The C rating that is listed for the 280a cells is at the individual cell and not the pack. So basing it against the overall pack size (1120ah) would be incorrect I believe since the voltages are not the same, 3.2 vs. 51.2v. You need to divide the overall charge by the number of individual cells that make up the battery since that is the amount of charge each cell would "see". Can be confirmed by measuring the current with a meter clamped over a bus bar (if it fits). ex. a 300a total charge rate from the sol-ark does not mean each cell is seeing 300a but a fraction of that based on the number of cells. 300/64 = 4.7a. You use that amount to calculate the c-rate that each cell is seeing, 4.7/280 or .017C. To put it another way you would need a charge of 17920a to reach the 1C rating that these batteries can support (280*64) !
Sol-Ark - Disconnect Strategy?
- Thread starter solardad
- Start date
The C rating that is listed for the 280a cells is at the individual cell and not the pack. So basing it against the overall pack size (1120ah) would be incorrect I believe since the voltages are not the same, 3.2 vs. 51.2v. You need to divide the overall charge by the number of individual cells that make up the battery since that is the amount of charge each cell would "see". Can be confirmed by measuring the current with a meter clamped over a bus bar (if it fits). ex. a 300a total charge rate from the sol-ark does not mean each cell is seeing 300a but a fraction of that based on the number of cells. 300/64 = 4.7a. You use that amount to calculate the c-rate that each cell is seeing, 4.7/280 or .017C. To put it another way you would need a charge of 17920a to reach the 1C rating that these batteries can support (280*64) !
Shawn Hayes
Learning Everyday
That seems logical but never seen that calculation before. I was using the method in my last post. That’s the only way I’ve seen c/rate calculations done on this site. But your calculation makes sense.
Yeah, I have a clamping meter but there is no way to get that around the bus bar to monitor the actual average going to each cell.
I know Batrium has telemetry and stats it records, so when I get more time with it I’ll see what data I can get.
Are you still get real time data through Batrium via R-Pi, (or something else) Grafana etc? If so, that and disconnect strategy will be my next projects.
Last edited:
I find the Batirum SOC reading (I use Batrium 4) is a lot more accurate that the Sol-Ark SOC but the other other stats like Volts and Current passing they are pretty much the same.
Yeah - Still using my RPI setup. Ironically the SD Card that I was using got corrupted a few months ago so my monitoring of my 12k was down. Batrium was still up since I use a different RPI for that. I just redid the setup this AM with a new SD Card and got it monitoring again. I use a RPI zero W for the node-red connection (reading the monitor) and then send the data to a RPI 4 that has the influxDB and grafana setup to store and display the data. I found having them all run on the RPI zero was a performance hit.
Note - RPI came out with an updated RPI zero that is something like 3x as powerful for $15. https://www.cnx-software.com/2021/10/28/raspberry-pi-zero-2-w-and-zero-w-features-comparison/
I have been somewhat out of the loop doing additional automation since everything just works great for my scenario as-is. I did check out the original site recently that provided all the initial instructions and assistance ( https://powerforum.co.za/ ) and they are now on to some pretty cool automation via Home Assistant that is all greek to me at the moment. I hope to dig into it when winter comes around.
Shawn Hayes
Learning Everyday
Well I’m still learning the Batrium, it’s control options, accurate readings (soc) as there are many ways to set and change things.
Not sure if mine is working (setup) differently than yours, but my state of charge display on Sol-Ark comes directly from the Batrium. Also, in the Sol-Ark battery settings, the absorption, float numbers are the same. I can change them in Sol-Ark, but they will change back to whatever is set in Batrium. That number in Batrium, whether I set it at 53.8 or 54 V, in (->remote settings, -> Charge targets, normal voltage) is white sol-ark shows as my battery charge targets as absorption, float etc. How do you get your absorption and float set differently, it appears my Batrium makes that number identical unless I have something set incorrectly? See photos, thoughts?
I find it interesting that you are using two raspberry pi too transfer data. I had purchased a raspberry pi four and struggled with getting the information from Sol-Ark to the raspberry pi and making it work. So I guess in the future I will revisit that. I will have to look into the newer raspberry pi zero you linked to.
Hope I can get that working too.
I think a disconnection strategy is probably most important.
I am seeing the same behavior, voltage settings being the same on the inverter side when BMS is hooked up. Not concerned since the inverter would be controlled by the BMS.
I ended up doing some testing this AM and confirmed a few things and are happy with the results based on tests Batrium outlined on there site: https://wiki.batrium.com/en/integration/inverter-testing
The testing (from link) focuses on voltage but I also setup some parameters in my CHARGING tab, enabled Max Cell volt and Shut SOC %. Goal was to basically turn off any charging if certain limits were reached. I tested with the same method by setting values that were below the current levels of my pack and the Sol-Ark responded once that level was reached and again switched the values back and the inverter started the charge back up- pretty slick.
For reference my REMOTE settings:
So the big win now with this setup is any rouge cells that decide to spike the BMS will tell the inverter to stop charging until the cell voltage drops or can be burned off by the BMS. (I have yet to have that issue but having this in place provides another piece of mind). The Sol-Ark has done a great job in not overshooting the pack voltage on its own but now having it based on the SOC & pack voltage is another layer of monitoring and security.
I still have a question on the SOC% and equivalent voltage reading since this chemistry is tricky to map but I think i am close enough for now by staying well above/below the limits of 2.5 and 3.65v.
NOTE - the missing nominal _pack size value on the Sol-Ark side was highlighted as a scenario by others. the 'fix' was to turn off then turn back on the Battery LI option on the Sol-Ark side. Did not work for me but I am not concerned, similar to the matching voltages, everything is controlled by the BMS when that setting is enabled.
This is not correct. His calculation before was (360/(280*4)). Your calculation would only be correct if they were all in parallel.
Right... Thanks for the clarification. In this case he would be pushing a max of .32C which is still well below the 1C rating of the battery.
Shawn Hayes
Learning Everyday
Thanks for the clarification as well. I was pretty sure the calculation I was using was correct for my 16s4p set up. However it did have me second-guessing myself for just a minute. Based on the charge parameters that I have set in Batrium/Sol-Ark, and observing the rate at which my batteries are charging, I felt that my initial calculation was the correct one. However, being relatively new to a lot of this, will always make you second-guess yourself, sometimes.
Shawn Hayes
Learning Everyday
I think you and I have similar settings based on the two photos you posted above. Our settings are different but with a similar idea that if anything happens outside of the certain parameters, Batrium will tell Sol-Ark to stop sending charge.
As for the amp hour pack size, and Sol-Ark, I had the same result. No matter what I have changed, the amp hour battery display in the lithium battery screen, remains zero.
Shawn Hayes
Learning Everyday
So did you read my postings and have you got it sorted?
As you asked, I posted details of my experience here:
Sol-ark + batrium canbus integration workaround until its supported
as the title says i played around with integrating sol-ark with batrium because it was raining and i'm tired of waiting for official support, i played with batrium pylontech and some other emulators but they didnt work very well or at all, eventually i set it to "project lychee" protocol that...
diysolarforum.com
@Shawn Hayes Did you get our Batrium settings finalized? Curious how you system is performing this time of year given the amount of PV you have an battery bank size. Are you finding you are reaching 100% before the end of the PV day? Seems like if I have 2 or 3 good PV days my bank is full and I start running around trying to do extra laundry to not waste any PV.. ;-)
Shawn Hayes
Learning Everyday
Well I would not say that my settings are finalized. I guess I’m more in the tweaking stages to some degree? I had four of batteries that would dip a little bit lower than others at night. I had a few spares and I swapped out two of the lowest ones. So now when the bank gets to 3.34-3.35v @85%, one battery I swapped out starts to run to 3.38-3.39. (target 3.35-3.36 @90%) So I will swap that one out today with another spare that I have.
I would say that I will not ever get to 100%. As 3.35 V and 90% SOC is typically. That is my target. So far typically, on the high side I’m averaging 90 to 92% SOC, on the low side somewhere around 76-79% early in the morning before the sun hits the panels. My charge voltage is set to 53.9 V. Basically so far, I’ve had no trouble with charging my batteries back to the 90% range.
The caveat being we are not full time living there yet. We were still waiting on some furniture to arrive that was ordered, However it did show up this week. I am running everything as normal with the exception of whatever appliances (tv, w/d, micro, etc) we would use when living there full time. That would not be a lot more draw than is currently being used. Things like the dishwasher, washer and dryer would be run during the day rather than at night. Something I’ve done to try to save a little bit of energy at night, we have a freezer and two refrigerators, they are both on timers and go off at midnight and come on at 7 AM. During the daytime I’ve been running the air condition at 72°, as the sun goes down I will turn the thermostat to 75° at night. Just to keep some of the battery draw to a minimum. We have a very well insulated home. We have 2 x 6 exterior walls with spray foam insulation on every exterior wall including the roof. I live in the southeast, so it doesn’t get really cold here in the winters. But we have really hot, humid summers. So far I am pleased with how everything is working. I actually still have a 30 minute paid tech-support session with Batrium that I have yet to use. I am trying to save that until I have questions that I cannot figure out?
Judoal
New Member
Recently I did a battery (AGM) test by discharging the battery bank 15min. The test required measuring Voc after the discharge period. I asked support how to disconnect the batteries safely. The answer was switch no battery in spite of what the manual stated. But when I did that, all power to the loads was cut off causing no end of problems. Sol-ark support sucks. Next time I just put the system on bypass and disconnect the batteries. I think selecting no battery and deselecting activate battery could work as well
