Request for info: What is the relationship between the UART commands to the indoor board and the indoor/outdoor communication?

[backcountrymountains]

I'm interested in controlling the outdoor unit without an indoor unit for a weird project. I was wondering if anyone knew the correlation between the UART control protocol documented here and the commands/information between the indoor and outdoor unit.

I have read that the communication between the two operates at a higher voltage (maybe 60v) but still might be some form of UART and I wondered if it might be the same information or at least the same protocol but at a different voltage. See this guys hacking here and here for a Gree unit.

Has anyone ever hooked up an optoisolator to the high voltage indoor/outdoor signal line to try and snoop the messages being sent between the indoor and outdoor units?

There exist tools like Dr. Smart and Dr. Smart that are able to control the outdoor unit independently of the indoor unit, but they are also proprietary.

According to this video it is possible to connect and control the outdoor unit using Dr. Smart on a 5v communication port, without the indoor unit. I wonder if that 5v protocol is the same as the indoor unit protocol.

Thanks.

[fmck3516]

@backcountrymountains I just started to reverse engineer the S-wire between IDU and ODU: https://medium.com/p/8b0276fdfb8a

[backcountrymountains]

@fmck3516 Heck Yeah! I am not a smart person but I really think it's possible to have Claude write a script to correlate setting parameters using wifi control and the info coming off of the UART connection. I also wonder how well Claude could interpret the signals if you showed it this repo and the other sources I listed. Heck, Claude might already have it all in it's training data and be able to get it working without any additional prompting.

[fmck3516]

@backcountrymountains I'm also pretty optimistic that this protocol can be reverse engineered. I'm currently waiting for the delivery of one of those inverter testing tools. Once it arrives, I'll know the exact value of any parameter at any given time, which should make it easy to trace it back to one of the data frames.

BTW: I had no idea that these testing tools exists until I saw this issue.