Reverse engineered Votronic RS485 protocol.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Tim Weber bd1e57b881
Merge pull request 'don't overwrite Interface.dump' (#2) from heeplr/votonic:main into main
4 months ago Update water/temperature sensor observations 3 years ago don't overwrite Interface.dump 4 months ago


This project plans to reverse engineer and document the protocol used by the inverters, solar charge controllers, tank sensors, displays etc. produced by the German company Votronic. These products can be found in recreational and expedition vehicles, boats and so on.

Note: There is also the S-BC Bluetooth Connector available from Votronic. If all you want to do is interface your setup with a smartphone, this might be your way to go. When I started working on this project, this component didnt exist yet. And now, I kind of want to finish it anyway.


Work in progress.

Ive figured out the physical/electrical setup (RS485 with RJ45 connectors) and am currently trying to make sense of the serial protocol thats spoken on the wire.

Hardware Im Using

Bus System VBS2 Master Box

This is the heart of the system, with one wire screw connection bar each at the top and bottom, interfacing to non-bus components.

It also has three RJ45 connections labeled “Bus”, which are (in my setup) connected to

  • the Display / control unit
  • the Equipment Adapter and
  • the Solar/Battery Charger

Battery Charger Triple VBCS 45/30/350

Solar Charge Controller and Battery Charger. Connects to both the ignition and RV batteries. Also has a bus connection.

I dont mess with this thing, it has big wires.

MobilPOWER Inverter SMI 1700 Sinus ST-NVS

Does the 230V shizzle. Isnt directly connected to the bus, but via the EQ-BCI adapter described below.

Equipment Adapter EQ-BCI

This is a matchbox-sized device that has two RJ45 jacks labeled “Bus” and one smaller jack (RJ12? have to check) labeled “B2B/Charger/Inverter”. It apparently translates between whatever protocol my inverter is using and the bus protocol analyzed in this project.

Since it has two bus connections, of which only one was used when I got my camper van, it was the perfect candidate to start reverse engineering and tapping into the bus. Looking at the chips inside the Equipment Adapter made me realize that the bus is using RS485, and looking at the datasheets of the ICs allowed me to reverse engineer the RJ45 pin mapping.

Board Contents

The top of the board contains:

  • 6N137 optocoupler
  • 25V 100µF capacitor
  • ZTT 8.00MT 8MHz ceramic resonator
  • diode labeled “Z 15”

The bottom of the board contains:

  • 7LB176 68M AKY5 RS-485 transciever
    • pin 1 is at the bottom left (holding it so that the label on the IC is upright)
  • 2× 690Y 2951 CMC voltage regulators, one for each bus port, apparently for power supply
  • TCLT1002 V 727 68 optocoupler

Values of Interest

On the control unit, I can see the following values:

  • RV battery
    • remaining capacity percentage
    • voltage
    • current (negative if discharging, positive if charging)
    • remaining capacity in Ah
  • starter battery
    • voltage
  • fresh water tank level percentage
  • grey water tank level percentage
  • indoor temperature
  • outdoor temperature
  • clock
  • solar current in A
  • status LEDs
    • 230V external power
    • charging battery (from 230V, I presume)
    • inverter
      • providing 230V
      • overload
      • manual
      • automatic

However, the indoor temperature sensor is connected directly to the display, so its quite possible that its value may not be available via the bus.

Physical Connection

Most of the products Im using connect through an 8-wire RJ45 jack. You can use cheap (flat) phone cable or twisted-pair (Ethernet) cables to connect to them.

When using a cable with T568B colors, pin 1 uses the white/orange wire; T568A use white/green on pin 1.

The pinout looks like this:

  • 1: not sure, seems not connected or reserved
  • 2 & 3: Vin (power supply)
  • 4: RS485 A wire
  • 5: RS485 B wire
  • 6, 7 & 8: ground

RS485 / UART properties

The wire protocol has a transfer rate of 19,200 bits per second, 8 bit, with even parity and 1 stop bit.

Linux users can use the following command to set up their RS485 interface:

stty -F /dev/ttyWHATEVER 19200 cs8 -cstopb parenb -parodd raw

Im using this RS485 USB adapter.

If you want to look at the raw data using hexdump, Id use something like

stdbuf -i0 -o0 -e0 hexdump -Cv /dev/ttyWHATEVER


Most of this is speculation. Dont rely on it yet. All byte values are written in hexadecimal.

There seems to be no collision avoidance on the bus. Devices seem to talk whenever they want, even while another transmission is already running. This leads to something like 5 to 10 % of packets looking wrong: Theyre longer than expected because theyre two packets mushed together, where one starts somewhere in the middle of the other.

Every packet on the bus Ive seen seems to be 9 bytes long and consists of the following parts:

  • A start byte, which is always aa.
  • Three bytes of header data which could contain a “from” address, “to” address, type, mode or a combination thereof. Not yet understood.
  • Probably the length of the payload; this was 03 in all non-collided packets I saw.
  • The payload; this was three bytes long in all non-collided packets I saw.
  • A one-byte checksum, calculated by starting with 0x55 and then xor'ing each byte in the packet (including the aa) on top of that. (Thanks Manawyrm for finding it out!)


  • When switching the display to “starter battery voltage”, I noticed the repeating packet aa 62 f4 44 03 03 16 05 3e, while a voltage of 13.0 was displayed. 05 16 is equal to 1302 in decimal.
  • In aa 62 f4 0c 03 03 16 05 76, 05 16 again corresponds to a displayed voltage of 13.0 (its 1302 in decimal), but this time for the RV battery.
  • Displaying water tank percentage leads to packets like aa 62 f4 14 03 02 1a 00 66 (1a being the fresh water percentage of 26 at that time) and aa 62 f4 18 03 02 3d 00 4d (3d corresponding to 61 % grey water). The packet with 14 seems to be the fresh water, the one with 18 the grey water value.
  • For temperatures of 20 °C indoors and 8 °C outdoors I was noticing the packets aa 62 f4 30 03 02 4b 00 13 and aa 62 f4 34 03 00 e5 00 bb. Manipulation of the indoor temperature seems to have no effect on these values, but since the indoor temperature sensor is connected directly to the display, I dont expect it to change. Ill try to find the outdoor sensor, though.
  • I couldnt identify a package responsible for updating the clock each minute, so I guess the display has its own RTC (the clock survives disconnecting the display) or is initialized on boot from the master.
  • When displaying the current drawn at the moment, packets like aa 62 f4 0c 03 02 ac ff 37 appear. ff ac is twos complement and means -84. -8.4A was the current consumption displayed at this moment. This observation also works for other negative and positive values.
  • aa 62 f4 0c 03 05 5d 00 3e seems to correspond to a “remaining capacity” value of 93 Ah (00 5d).
  • In aa 62 f4 0c 03 06 3a 53 09, the 3a seems to correspond to the “remaining capacity” value of 56 % (confirmed with other percentages). The byte following it fluctuates over time, getting smaller the more 12V power Im consuming at that moment. If my consumption is high enough, the byte falls below zero and wraps around. Im not sure what it means yet.
  • A message like aa 22 0c f4 03 06 00 00 20 is sent before every “RV battery capacity percentage” message arrives: Its the request for that value. Ive tried sending this message and indeed got a capacity percentage back.


These are the messages that Im fairly sure of. For most of these, some bits or bytes are still unclear, these are marked as ??.

## is the payload length byte and usually 03.

  • 0c ## 02 LL HH: RV battery current consumption, where LL HH are a 16-bit twos complement deciamp value (e.g. 86 ff for -12.2 A)
  • 0c ## 03 LL HH: RV battery voltage, where LL HH are a 16-bit centivolt value (e.g. 16 05 for 13.02 V)
  • 0c ## 05 LL HH: RV battery capacity remaining (Ah), where LL HH are an unsigned integer (e.g. 58 00 for 88 Ah)
  • 0c ## 06 XX ??: RV battery capacity remaining (%), where XX is an unsigned integer (e.g. 37 for 55 %)
  • 10 ## 02 LL HH: solar current input, where LL HH is a 16-bit unsigned deciamp value (e.g. 06 00 for 0.6 A)
  • 14 ## 02 LL HH: fresh water tank percentage, where LL HH is an unsigned integer (e.g. 1a 00 for 26 %)
  • 18 ## 02 LL HH: grey water tank percentage, where LL HH is an unsigned integer (e.g. 3d 00 for 61 %)
  • 44 ## 03 LL HH: starter battery voltage, see RV battery voltage above


Why are you doing this?

Because I own a camper van containing some of their products, and Id like to be able to interface with them in order to read power and water levels etc.

Theres a typo in the project name!

No theres not. For trademark reasons, this project is named after vodka tonic. Because why not!