Little late,but...better than never
So finally explanation of my idea for controlling the AC.
Since it is a DC inverter,simply applying voltage to the compressor,doesn't work,because there's a PCB which converts single phase 230V 50Hz mains to 3phase voltage 120V (according to my memory,it could also be e different voltage,but it's there about) with variable frequency,usually 20-130Hz.Making VFD from scratch,and in home for such an application (kW of power,high voltage and current) combined with all the necessary protections is hard and time consuming,so there should be an easier way.Than i realized that all the above is made in the factory and is sitting in front of me,in the AC's case.So how to use it?.Simple-fake the control PCB of the AC,so it "thinks" something that in reality is not true.More precise-the room temperature,and the discharge pipe temperature.Since i need only an on/off mode of operation (or full-throttle),not a real speed regulation,it's sufficient only to simulate the pressing of the power button of the remote control to switch it on or off.I made that by capturing the codes sent by the RC (power on,power off,temp.up,temp.down and mode (cooling/heating) with the IRLib for Arduino(
https://github.com/cyborg5/IRLib) and send them in the appropriate time to the AC by an IR LED,which i temporarily installed in front of the IR receiver of the AC,but in the final version i'll remove the receiver and solder the wires directly to the AC's PCB.
The next thing is to make the control unit "thinks" that the temperature of the discharge pipe and the room temp are correct for heating mode.The way i solved that problem is by replacing the original room & discharge thermistors with fixed resistor.The original discharge pipe thermistor was NTC 10k,which i replaced with 4.7k,and the room temp thermistor was 25k (or there about),which i replaced wtih 33k resistor.I have checked thermistor's values with a cup of hot and cold water,and chose the values of the resistors,so that if the room temp is 15C and the discharge temp is about 55C.Now when the AC thinks that it's cold in the room,it runs at maximum power,and when the discharge temp it's not changing,it doesn't go in to defrost mode,since it's not detecting temperature/pressure drop due to the icing of the exchanger.So the next problem to be solved is how to really defrost when it's needed?Unlike the compressor,the four-way valve and the fan are simple devices-it only need mains to operate
.So the most obvious way to control them is to apply or cut-off the voltage on them.I have made this by putting a relay in series with the valve and with the fan,so i can turn them off independently of the AC's control board.In that way ,for example when the AC operates in heating mode,the valve is powered.When i need a defrost,i simply cut-off the power of the valve and the outdoor fan,using the relays,wait until the exchanger temp reaches 6C,than turn on the fan to blow out the moisture from the exchanger and 10sec after that i turn the valve on again.That way the AC thinks,that it's heating the room,but it's really in defrost mode,and when the defrost is done it really starts to heat the tank
.Luckily the control board of the AC is simple,so it doesn't watch for outdoor fan's speed and no additional circuitry is required to emulate the working of the fan.The last problem i've solved was how to determine when the compressor is running or not,since again because of the fact that it's an inverter,the outdoor unit is supplied all the time,and simple opto-coupler circuit for detecting voltage is not going to work here.So i've placed a current transformer to measure the AC's current and in that way determine if it's ON or OFF.
That's all for now,because it's bed time.Maybe tomorrow I'll put some code with explanations and some pictures.Good night