can i get somewhere the protocol for the Benwin A/C and Midea A/C or do i really need to use a solution like @
dougy83
is suggesting?
Depends if you can find someone who has already done it or not?.
However, I would suggest
NOT using a photodiode to read it, use an IR sensor IC instead, this strips off the carrier frequency, and provides the actual data that you need. The carrier frequency itself isn't very important, around 38KHz seems pretty universal and works with everything (except some very old B&O TV's, which used 100KHz for some bizarre reason).
I've reverse engineered various remotes over the years, all using IR receiver IC's to demodulate the signal - originally I used a 386 DOS Laptop parallel port, and a program I wrote in Turbo Pascal. Since then I've used the PC audio input and Audacity (attenuate the signal from the sensor first), and more recently I've also used a PicKit3, which has a simple logic analyser built-in.
I've now got a couple of the cheap Chinese logic analysers, so I'd use those if I was doing one now - mainly due to memory constraints on the PicKit3, which only barely has enough space to do it.
As examples, here are a couple of the PicKit3 screen shots - these were 'setup' and 'enter' respectively.
In this particular project, which I did for a company, I was only interested in transmitting a duplicate (without carrier), to create a minimal wired set of buttons to replace the IR remote.
Interestingly, as the IR sensors work by effectively pulling down an O/C output pin, I was able to simply connect an O/C output in parallel with the sensor, so it worked from either the new manual buttons, or the old IR control as well. So it made it easy for any changes to be made, that weren't possible from the new minimal manual buttons. Basically they wanted an 'idiot proof' wired control system, so the people working it couldn't mess with the settings, and couldn't lose the remote.