• python prototype

    Alexander Krause • 03/12/2024 at 15:19 • 0 comments

    Today I started to connect the bus to my PC.


    A USB/USART adapter from FTDI was my choice since I got tons and various PCB's laying around.

    The FTDI can be configured for RS485 to enable a IO when you're sending data. I tried the interface with the two transistors which was mentioned at the afterburner project but that didn't work for me. (I used mun2214 and mun2112 transistors).

    I think a normal RS485 would do when you're using A against GND, so that's what I did next.

    The 3k resistor is for protecting purposes.

    So anyway, using a simple hexdump with 4800 baud gave me the bytes I saw in my analyser yesterday. So I decided to have a better look at the frame structure.

    It seems so be a quite simple structure but I'll have to find out what each byte stands for though.

    I spent some minutes to write a little parser to check if my assumption for the link-layer structure is correct.

    So far so good. There seem to be a runtime counter which increases with every frame once the heater is running. The remote sends 06 and 02 commands. The first 06 is for turning on, the second for turning off the heater.

    I've uploaded the current version of the script to the files section.

  • basic protocol structure

    Alexander Krause • 03/11/2024 at 14:12 • 0 comments

    Some days ago a friend told me about the heaters from Vevor and after some search I found this page very promising: http://afterburner.mrjones.id.au/ 

    So I hoped the Vevor ones have the same protocol structure and things would be quite easy.

    Today I started with some digging and also found the 3 wires on my heater. They have these functions:

      * black - Ground

      * red - VCC / +5V

      * blue - serial communication

    I've made a Y-cable and then started with my DS203:

    So far so good. The short frame seems to be the request from the remote and the second longer is the response. This is the same communication the devices of afterburner use. Also the hardware for the serial interface seems to be the same - that makes things a bit easier.

    Next, I measured the smallest bit length:

    It seems to be something at 200us (0.2ms) which would mean a baudrate somewhere at 5000 baud. The afterburner devices are running at 25k - so there is the first difference.

    To get a better view of the frames I decided to use my logic analyser (ZEROPLUS Logic Cube ) and the linux tool sigview.

    It looks like a bit width of 210us seems to fit better - that'd be 4800 baud which is more common than 5000 or 25.000 .

    I also noticed, the protocol is looking different.

    The Vevor request starts with 0xAA 66 02 but the ones from afterburner are 0x78 16 00 - so the protocol is different :-(

    So it seems I've to start from scratch.