Split system only dropping 2c then outdoor unit turns off

Ok you are going to all laugh but it is a B&Q system (easyfit) been in for many years. I have 3 of these so can swap PCB's etc. It is the AF9 model so corner mount looks like a bit like a speaker if anyone familiar with it?

So what happens is when you first call for cooling the unit stays on for around 7 minutes achieving a 2c drop according to the display on the front of the indoor unit. At this point the relay clicks and the compressor switches off. Then you get a delay and the cycle continues albeit subsequent cooling cycles are much shorter and it appears like there is a hysteresis of 1 deg C.

So it looks like comms is ok between indoor and outdoor unit. Have took the outdoor unit cover off and there is no logic board or PCB present. Yet to take indoor unit apart but guess this is where the logic board is. Included are the pics of what is in the outdoor unit.

Are these symptoms likely to be a low refrigerant level or logic issue such as thermistor mis reporting?

Any help much appreciated, such as service manual so I can check thermistor ranges etc.

Could be both but would confirm that all logic is performed at indoor unit.

Ok thanks will start with changing PCB between units if that dont fix it then will change thermistors one by one to try and find the rogue one (assume one to detect indoor temp and 1 each on high and low side). Going to struggle to get anyone out I guess as I will be sneered at for the B&Q special. TBH they have served me well so far, cools the rooms nicely runs quiet just no parts backup not sure if any diagnostic ports or lights. Do you know who B&Q sourced them from.

I think it is good to start with changing the PCB. Or-else you may also consult an expert for correct guidance.

I have got to the PCB and measured the resistance of thermistors RT1 and RT2 between a non working unit and working unit.

RT1= RAT
RT2= ICT

Working unit
RT1 (RAT)= 5.25k
RT2 (ICT)= 4.95k

Non working

RT1 (RAT)= 9.26k
RT2 (ICT) = 4.75k

Is RAT room air temp if that is the case then maybe that is the issue as the two rooms were very close to the same temperature they are located on the same side of the house. Assume ICT means indoor coil temperature so they were very close resistance readings so they are ok and I assume sort of proves that the room temperatures were more or less the same? Do you think I am on the right path?

RT2 indoor coil temp was at fault. Today RT2 was 11.8k even when it hasnt been on. changed out to a known working thermistor and compressor is staying on and maintaining the room temp around the set point. Output temp was 4.7c when I set the indoor temp to 18c this caused a slight bit of frost on the indoor coil. When set to a more normal 21c no frost on the coil. The thermistor goes in a pocket that is braised onto the indoor coil (just a bit of pipe on the side).

22c it reads 5.51k
10c 9.25k
9c 9.69k

anyone got any ideas where i can get a thermistor that is around these ranges?

Have a look at the millions of variants available on eBay mate.
Cheap ran easy fix up.
Search for ntc thermistor

Having trouble finding anything that actually lists the resistance @ temp x. Have took pot luck on one that was on amazon for £2.37, looks the same, same plug but no data sheet so who knows until I get it.

Doh I was stupid, the one I ordered was a 20k needed 5k.