I cannot get my EG4-LLs to talk to Solar Assistant for the life of me

[mike616]

I am trying to connect my EG4-LL (24V) batteries to Solar Assistant. I have two questions that I cannot figure out.

1. How do I wire up an ethernet cable to the RS485-to-USB adapter?
2. Given the EG4's wonky DIP switch issues, how can I be sure I'm on the right address (address 2 per Solar Assistant's guide).

For question 1, I've followed this image:

I am wiring port 5 to the A+ terminal, 4 to the B- terminal, and 8 to the GND terminal. I've also tried flipping 4 and 5. I get nothing from the batteries in Solar Assistant or in EG4's BMS Tools.

I've also tried using this cable with the same results.

For question 2, I have no way to know what DIP setting I'm supposed to be on for my first battery. Multiple posts on here say that people have their DIP switches flipped or mirrored. I've tried the following permutations of address 2 and got nothing:

1 2 3 4
D U D D
D D U D
U U D U
U D U U

Is there any reliable way to determine what address the battery BMS thinks it's on? I don't see anything in the limited LCD screen options. I've tried using `minicom` on my Linux machine and BMS Tools in a Windows VM and just nothing at all.

I've been at it for three days now and have idea what to try next. Please help!

 

[mike616]

Some other stuff I found out today which might be interesting:

I tested continuity across all 8 pins of the RS485 port on the battery, using the image above as a reference, the following pins are bridged:
1 - 8
2 - 7
3 - 6

I also checked voltage across every pin to every other pin and found the following:

Voltage (ABS)

	Pins 1 - 8​	Pins 2 - 7​	Pins 3 - 6​
Pins 1 - 8​	X	4.16	0.44
Pins 2 - 7​	4.16	X	4.6
Pins 3 - 6​	0.44	4.6	X

4 and 5 have no voltage across any other pin.

What's strange to me here is that 4 and 5 are supposed to be the data signals but they have no electrical potential across any other pin. Yet all the other pins (which should be unused in half-duplex RS485) do have some voltage.

So I tested every combination of these three pin pairs into the three ports of the RS485 adapter. Same results.

Not sure if any of this is relevant but it took me a long time to get all this info and I figured I'd share it.

 

[Brett V]

Deleted

 

[Brett V]

you can’t measure voltage. It’s very slow speed data but still not something you can see without an oscilloscope

 

[Brett V]

Never mind…I was thinking it was RS232

 

[timselectric]

It's possible that one of the dip switches is faulty.
Try using one of the other batteries as the master.

 

[mike616]

Never mind…I was thinking it was RS232

No you're probably right. An oscilloscope would have been the right tool for this. I just wanted to see if there was anything going across those pins.

 

[jberger]

Just thinking out loud since you have a couple of days into this, wall of questions below.

Have you tried using the cable that came with batteries instead of the 3rd party?
Are you sure you are using the right BMStools package for the battery?
Are you powering the battery off and back on after changing the DIP switches?
Any change you could try windows on bare metal and not have to deal with the VM configuration?
Are you sure you have the correct baud for the connection and are you letting BMS tools scan the entire range of com ports and ID's?

I struggled the first time because the original manuals for the 12 volt units had the wrong DIP settings listed. So I ended up just connecting to one battery with the factory cable and walking all of the DIP's to make sure I had the correct ID's and baud rate. Once I had BMS tools working, then I dealt with the troubleshooting to get it working in SA.

 

[webbbn]

4 and 5 have no voltage across any other pin.

What's strange to me here is that 4 and 5 are supposed to be the data signals but they have no electrical potential across any other pin. Yet all the other pins (which should be unused in half-duplex RS485) do have some voltage.

RS485 is differential and should not be connected to any other pins. That's one of the reasons that it's used in these applications, so you can keep your battery communications interface electrically isolated from the battery.

I believe modbus is also a request/reply protocol, so you also will likely not see any communications from the battery over those wires until, whatever you're connecting to it sends a valid modbus request that the battery is configured to receive and respond to.

 

[Brett V]

Is it possible the USB on the Raspberry Pi doesn’t have enough current to power the converter?