallons of diesel for my tests. You have 1000 miles to go.

Um, yes, actually! I am going up in two weeks to check it out, then stay for a few days. I do really think there is something not quite right with the way the recharge is happening and I want to get to both of those batteries and check them out. Plus I forgot to put the brake on the windmill when I last visited, there are a couple of gates that did not get closed properly, etc., etc.