diy solar

diy solar

Chinese BMS+Deye 12kw 3 phase help (BMS DR-1363/DR-Power/DR_application)

Steemsmij88

New Member
Joined
Nov 12, 2023
Messages
12
Location
The Netherlands
Hey Guys!

I am having trouble understanding the poorly Chinese-translated BMS of my 51.2V 16S LifePo4 batteries.
Just to be sure I want to make clear that the provided manual is poor and does not include any information on the settings inside the BMS.
After a few troubles, I managed to make contact using a USB to RS485 cable to my batteries (more info here)

I am hoping someone can assist me or nudge me in the correct way for the things I want to achieve:
  • a charging voltage of 3.45V per cell or 55.2V for the 16S pack.
  • a floating Voltage of 3.35V per cell or 53.6V for the 16S pack.
  • cell balancing to start at 3.4V per cell
The reasoning behind the exact voltages is not my goal in posting this thread. Charge settings etc are things that are different for a multitude of reasons but the biggest reason behind this thread is to understand 'where' my settings are in my BMS.

I will post a few screenshots of the BMS page, maybe these are enough for people to assist me in the search for these settings. In case someone wants to dig deeper into the program I can post a download link to the program I am using. This file is provided by my battery manufacturer.
People downloading this file can use it without having a battery connected. Of course, all the settings will be 'blank' but you can help me search for the settings deeper into the program without me having to post all the screenshots here.

I hope you guys can help me out.
 

Attachments

  • 1.PNG
    1.PNG
    51.3 KB · Views: 61
  • 8.PNG
    8.PNG
    20.6 KB · Views: 52
  • 7.PNG
    7.PNG
    18.1 KB · Views: 46
  • 6.PNG
    6.PNG
    20.1 KB · Views: 46
  • 5.PNG
    5.PNG
    19.3 KB · Views: 42
  • 4.PNG
    4.PNG
    11.1 KB · Views: 38
  • 3.PNG
    3.PNG
    21.3 KB · Views: 37
  • 2.PNG
    2.PNG
    24.3 KB · Views: 54
Last edited:
  • a charging voltage of 3.45V per cell or 55.2V for the 16S pack.
  • a floating Voltage of 3.35V per cell or 53.6V for the 16S pack.
These are charge controller settings and should not be attempted with BMS settings. You won't find these settings in BMS configuration.
Charge settings etc are things that are different to a multitude of reasons
What reasons do you have to not control charging with your charge controller?
 
MisterSandals,

It seems to me that you nudged me in the correct way. Thank you for that!
It is only recently that I am the proud owner of a Deye 12kw 3-phase hybrid inverter with my 8x 52.1V 100ah batteries. After a lot of tutorials and a lot of information, I am afraid I thought the charge settings were given by the BMS.

Maybe you can understand where this misunderstanding stems from:
I have attached some pictures I took from my Deye inverter and inside the Battery settings (Battery Settings.jpg) I can choose what kind of communication my inverter makes with my batteries:
  • Lithium (uses the BMS port and communicates directly with the BMS)
  • Use batt V (uses the battery Voltage as a main parameter. When choosing this mode I get to set charging voltage settings (using voltage.jpg)
  • Use Batt % (uses the battery %, I think a derivative from the Voltage)
  • No Batt (the inverter neglects the battery and does not use it.

Using Lithium (BMS port and BMS communication)
Whenever I use the Lithium option, I can't change any charging voltages, float voltages, no absorption voltages. It looks like it is then regulated by the BMS. This is illustrated in the attachment 'using BMS in settings.jpg' here you see the Battery Charge Voltage is set to 57V, and the Charge- and Discharge Current Limit is set to 640A (8 batteries x 80A). This made me think the BMS is trying to pull 57V.

On the left-hand side of the same attachment, you can see that I turned on Grid Charging for demonstration purposes. The inverter is charging the batteries with 206A (divided over 8 batteries) at a voltage of 53.83V. I can't place this Voltage since I don't see it anywhere.

Use Batt V (using the Voltage as a main parameter and no BMS communication)
When I choose the option to use the battery voltage as a parameter, I can enter more settings. It lets me set absorption and float voltages.
I set them according to my wishes (55.2V absorption and 53.6V float). When I now try to do a 'grid charge' again for demonstration purposes it looks the same:

My coming plan
To fully understand the charge settings of the Deye inverter I installed Home Assistant and now I can accurately measure and log the Voltages and Amperages of the inverter and the batteries. I want to do the following:
  1. Discharge my batteries to 10% and Grid Charge to 100% using the Battery setting 'Lithium'. So with BMS communication and the presumed 57V charging voltage.
  2. Discharge my batteries to 10% and Grid Charge to 100% using the 'Use Batt V' setting. So without BMS communication and the presumed absorption (55.2V) and float voltages (53.6V).
I then can report my findings here for some more help.

For now: I really appreciate your help!
 

Attachments

  • when using BMS in settings.jpg
    when using BMS in settings.jpg
    1.8 MB · Views: 17
  • using BMS in settings.jpg
    using BMS in settings.jpg
    1.8 MB · Views: 14
  • battery settings.jpg
    battery settings.jpg
    1.6 MB · Views: 15
  • using Voltage.jpg
    using Voltage.jpg
    1.9 MB · Views: 13
I have attached some pictures I took from my Deye inverter and inside the Battery settings (Battery Settings.jpg) I can choose what kind of communication my inverter makes with my batteries:
This is outside my experience. You should start a thread with "Deye inverter" and "battery settings/comms" in the title to attract folks who can help.
I'd also explain what battery and BMS you have.

Sorry, i wish i could help with your specific issue. Someone here will have what you need.
 
  • a charging voltage of 3.45V per cell or 55.2V for the 16S pack.
  • a floating Voltage of 3.35V per cell or 53.6V for the 16S pack.
You can set those in your inverter under Battery settings, page batt set3, the same that is in your picture with the name "using Voltage.jpg", which you have already done.

Whenever I use the Lithium option, I can't change any charging voltages, float voltages, no absorption voltages. It looks like it is then regulated by the BMS. This is illustrated in the attachment 'using BMS in settings.jpg' here you see the Battery Charge Voltage is set to 57V, and the Charge- and Discharge Current Limit is set to 640A (8 batteries x 80A). This made me think the BMS is trying to pull 57V.

On the left-hand side of the same attachment, you can see that I turned on Grid Charging for demonstration purposes. The inverter is charging the batteries with 206A (divided over 8 batteries) at a voltage of 53.83V. I can't place this Voltage since I don't see it anywhere.
These are the limits that the BMS sets, it just shows what the BMS limits are. Actual charging and discharging is limited by the inverter in your case, since 240A<640A.
The 53.83V and 206A are what the inverter is currently sending to the batteries, 206A at 53.83V. This changes depending on the battery SoC, load, BMS limits. I would guess your current loads with the inverter are at about 1kW, since the inverter limits the battery charging and loads to a total of 12kW (assuming you have the sun-12k-sg04lp3-eu) when you charge from the grid. With solar+grid, you can go over that limit.

  • cell balancing to start at 3.4V per cell
This you can set in the BMS, personally I'd set the starting voltage difference lower than 30mV that you currently have.
 
Alright, I have found some stuff regarding the charging pattern of my Deye 12kw 3phase hybrid inverter.
It looks like there is a bug in the software of my Deye inverter. It's not BMS-related.

To keep everything neat in terms of forum subjects, I will no longer go into the questions here that are charge/inverter related.
These questions will continue in a new forum thread in the correct subject 'All-in-One Systems'. I will link the thread in this post for future troubleshooters.

BMS question
These are the limits that the BMS sets, it just shows what the BMS limits are. Actual charging and discharging is limited by the inverter in your case, since 240A<640A
It's nice that you agree with me that the BMS gives out these parameters to my inverter. That is the exact thing that I am looking for inside my BMS. Since I want to set this value to 55.2V so that I can run my inverter in 'The BMS communication method'.

Can anyone who has some experience in these Chinese BMS-es try and help me in the right direction? In my opening post, there are some screenshots included, but not all setting fields are shown. What is that charging current called in other BMS-es?

personally I'd set the starting voltage difference lower than 30mV that you currently have.
I have lowered the voltage difference to 10mV. From my understanding; this 10mV is the maximum cell voltage difference the 'balancer' tries to keep the cells right?

Thanks guys! Much appreciated!
 
It's nice that you agree with me that the BMS gives out these parameters to my inverter. That is the exact thing that I am looking for inside my BMS. Since I want to set this value to 55.2V so that I can run my inverter in 'The BMS communication method'.
You already have communications working, since the inverter recognizes the values sent to it. There isn't really much sense in setting the BMS limit to 55.2V, this way it would disconnect when you go over that value, which might happen when you're charging. Think of the BMS as a line of defence for the batteries, it is there to prevent damage to the batteries, not regulate their day-to-day use (this is the inverters job).

Personally, I use the same inverter with JK BMS without communication, I have my BMS limits at 2.8V lower limit and 3.6V upper limit at the cell level, so 44.8V and 57.6V at 16S respectively. Since my inverter charge and discharge levels are respectively lower and higher, I never hit the BMS limits, so the batteries won't need to disconnect. I have float at 54.8V, absoprtion and equalization at 55.5V (so the inverter charges up to 55.5V, then lets the battery drop until the float level and keeps it around that level). This way, the battery can still accept a surge current when I turn off a bigger appliance or machine. With the batteries disconnected, you'll need to rely on just the grid.

I have lowered the voltage difference to 10mV. From my understanding; this 10mV is the maximum cell voltage difference the 'balancer' tries to keep the cells right?
It starts the balancing once the difference between cells is 10mV or more. JK BMS balances until they are within 1mV, I'm not sure about the accuracy of your BMS. Since you should only do this at the top end of the charging curve, (starting at 3.4V for example), even if it balances to 10mV, the difference will pretty much disappear whenever you apply even a smallish load on the battery.
 
Thank you so much, for the great explanation. I'm still learning.

Alright, I have found some stuff regarding the charging pattern of my Deye 12kw 3phase hybrid inverter.
It looks like there is a bug in the software of my Deye inverter. It's not BMS-related.
I know I should make a new thread with this question, but now the 'bug' in the software I found kind of pops up naturally for me:

Since I think we have the same inverter Deye 12kw 3phase SUN-12K-SG04LP3-EU, I want to ask you something about its charging possibilities.
For your information, I am on the software version: HMI 1001-C037 MAIN Ver 2005-1140-1807

When using the 'Batt Voltage', so no BMS communication, we can specify the charge parameters the Deye uses (using voltage.jpg) out of my opening post). Here you can set the floating and absorption voltages like we talked about. I noticed something strange when I set my float to 53.6V and my absorption to 55.2V. Then I go to TOU schedule and I notice that I am not able to set the desired voltage to 55.2V (absorption). The maximum voltage I then can set in the TOU table is my float voltage of 53.6V. Did you notice that as well?

So, given it is wintertime where I live and there is not enough sun to charge the batteries, I tried out my charging settings using the TOU table with 'Grid Charge'. When I then accept the TOU set voltage of 53.6V that then is the maximum charge voltage the Deye inverter is supplying to the batteries. This feels kind of strange to me since I specified the Deye to 'absorb' with 55.2V.

I hope you understand my question here.
 
Last edited:
Since I think we have the same inverter Deye 12kw 3phase SUN-12K-SG04LP3-EU, I want to ask you something about its charging possibilities.
For your information, I am on the software version: HMI 1001-C037 MAIN Ver 2005-1140-1807
I'm on HMI 1001-c020 MAIN Ver 2005-f098-1807. Since I'm not aware of a public changelog, I'm not sure if or whether there are any significant differences.
When using the 'Batt Voltage', so no BMS communication, we can specify the charge parameters the Deye uses (using voltage.jpg) out of my opening post). Here you can set the floating and absorption voltages like we talked about. I noticed something strange when I set my float to 53.6V and my absorption to 55.2V. Then I go to TOU schedule and I notice that I am not able to set the desired voltage to 55.2V (absorption). The maximum voltage I then can set in the TOU table is my float voltage of 53.6V. Did you notice that as well?
I use Solar Assistant to monitor and make changes, since that integrates to Home Assistant quite nicely. It's also easier to do all the changes that you'd want, even when away from the inverter. In Solar Assistant, I can set the voltage to whatever level is within the parameters (so, between 40 to 60V), however it won't charge higher than the absorption voltage. When it hits absorption, it drops to float and won't go above that.
I just checked, if I try to do it manually from the inverter, it won't go above float as well. If you're just trying to top them up once in a while, you can raise the float voltage temporarily. I do that once or twice a month if I see some cells go out of balance (I have 2 16S6P banks, so there's bound to be some more balancing issues). However, you will only miss at most 1% of the total energy if you just charge to float. The only noticable issue would be a slightly lower charging speed at the top end.
 
Superb!

I am in the process of installing Solar Assistant as well. My SA inverter cables and battery cables are arriving tomorrow.

So you are using SA and have no communication with the BMS. Do you use 'inverter values' in SA then, or 'Emulated BMS'?
In that case, I could try that as well with my current USB-RS485 cable I have right now. And if that works great, I could just leave the BMS to do its job on its own without communication with the inverter.

Another question, since we have almost the same setup. Do you also see a slight voltage difference between the inverter Voltage value and the actual batteries themselves?
 
Last edited:
So you are using SA and no communication with the bms. Do you use 'inverter values' in SA then?
In that case I could try that aswell with my current usb-rs485 cable I have right now. And if that works great, I could just leave the bms to do it's job on its own without communication to the inverter.
I used to use the JK BMS adapter, that would show all the voltage values in SA as well, it's a bit more accurate with the SoC calculation. But "inverter values" does work, if you just want to see the pack voltage, current and SoC.
Seeing as your BMS does communicate with the inverter, you can just use "inverter values" as well, or if you want to, you should be able to add the battery separately to SA as well, just need to see what protocol it uses (something from the RS485 protocols here: https://solar-assistant.io/help/battery/configuration), though you will probably need a splitter, unless the BMS has 2 RS485 ports.
Another question, since we have almost the same setup. Do you also see a slight voltage difference between the inverter Voltage value and the actual batteries themselves?
Yes, there is up to 0.5V difference (voltage is higher on the inverter side), it somewhat depends on the load, but even at rest there is a significant difference, like 0.2 to 0.3V. For some reason, the difference gets a lot lower if I disable one battery bank (doesn't matter which one).
Even when measuring from the battery connections inside the inverter, the voltage is almost the same as the BMS shows, so something is off on the inverter side.
 
That is 12x 16S batteries total which would require 12 BMSs. Maybe you have 2x 6P16S batteries and 2 BMSs?

How big are the cells (Ah)? I would love to see a pic.
2x 6P16S indeed, with 2 BMSs. They're 60Ah S168 cells, green wrapped cylindrical LiFePO4. One bank measures about 38x35x100cm (15″ x 14″ x 40″), plus accessories. I want to clean up my build before I'm comfortable posting about it, I'll probably make a post in the show and tell forum once I'm happy with everything.
 
Back
Top