houseofancients
Solar Wizard
i did and do, seems to work well here so far, but still testing
Seplos CAN BUS RS485 48v 200A 8S-16S BMS
- Thread starter Firechief
- Start date
houseofancients
Solar Wizard
havent tested comms from the canbus tot rs485 to mpp solar/voltronic yet
Ok, it is the pylonthec Rs485 comms implementation to mpp solar I would to test later.
houseofancients
Solar Wizard
using pylon on voltronic side, rm485 on seplos side.
this works .
however, have not tested 2 bms's
1 16s bms canbus version -> 1 16s rs485 -> voltronic yet
Good news, thank you. That's interesting, having a mix of canbus and rs485 version depending of who is the master, the battery bank could be use with voltronic or deye inverter !! I need to ask seplos if this configuration is supported. I'm tempted to say yes since the battery inter comms is the same ?
houseofancients
Solar Wizard
i doubt it is supported, however , since the rs485 model has been out for a short time, loads of folks have bought 1 or in many cases more that 1 canbus version, where they need the rs485 for inverter comms, so well worth testing
anyway , life has been busy, have gotten around to it yet.
btw meant to ask, the box you designed, what is the width of it ?
hopefully not more 48 cm ?
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,714
That is what I plan to test also once I receive the RS485 version. Interesting that now that I bought Seplos, JK's new BMS's now support RS485 also. ( I have the older CAN versions, which dont work with MPP).
Anyway, I am hopeful that since only the master communicates with the inverter, I can use the older CAN versions behind the RS485 version master. I guess it will come down to if the new BMS has two RS485 ports instead of a CAN port for Inverter and RS485 for inter-battery communication.
I have the RS485 BMS version, it as a dedicate RS485 port where the CANBUS was before, and and a pair of RS485 for inter battery communication, documentation is available in resource section of the forum. since the inter comms is the same as before, there is no reason why it would not work in a mix configuration
.I like JK too, but their protocol doesn't support addressing. it means you need to have a serial port for each BMS., bcoz you can't have all BMS connected to the same RS485 line.
It is not suppose to be always activated, but only a short time before the FET connect the battery. If you want to know the glory detail better to ask seplos
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,714
Thank you VERY much for confirming! That is awesome news! You made my day.I have the RS485 BMS version, it as a dedicate RS485 port where the CANBUS was before, and and a pair of RS485 for inter battery communication, documentation is available in resource section of the forum. since the inter comms is the same as before, there is no reason why it would not work in a mix configuration
I like JK too, but their protocol doesn't support addressing. it means you need to have a serial port for each BMS., bcoz you can't have all BMS connected to the same RS485 line.
houseofancients
Solar Wizard
"there is no reason....."
i agree, and have been saying that, however, still like it to be conformed
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,714
I will test it as soon as the new BMS arrives.
houseofancients
Solar Wizard
yeah, providing life gives me some time, so will i
I´ve tested it now and have written a lot with a technician from Seplos:
The hardware and the functions are good and working for a really good price, but the software is bad.
Seplos is doing wrong thinks with the parameters and what they are doing with the CVL respectively how they interpret it is simply wrong.
There are four relevant parameters in the software which can set:
-Monomer high voltage alarm & protection (cell)
-Total high voltage alarm & protection (battery)
At "monomer high voltage alarm" the BMS sends a CCL of 10A - no control after that if that could be enough. If it is still to much the cell goes up to 3,65V (= monomer high voltage protection) and BMS cuts out - thats not a drama but also not the fine-way
At "total high voltage alarm" the BMS sends a CCL of 0A and is setting the SOC to 100% - so charging ends up here without gradual current reduction!
At "total high voltage protection" the BMS cuts out - BUT: They are sending THIS VALUE also as CVL to the Inverter!
So the goal for the inverter (=CVL) is the same value at which the BMS cuts out - but anyway it will never reach it because at "THV-alarm" he gets the signal CCL = 0A! That doesn´t make sense to me.
This regulation is really bad when you have a Victron-System because Victron can NOT react to the CCL when charging from the MPPTs/Solar (CCL has only a effect when charging from the grid with the Multiplus or when grid feed-in is disabled)! So whats happening at Victron:
When a single cell-voltage is getting too high the BMS can only act when it reaches the protection-value (3,65) - cut out!
Because CCL is "disabled" the inverter charges up to CVL (= protection-value!) and would then go down with the current step by step - he would but its not possible because - cut out! ?
So what´s left for me: A stupid BMS with a better SOC-calibration (compared to a BMS without communication and no installed BMV), a Display where i can see the cell voltages and a software with the function to record data in liftime...
I hope this whole thing applies only to me and my BMS, otherwise i cannot understand why I´m the only one who´s criticizing this ?
(REC for example is doing it perfect - but it´s so expensive :/ )
The hardware and the functions are good and working for a really good price, but the software is bad.
Seplos is doing wrong thinks with the parameters and what they are doing with the CVL respectively how they interpret it is simply wrong.
There are four relevant parameters in the software which can set:
-Monomer high voltage alarm & protection (cell)
-Total high voltage alarm & protection (battery)
At "monomer high voltage alarm" the BMS sends a CCL of 10A - no control after that if that could be enough. If it is still to much the cell goes up to 3,65V (= monomer high voltage protection) and BMS cuts out - thats not a drama but also not the fine-way
At "total high voltage alarm" the BMS sends a CCL of 0A and is setting the SOC to 100% - so charging ends up here without gradual current reduction!
At "total high voltage protection" the BMS cuts out - BUT: They are sending THIS VALUE also as CVL to the Inverter!
So the goal for the inverter (=CVL) is the same value at which the BMS cuts out - but anyway it will never reach it because at "THV-alarm" he gets the signal CCL = 0A! That doesn´t make sense to me.
This regulation is really bad when you have a Victron-System because Victron can NOT react to the CCL when charging from the MPPTs/Solar (CCL has only a effect when charging from the grid with the Multiplus or when grid feed-in is disabled)! So whats happening at Victron:
When a single cell-voltage is getting too high the BMS can only act when it reaches the protection-value (3,65) - cut out!
Because CCL is "disabled" the inverter charges up to CVL (= protection-value!) and would then go down with the current step by step - he would but its not possible because - cut out! ?
So what´s left for me: A stupid BMS with a better SOC-calibration (compared to a BMS without communication and no installed BMV), a Display where i can see the cell voltages and a software with the function to record data in liftime...
I hope this whole thing applies only to me and my BMS, otherwise i cannot understand why I´m the only one who´s criticizing this ?
(REC for example is doing it perfect - but it´s so expensive :/ )
I received response from Link Liu:
I wanted to confirm if I understand this correctly:
H confirmed that this is how the precharge works.
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,714
i did exactly.. but i dont know why deye is not reading that i have two battery ...its reading BMS1 only :/
Similar threads
