Seplos CAN BUS RS485 48v 200A 8S-16S BMS

OK, it looks a lot better with that number. But you'll either need to find a way to force the battery to get to low voltage disconnect or manually set the correct numbers. There should be a number for residual capacity (remaining capacity). Or at least some other editable number that would allow you to set correctly the SOC.

This is the quote from the official documentation of the BMS that leads me to the conclusion that once fully charged and then fully discharged the capacity and the SOC will be calculated properly:

BMS calculates the remaining capacity in real-time. The BMS get the
capacity at the first time when the battery pack complete a full charging
and discharging cycle. And the SOC calculating accuracy is ±5%.

Click to expand...

I can try setting the inverter to lead acid mode; it might allow me to customise the disconnect voltages

This should work too. As long as it is the BMS that cuts out on low/high voltage, but not the inverter then the calibration should go fine.

brum said:

This should work too. As long as it is the BMS that cuts out on low/high voltage, but not the inverter then the calibration should go fine.

Click to expand...

If I set it to ‘default’ I have the option to specify these parameters which have these default values and minimum allowed to be set

Low V protection 44V - 42.5V
Min discharge V 46V - 43.5V
Empty discharged 46.7V - 42.5V


The ‘total_voltage undervoltage protection’ on the bms is set to 43.2V

Edit: I noticed if I set the battery to default it goes back to ‘general lithium’; if I disconnect the can cable to the bms it stays on ‘default’, however after a couple of minutes the inverter goes into a fault mode

Now I cannot connect to the BMS using Battery Monitor anymore; I get a time out error.
I can still connect via bluetooth
Has anyone had this problem?

I had to reset the bms to be able to connect

I still have problems hitting 100%





It looks like the voltage drops when the current drops to 9A, hence the ‘total voltage overvoltage protection’ is not triggered

Looks like the inverter is charging to a specific voltage and this is likely not coming from the battery BMS. I would go over the inverter settings and documentation, and try to understand what is the charging algorithm used by it.

brum said:

Looks like the inverter is charging to a specific voltage and this is likely not coming from the battery BMS. I would go over the inverter settings and documentation, and try to understand what is the charging algorithm used by it.

Click to expand...

Tonight I tried making ‘total overvoltage alarm’ and ‘total voltage overvoltage protection’ closer to each other: 55.8 and 56V (and the same with their ‘monomer’ versions)
It has now reached 100% and kept in standby for the remainer of the off peak period

My theory was that when the overvoltage alarm makes the current drop to 9A the previous settings meant that it would make the total voltage drop as well, so it would take ages to charge to 100%

I think I need to find the correct delta between overvoltage protection and alarm that makes the current drop but still reach 100% SoC

Has anyone noticed that if you click on ‘history records’ it shows all the saved alarms but the latest dates and times are in the future?

cajocars said:

Now I cannot connect to the BMS using Battery Monitor anymore; I get a time out error.
I can still connect via bluetooth
Has anyone had this problem?

Click to expand...

That could be a number of things. Cable is bad, port config on the computer etc...

One thing I've found with these BMSs is that the on board RJ45 connectors are pretty fragile. I accidently stepped on a cable a few weeks ago bench testing a new battery I built and bent one of the RS485 connectors 90 degrees upward bending all the pins inside. De-soldering and replacing it wasn't a big deal but now I feel like I need to be super careful. The only thing really holding these sockets to the board are the pins. There are no metal tabs soldered down or through the board to reinforce them, and those little black plastic nubs through the board don't do much at all to strengthen them.

Thought I'd check on the system today and noticed that although the battery seems be charging ok from the PV, it's not charging at night (E7) and not really discharging during the evening. Looked back though the inverter history and it seems to have started this weird behaviour a couple of days ago. There are no faults logged on either the BMS or the Luxpower. Any ideas guys?

Seems to be in Standby mode too. How do I get out of that?

Standby mode means that no charging/discharging current is flowing through the BMS. Start charging or discharging and it will change by itself.

Turns out it was a frequency issue with the grid. Emailed LuxPower and they tweaked the inverter.

I've just built my Seplos Mason 280 and reached 100% SOC for the first time when the below happened. I fully intend to read this thread and get caught up. But in the short term can anyone tell me if the below anything to worry about, as it looks catastrophic? It started with one cell initially and I assumed it was an issue with that cell, but then all cells intermittently started getting "Monnomer overvoltage protection".

danmc said:

I've just built my Seplos Mason 280 and reached 100% SOC for the first time when the below happened. I fully intend to read this thread and get caught up. But in the short term can anyone tell me if the below anything to worry about, as it looks catastrophic? It started with one cell initially and I assumed it was an issue with that cell, but then all cells intermittently started getting "Monnomer overvoltage protection".

Click to expand...

If this is the 10E version (with bluetooth); momomer overvoltage protectiom and total voltage overvoltage protection are the only triggers for 100% capacity
I.e. the bms will reset to 100% whenever one of those is reached

In my case, I think my inverter uses the bms SoC value to know when to stop so I either charge to 100% each time and hit one of those protection or charge to something lower like 95% and only charge to 100% once a month or so… I guess I’ll do the latter

cajocars said:

If this is the 10E version (with bluetooth); momomer overvoltage protectiom and total voltage overvoltage protection are the only triggers for 100% capacity
I.e. the bms will reset to 100% whenever one of those is reached

In my case, I think my inverter uses the bms SoC value to know when to stop so I either charge to 100% each time and hit one of those protection or charge to something lower like 95% and only charge to 100% once a month or so… I guess I’ll do the latter

Click to expand...

No mine is the '1101-SP16' if that means anything to anyone. I believe it's the model just before the bluetooth BMS. Perhaps that works in a similar way. Looks like i've got some reading to do.

danmc said:

No mine is the '1101-SP16' if that means anything to anyone. I believe it's the model just before the bluetooth BMS. Perhaps that works in a similar way. Looks like i've got some reading to do.

Click to expand...

in that case, it looks like 100% is when it hits 56V, regardless of what you set in the parameters; so if your overvoltage protection is lower than 56V you will never see 100%; also if you set it higher than 100%, it will keep charging but still show 100%; and when you discharge it, it will show 100% for a bit until you go below 56V
at least based on what Andy says in his video

cajocars said:

in that case, it looks like 100% is when it hits 56V, regardless of what you set in the parameters; so if your overvoltage protection is lower than 56V you will never see 100%; also if you set it higher than 100%, it will keep charging but still show 100%; and when you discharge it, it will show 100% for a bit until you go below 56V
at least based on what Andy says in his video

Click to expand...

I think the behaviour is related to firmware version rather than model. I too have the same 2.8 firmware as Andy. That would explain it. It's sitting at around 54v and just switching my charge controllers on and off. I'll add this to the list of other peculiarities I need to investigate since my system went online yesterday.

I suppose this is half the fun and why we do it. ?

Does anyone know what the latest FW is for the 10D version of the board?

This is a quote from their site:

A:Not allow to upgrade the Firmware program without permission,after sales service is not available if its cause any damage.

Click to expand...

Based on the above I suppose firmware upgrades are not a user-friendly thing and are used only in case of issues.

Well it looks like I’ve still got problems with my setup. Battery charges from PV and sometimes from E7 but is exceptionally flakey when it comes to discharging. Luxpower assure me that from the data, the inverter is calling for DC but the BMS just sits there in standby mode most of the time with a short token discharge when it feels like it. Guess I’ll have to try and get some sense out of Seplos ?

Check the voltage at the battery terminals after the BMS. If the BMS cuts out the discharge path you'll see low voltage there. I doubt that this is the BMS. If it is working correctly the discharge will not be limited unless the SOC is at 0% (which is the same as having a cell below the specified low voltage protection level).

And to confirm Andy's finding on the SOC drifting if you don't fully charge the battery pack regularly. I had about a month of battery use without getting to either 100% or 0%. Yesterday I charged it fully and found out that the SOC has drifted about 30% off.

How it went - during the charging process, the SOC got to 98.2% and then the battery took almost 100Ah before the SOC started moving up again. This happened as soon as a cell got to the 3.5V point (my high voltage warning threshold). When a cell hit the 3.6V (the high voltage protection threshold) the SOC got to 100% and the BMS disconnected the battery.

I never discharged below 45%. Seems like this has been 45% SOC at the beginning, but this has slowly drifted toward 15% before yesterday. My goal was to not get the battery completely drained during the cold months when I was trying to keep it warm and above 15C.

The algorithm utilized by the v16 BMS (10E) seems good on this. It can't easily track the SOC perfectly without frequently detecting the 0% or 100% SOC points. I like how it did not go to 100% before actually getting the cells fully charged. This would most likely prevent the inverter from further charging. I also like the approach for calibrating the 98.2% SOC at the point where the cells are hitting the warning threshold and the 100% SOC at the point where the BMS is hitting the protection threshold.

What would be better done would be to have more precise detection of the flowing current. But this would mean a shunt with higher resistance and more losses. To me, the design that Seplos took seems just right. I would also appreciate it if they include that detail in the documentation - I would know what to expect without having to reverse-engineer it.

Hello
I read about setting the BMS to ginlong
But I do not find this setting at all
I have a 10c.

Ampster said:

I would ask Seplos to give you examples of communication with Victron or Fronius. I would not assume all CANBUS or 485 devices can communicate with each other unless you are an engineer who is familiar with those protocols. There is a user here who has successfully communicated with a Victron and an Orion BMS.

Click to expand...

Check out Andy on his "offgridgarage" youtube channel to see if he has an update on his seplos experience.