Sol-Ark Inverters 8 and 12K

[newbostonconst]

I would order extra cells....I have needed 2 extra's one was my fault and the other wasn't so the vender replaced for free.....

Also note most inverters will charge with 17 or 18 cells. The Sol-Ark 12k will charge with up to 63 volts....so 63/3.5 says you can have 18 cells. I have a 17 cell balancer but haven't seen many BSM's that are more then 16 cells.

A 280ah * 3.5vdc = .980 kWh.....which is ~1kWh so a 17 cell pack is about ~17kWh.....

 

[Haugen]

I would order extra cells....I have needed 2 extra's one was my fault and the other wasn't so the vender replaced for free.....

Also note most inverters will charge with 17 or 18 cells. The Sol-Ark 12k will charge with up to 63 volts....so 63/3.5 says you can have 18 cells. I have a 17 cell balancer but haven't seen many BSM's that are more then 16 cells.

A 280ah * 3.5vdc = .980 kWh.....which is ~1kWh so a 17 cell pack is about ~17kWh.....

@Ulmo posted the specs in post #14. The nominal voltage for the SolArk is 48V.
If you read Will's sticky posts concerning charging profiles for different banks for LiFePO4 cells, he specifically addressed 48V as a 16S setup. I'm not really interested in pushing the limits on my new expensive solar installation for a couple extra Watt hours. There is a good chance that I will add another 16 cells later in a parallel bank. With the low cost, there is every reason to expand to 28kWh,especially when I switch to an electric vehicle.
BTW, nominal voltage for LiFePO4 cells is 3.2V. To get them to a full charge, you use 3.65V. That means you need almost 59V to properly charge 16 cells.

 

[newbostonconst]

@Ulmo posted the specs in post #14. The nominal voltage for the SolArk is 48V.
If you read Will's sticky posts concerning charging profiles for different banks for LiFePO4 cells, he specifically addressed 48V as a 16S setup. I'm not really interested in pushing the limits on my new expensive solar installation for a couple extra Watt hours. There is a good chance that I will add another 16 cells later in a parallel bank. With the low cost, there is every reason to expand to 28kWh,especially when I switch to an electric vehicle.
BTW, nominal voltage for LiFePO4 cells is 3.2V. To get them to a full charge, you use 3.65V. That means you need almost 59V to properly charge 16 cells.

3.65 is to high....you don't get much more capacity above 3.5
3.65 is used to balance and maybe do a capacity test, at 3.65 you run the risk of over charging a cell especially when you have 16 and the chance of having a high cell is great when it is hidden in 15 other normal cell voltages. I know from experience….

The Sol-Ark is rated to 63 vdc charging in the spec sheet. Keeping batteries between 50% and 80% charge will make them last the longest. So buying extra cells and having them in the pack will enable you to run longer and not have as much stress on the pack and lower the chance of over voltageing the pack and puffing a cell. This will also increase the pack capacity so you don't have to buy a second pack of the same large size and expensive cost if you are on the edge of capacity and also when the pack starts degrading years later. This can also get you by longer and thus pack cost will likely be a lot cheaper later. Paralleling 2 packs of different life and possibly chemistry might not work in the next few years. Lastly the higher voltage is more efficient to run because the inverters don't have as hard of time upping the voltage to 120/240 VAC, this is shown in many spec sheets and especially comparing the efficiency of a 24vcd inverter to a 48vdc inverter. And having extra cells gives you the chance of removing one cell latter as needed if one goes bad, it is easy to add a cell at the start and harder later.

Each Cell added gives you ~1,000 watt hours, not just a couple.

The only negative is having to get a 17 or 18s BMS/Balancer.

But do what you want, I am not trying to force you into do anything, it is more about discussion and learning and others learning. Have a great day.

 

[Haugen]

@newbostonconst I see your point and having margin on important system parameters makes sense. I appreciate your input.
Doubling the voltage from 12 to 24, or 24 to 48 means that your current for the same power draw will be half. This means you can use much smaller wire and save significantly on wire cost and there is less voltage drop across the length of the wire. The 3.2V difference in voltage is not going to change the wire gage or make any significant infrastructure improvement. SolArk requires a 48V battery, so that is set in stone.
It's true that there is very little capacity to gain by charging between 3.5V and 3.65V. It was never my intention to charge to those levels, but I wanted my charger to have capacity above my limits to not stress its components. I'd rather have issues with my $80 cells, or $200 BMS than with my $8000 SolArk. As you probably read, with only 7.6kW of solar, there is a lot of room to expand later.

My future battery upgrade will be due to a much higher power usage from an electric vehicle. It will be accompanied by more panels. Probably another 5kW or so. By then, I'll have seen how efficient my panels are and whether I risk overpowering the SolArk, or whether I'll just get close to the 12k limit.
I won't be adding cells to the same bank, but adding a new bank with its own BMS. Worst case scenario, I'll get 32 new cells and rebuild it all. Even that is cheaper than buying the pre-built solution from the big solar companies.

 

[Craig]

I don't know why they can't make these things work with a generic BMS it can not be that hard.

 

[Craig]

3.65 is to high....you don't get much more capacity above 3.5
3.65 is used to balance and maybe do a capacity test, at 3.65 you run the risk of over charging a cell especially when you have 16 and the chance of having a high cell is great when it is hidden in 15 other normal cell voltages. I know from experience….

The Sol-Ark is rated to 63 vdc charging in the spec sheet. Keeping batteries between 50% and 80% charge will make them last the longest. So buying extra cells and having them in the pack will enable you to run longer and not have as much stress on the pack and lower the chance of over voltageing the pack and puffing a cell. This will also increase the pack capacity so you don't have to buy a second pack of the same large size and expensive cost if you are on the edge of capacity and also when the pack starts degrading years later. This can also get you by longer and thus pack cost will likely be a lot cheaper later. Paralleling 2 packs of different life and possibly chemistry might not work in the next few years. Lastly the higher voltage is more efficient to run because the inverters don't have as hard of time upping the voltage to 120/240 VAC, this is shown in many spec sheets and especially comparing the efficiency of a 24vcd inverter to a 48vdc inverter. And having extra cells gives you the chance of removing one cell latter as needed if one goes bad, it is easy to add a cell at the start and harder later.

Each Cell added gives you ~1,000 watt hours, not just a couple.

The only negative is having to get a 17 or 18s BMS/Balancer.

But do what you want, I am not trying to force you into do anything, it is more about discussion and learning and others learning. Have a great day.

I like this approach. Makes life easy. If they would just make a simple switch that a BMS could use to cut charger or discharging when necessary

 

[fisherus]

I don't know if anyone is still researching Sol-Ark 12K systems on this forum but because of changes Sol-Ark has made available in the 12K, I thought I would inform you about them. The Sol-Ark 12k now is available to charge 63V, up from 59V. I am using one for my LTO 64.8V SCiB batteries using a 60V Nominal and 48V Minimal which keeps the batteries in their "sweet spot " range of 2.5V-2.0V per cell. With a C rate of 10C, my battery bank is pushing more power with the 20 64.8V 45Ah 24 cell batteries than I could ever have imagined. The ability for starting well pumps, compressors and welders on my off-grid farm has been amazing to see what the ability will be once everything is connected. I'll post pictures and more information once I have it going full force.

 

[Haugen]

I don't know why they can't make these things work with a generic BMS it can not be that hard.

SolArk does! That's why I choose it.
The BMS does its job and the SolArk works with the power available. Communication optional

 

[Craig]

SolArk does! That's why I choose it.
The BMS does its job and the SolArk works with the power available. Communication optional

Can you elaborte on how it works. Are you talking can us or just a switch controlled by some sort of voltage.

 

[Craig]

SolArk does! That's why I choose it.
The BMS does its job and the SolArk works with the power available. Communication optional

If BMS detects cell over voltage how would it tell sok-ark to stop charging?

 

[ArthurEld]

If BMS detects cell over voltage how would it tell sok-ark to stop charging?

I haven't figured out how to do that. But when using multiple batteries I don't want the charging or discharging to stop.
If something caused all batteries to stop, I hope Sol-Ark can live through all batteries disconnected.
Hopefully we can come up with something better by the time I actually install my system. But I'm fairly comfortable with what I described.

 

[Craig]

I haven't figured out how to do that. But when using multiple batteries I don't want the charging or discharging to stop.
If something caused all batteries to stop, I hope Sol-Ark can live through all batteries disconnected.
Hopefully we can come up with something better by the time I actually install my system. But I'm fairly comfortable with what I described.

Yes with multiple batteries you need to be able to switch of each battery individually

 

[Haugen]

Can you elaborte on how it works. Are you talking can us or just a switch controlled by some sort of voltage.

I don't have the exact answer yet, but the installer tells me that the SolArk "plays with" your battery the first couple days after coming online and it figures out what battery is out there.
It sounds like what you have to do for your BMS to accurately know your SOC. I think it must be counting coulombs so that it knows how much energy is available from that source.

Once the system is installed, ill be able to give better answers. It would be nice if my BMS CAN bus comes up and is able to talk to it. My hopes are not high though.

 

[Craig]

I don't have the exact answer yet, but the installer tells me that the SolArk "plays with" your battery the first couple days after coming online and it figures out what battery is out there.
It sounds like what you have to do for your BMS to accurately know your SOC. I think it must be counting coulombs so that it knows how much energy is available from that source.

Once the system is installed, ill be able to give better answers. It would be nice if my BMS CAN bus comes up and is able to talk to it. My hopes are not high though.

State of charge has nothing to do with individual cell voltage levels. I'm sure it will be able to figure out how many AH pass through but if any 1 cell gets out of line counting columbs does nothing. Sorry to keep asking these questions but I really like the sol-ark inverter but if it cant stop charging when one cell is too high or low it creates complications.

 

[Haugen]

If BMS detects cell over voltage how would it tell sok-ark to stop charging?

The BMS would cut off charging in that situation.
I chose a BMS with 2A of active balancing to try to minimize this.
Tge SolArk would notice that the battery stopped drawing current, and the BMS would turn charging back on when the cell was back in balance.
I hope to not have the BMS acting too frequently. The SolArk should have settings that will stop charging and discharge before the BMS limits are reached.

 

[Haugen]

State of charge has nothing to do with individual cell voltage levels. I'm sure it will be able to figure out how many AH pass through but if any 1 cell gets out of line counting columbs does nothing. Sorry to keep asking these questions but I really like the sol-ark inverter but if it cant stop charging when one cell is too high or low it creates complications.

If you don't trust your BMS, go buy one of the batteries that they have proven to work with the SolArk. They are a bit more expensive, but it will ensure that the CAN bus talks.

 

[Craig]

If you don't trust your BMS, go buy one of the batteries that they have proven to work with the SolArk. They are a bit more expensive, but it will ensure that the CAN bus talks.

I trust my BMS 100% I just think there us a much easier way than using canbus all the Sol-Ark needs to know is weather to allow charging or not. There is no need for in depth conversations. I guess what I am not explaining is I dont want to run 200 Amps through a BMS I want the bms to read the cell voltages and if one is off just tell the Sol-Ark to stop charge or discharge. So back to my original point why cant they make these things work with a generic BMS

 

[ArthurEld]

If you don't trust your BMS, go buy one of the batteries that they have proven to work with the SolArk. They are a bit more expensive, but it will ensure that the CAN bus talks.

I have thought about buying one made to communicate with Sol-Ark. And their price starts looking a lot better if I have to use expensive BMSs.
Alt E store sells batteries with their Sol-Ark package but I don't know if they communicate. I should ask.
This problem has been beating me up in the back of my mind for months.

 

[Haugen]

I trust my BMS 100% I just think there us a much easier way than using canbus all the Sol-Ark needs to know is weather to allow charging or not. There is no need for in depth conversations. I guess what I am not explaining is I dont want to run 200 Amps through a BMS I want the bms to read the cell voltages and if one is off just tell the Sol-Ark to stop charge or discharge. So back to my original point why cant they make these things work with a generic BMS

It is the job of the BMS to tend to the cells and stop charging/discharging in a situation where damage can occur. You are asking for this function to be handled by a device that is already an inverter, charge controller, and a transfer switch.
If you don't want to use CAN bus, you can use RS485, but that is a less structured communication bus. You should read up on CAN bus. Your vehicle likely has at least 1 and probably more CAN bus' operating. It's as easy as it gets to transfer this amount of data from one device to another. Think about it, it has to transmit the cell number and the voltage for every cell then the pack voltage, and the pack current every second or so. CAN uses ust 2 wires and termination resistors. If you tried to wire all those points, you would have a huge bundle to maintain, and have all the losses involved with the resistance of the wire giving you poor quality data.
Once you understand it, you realize that this is a simple, elegant solution to having 2 devices share information. Especially if there will be a little distance between them.

 

[Haugen]

I have thought about buying one made to communicate with Sol-Ark. And their price starts looking a lot better if I have to use expensive BMSs.
Alt E store sells batteries with their Sol-Ark package but I don't know if they communicate. I should ask.
This problem has been beating me up in the back of my mind for months.

I got wrapped around the axle with that myself, but after talking to SolArk and learning that communication via CAN bus is optional that it will work fine without it, I'm okay with it.