diy solar

diy solar

Sol-Ark 15K + SOK batteries = Weird SOC issue

Watts Happening

I call it like I see it.
Joined
May 3, 2022
Messages
819
Generally speaking I'm incredibly happy with my Sol-Ark and SOK batteries, I've just run into one issue and currently can't point the finger either direction so I'm at a bit of an impasse. Hoping to get some valuable feedback from other members with the same setup, or @HighTechLab (I've got a call in to your tech support, no rush, thought this may be easier). *edit* got a call back from Current Connected and certainly doesn't sound like it's a battery thing, but the question remains.

Basically I can't get my Sol-Ark to discharge the batteries below ~19-20% unless I disable closed loop communication. I was just on the phone with Sol-Ark going over it and essentially they ran out of ideas. If I take the inverter out of closed loop communication (unplug the comms wire) it will continue discharging, however with it plugged in the inverter begins to use grid power right at the 19-20% SOC. It won't let the batteries drop any lower.

While this is a pretty minor problem to have, I'd like to A) understand what the issue actually is and B) be able to utilize a bit more of the battery as we're often beginning to use a bit of grid power for only an hour or so in the mornings, a slightly lower depth of discharge would eliminate that as well as eliminate the batteries becoming full earlier and solar having nothing to power if I'm not paying attention. We'll likely have an interconnect agreement with the power company soon, but in the meantime it's just wasted power once the batteries are full.

Sol-Ark is going to talk to their engineers and the tech said, "maybe the BMS has some form of setting that calls for a charge when it hits that SOC". Generally I doubt the BMS is doing that, and I would think the inverter would ""say"" - "I understand you want me to charge you, but I can see that you're at 19% SOC and I'm not supposed to charge you until you hit 10%".

Any ideas as to even the right direction to look here? I tend to imagine it's something in the inverter, but before I waste any time I thought someone may have some insight. Thanks!
 
Last edited:
Yep. Corrected my @ above ^^.
Updated battery FW is always a good idea as long as you use comms.
 
If i remember correctly, SOK does not really do updated firmwares. But I am interested to see what Dexter has to say.
 
If i remember correctly, SOK does not really do updated firmwares. But I am interested to see what Dexter has to say.
There is software listed, and I believe I once used it to do some diagnostic stuff on one of the batteries but I don't see any mention of firmware.

 
I know this is not Solark or SOK but the EG4 18k will still grid bypass and charge the battery even in off grid mode if the batteries get low. I call it the "oh shit" mode.
Perhaps something like this is built into your equipment and no one knows it.
 
I know this is not Solark or SOK but the EG4 18k will still grid bypass and charge the battery even in off grid mode if the batteries get low. I call it the "oh shit" mode.
Perhaps something like this is built into your equipment and no one knows it.
That’s part of my suspicion, but again as soon as comms are removed from the battery, it works as expected and falls back on SOC parameters based on voltage. Has me confused to say the least
 
You might disable communications but you enable voltage control 😂.

Are any of the cells dipping below cutoff threshold? Maybe that is throwing the SOC off?
Can you download the BMS software and take a deep dive into the settings?
I haven’t gone that deep yet, truth be told it seems far too consistent to be something like that in my mind.

I asked Sol-Ark a similar question, whether a single battery could trigger charging or if it was an average and was told it’s based on the average, which makes sense because when I hit that ~20% SOC I may have one at 17% and another at say 22% with others in the middle. It doesn’t seem as though a single battery is triggering anything so I can’t imagine a cell would etc.

Plus, we’re a full 10% above the point at which it’s supposed to have the grid take over. If it were significantly closer I’d be more convinced.
 
This makes me think my post above about low SOC% setpoint might be the issue.
I’ll give it a look when I get home.

Even if that’s the case it then leaves me more confused with the inverter/battery communication and who’s really “in charge”.

Mainly just interesting
 
I haven’t gone that deep yet, truth be told it seems far too consistent to be something like that in my mind.

I asked Sol-Ark a similar question, whether a single battery could trigger charging or if it was an average and was told it’s based on the average, which makes sense because when I hit that ~20% SOC I may have one at 17% and another at say 22% with others in the middle. It doesn’t seem as though a single battery is triggering anything so I can’t imagine a cell would etc.

Plus, we’re a full 10% above the point at which it’s supposed to have the grid take over. If it were significantly closer I’d be more convinced.
A single cell going into ovp can cause this.
Looks to me there might be an cell imbalance that that pack

Disabling comms and going voltage mode is s very bad idea in my book

Comms is a safety feature and should, if the BMS's are properly configure also prevent a cell ovp by limiting the charge amps at the very top so the passive balancer can do it's job
 
A single cell going into ovp can cause this.
Looks to me there might be an cell imbalance that that pack

Disabling comms and going voltage mode is s very bad idea in my book

Comms is a safety feature and should, if the BMS's are properly configure also prevent a cell ovp by limiting the charge amps at the very top so the passive balancer can do it's job
I'm charging to top balance and then discharge to test this theory, but as it sits at ~93% SOC the largest imbalance I'm seeing on a single battery is 3395-3425mV aka 3.395 at the lowest and 3.425 at the highest. I don't know the individual cell voltages well enough to determine exactly how far out of balance that is, but my gut tells me that's not enough to trigger a low voltage issue?
 
I'm charging to top balance and then discharge to test this theory, but as it sits at ~93% SOC the largest imbalance I'm seeing on a single battery is 3395-3425mV aka 3.395 at the lowest and 3.425 at the highest. I don't know the individual cell voltages well enough to determine exactly how far out of balance that is, but my gut tells me that's not enough to trigger a low voltage issue?
That's definitely enough to cause an issue.
Still at the bottom of the curve vs climbing to the top.
 
Alrighty, I'll give her a solid charge and see if I can't get it back in line.
It will likely take a while based off my experience with my packs. You'll need to hold them at a voltage that will trigger balancing. I'm not sure what the SOK starts at, I'd guess ~3.42V/cell.
 
It will likely take a while based off my experience with my packs. You'll need to hold them at a voltage that will trigger balancing. I'm not sure what the SOK starts at, I'd guess ~3.42V/cell.
I'm pretty sure it's 3.45v, or 3.5v. I don't remember now, but I do remember thinking it was unusually higher than I would expect.
 
It will likely take a while based off my experience with my packs. You'll need to hold them at a voltage that will trigger balancing. I'm not sure what the SOK starts at, I'd guess ~3.42V/cell.
I suppose I can just unplug the comms cable to accomplish this and let it sit at 57.6v for a while?

I'm right at 99% state of charge, but what has me confused now is the charge current limit of 47A rather than what I would expect closer to 300A, any ideas how it would come up with that number? Discharge of 285A looks correct, there are (3) SOK batteries connected to the Sol-Ark, but the 47A number seems, uh, odd?

IMG_7676.jpeg
 
For now, I'll leave a chargeverter connected at 57v for a couple hours and see if that makes a difference cell voltage wise, however I'm still perplexed by the charge current limit as other threads indicate charge and discharge limits should be the same and are communicated from the packs. In my case it seems as though both should be 285A. I've CERTAINLY seen the Sol-Ark provide far more than 47A of charge to the batteries on many occasions.
 
I suppose I can just unplug the comms cable to accomplish this and let it sit at 57.6v for a while?
I would think so. I wouldn't go above 56.8V. You may not be able to get that high if cells run at the top, drop down a little bit if needed.
I'm right at 99% state of charge, but what has me confused now is the charge current limit of 47A rather than what I would expect closer to 300A, any ideas how it would come up with that number? Discharge of 285A looks correct, there are (3) SOK batteries connected to the Sol-Ark, but the 47A number seems, uh, odd?
Not sure on that one. I'm running open loop. Is there a tail current setting in the Solark? I'm not familiar with them.
 
SOKTool will tell you everything you need to know about the SOK battery (Rack mounted).
You will need an RS232 to USB Cable.
Note the "Balance Threshold(V)" voltage.

A higher charge voltage is OK until the battery gets to about 90% SOC then you will want to reduce the voltage to less than the 57.6 "Pack OV Alarm(V) voltage. Monitor the cells and if cells start getting close to the 3.6 Cell OV Alarm(V) you will want to further reduce charge voltage. You can monitor each cell (Realtime Monitoring tab) and watch as cells reach the balance threshold and start to balance.

The BMS will often times report 100% SOC even thought the battery is really not fully charged. The battery will be FULLY charged once you get cells to start balancing

sok.png


SOKTool

USB-RS232 Cable
 

diy solar

diy solar
Back
Top