PDA

View Full Version : How does an inverter fan coil communicate with its condenser?



kevin090972
09-01-2007, 03:40 PM
Hi Guys,

Can anyone please advice me how does an inverter's fan coil communicate with it's condenser? With a conventional aircon, the blower either cuts in or cuts out, very straight forward. But what about with an inverter system? What sort of information were sent across and is there any way we can test if the blower is at "high load" or sufficiently cold using just a multi meter?

I know that the inverter will go "full speed" when we initially turn the aircon on. During troubleshooting, if during full load and the inverter doesn't go full speed, something is wrong. But how will i know if the problem lies with the blower signal or with the condenser side? How do i test the blower signal?

Wll really appreciate your kind reply, Thanks!

Kevin

adams?
09-01-2007, 10:07 PM
what unit? the inverter normally only operates the compresser

Peter_1
09-01-2007, 11:20 PM
And the signal is mostly not coming form the indoor unit but from a pressure signal - or a converted pressure signal - of the LP of the outdoor unit.

I shoudl say, if you don't have the proper diagnostic tools from the manufacturer himselves, impossible to measure this, I don't speak even here not of repairing if you ever could find the reason for it.

Just replacing PCB's and sensors ist he only thing you nowadyas can do with the modern aircons

momo
09-01-2007, 11:33 PM
By fan coil we assume you mean indoor unit for cooling/heating ?
Assuming power being delivered to indoor unit: Interconnections on switched motor A/C units range from simple Earth, Live (compressor+fan) , Neutral <for cooling only> to E N Compressor Fan Reversing valve and wiring for temperature sensors (usually suction line) or more...

Consider what happens with an inverter: you must signal what cooling/heating power you want, also where is the "intelligence" of the unit inside or outside which decides speed, valve settings, defrost... or detection/indication of faulty conditions :cool:.

Communications usually becomes serial on a 4 cable format: E L N Signal (single or bi-directional).
This is where it gets manufacturer and model specific and I think it is time we started hacking into these signals since it could help with fault detection and modifications! :D
Anybody come across studies on the types of comms. used (super-positioning of signal on mains AC/DC, baud rates, data packets etc)?;)

This is one of the areas where good vs cheap engineering and design is important or you have fried PCB if a fault occurs on the mains or wiring.:)

inverter
10-01-2007, 12:09 AM
Theres lots of different ways. thermistor's, pressure transducer's,or a dedicated sub cool fridge circuit IE mitsi vrf.

kevin090972
11-01-2007, 08:14 AM
Hi Guys! Thanks for all of your kind replies! Cheers! Appreciate it!

laf100
13-01-2007, 12:44 AM
Many VRV/Inverter system faults can only be fully/properly diagnosed by connecting a laptop service tool, and using the manafacturers software, but training courses and the all important service mamuals are a big help too!

paul_h
21-07-2007, 04:14 PM
Modern a/cs are three wire interconnect, that is active, neutral and signal.
The indoor unit communicates signals to the compressor information regarding it's coil pipe temp, it's air on temp and tells the compressor what load it requires based on the difference between it's air on temp and what the desired setpoint is.
The compressor gets that info and runs at a speed required, but also uses it's own sensors (discharge temp, crankcase temp and outdoor coil temp) to over ride any signal from the indoor demands so it doesn't overheat or ice up.
For example, say on heat, if you want it to run at full speed, but the indoor unit pipe thermistor is above 50C, it will think it's too hot and start slowing down to bring the temp down, only ramping up again when it gets below a certain temp. The outdoor pipe thermistor does the same thing on cooling, slowing the comp down when the coil is too hot, or the crankcase or discharge sensor.
So if you want it full speed and it's not doing it for you, it could be either the indoor or outdoor thermistors telling it to slow down.
You really need a set of gauges and a touchprobe to check discharge, compressor crankcase, outdoor coil and indoor coil temps, or a bunch of your own thermometers strapped to those points to find out what the system is doing.
I've never seen them run flat out on start up either, they normally slowly wind up.
edit: wrong about the HP, they also monitor HP and cut back if thats too high.

I've never seen a I/u just send an incorrect signal though, I've hardly ever seen a circuit board tell it to run at the wrong speed. Either the chip/PCB gets fried, or a sensor goes open or short circuit, and it shows a fault code for that.
Exceptions have happened to me twice are, a mouse chewed through sensor wiring enough for it not to carry enough voltage, and an installer put a return air plenum on a ducted unit in such a way to crush the sensor cable.
I guess the only thing you could do with a mutlimeter to test the sensors is get a chart off the manufacturers stating what DC voltage should read across their sensors at certain temps to find out if they're in spec, that's what I had to do in either of those cases (both fujitsus).
If they're not in the correct voltage for temperature, you have a faulty sensor which affects comp speed. If they are in spec, and all the temps and pressures are in the correct range but the compressor isn't operating at the right speed, then you have a faulty control board.
As to which one it is (and I realise this is your original question), I'll have to bail out on that one as I haven't come across a half functioning inverter board, normally they're pretty dead or the good one has an error code for bad comms with the faulty one as more than just thermistor reading are often wrong, low power signal for example.
I have some info on fujitsus, like the serial is 110Vac, both putting out 20-55V dc each on opposite sides of the sine wave horizontal axis. Got information on the 'bits' it sends out and what they do, but no way to measure that out in the field and never needed to.

Contactor
21-07-2007, 09:31 PM
I once found a faulty discharge pipe thermistor on a Daikin VRV inverter compressor. However, in addition the main board was also faulty and without the service checker / laptop this would have taken much longer to sort out.

Mohamdy
22-07-2007, 09:48 AM
it desn't matter if you know from where the signal is comming to the inverter or not.
Look at the inverter input, it is an analogue input either current 4-20mA or voltage 0-10V so it should give 20mA or 10 V at full speed at the starting
if it is not maximum then there is a problem with input signal so follow it by sequence
Best regards

Peter_1
22-07-2007, 10:32 AM
it desn't matter if you know from where the signal is comming to the inverter or not.
Look at the inverter input, it is an analogue input either current 4-20mA or voltage 0-10V so it should give 20mA or 10 V at full speed at the starting
if it is not maximum then there is a problem with input signal so follow it by sequence
Best regards

Mohamdy, on small inverter driven units used for AC's, you will not find any 4-20mA nor 0-10V converters.
Everything is done with thermistors.
What you're describing is the working principle of an industrial VFD system driven by external devices.

Peter_1
22-07-2007, 10:38 AM
...,or a dedicated sub cool fridge circuit IE mitsi vrf.

Inverter, I'm eager to learn more how this works.

I was once told that Mitsubishi measures the evaporating pressure via the corresponding evaporating temperature.
A small capillay is injecting liquid in a small chamber and is connected to the suction again. So a small evpaorator.
In this small 'evaporator' is a thermistor mounted which temperature is converted again to the coresponding pressure via a simple software conversion.

This is in fact a cheap solution and isn't prone to overpressure an I think that this can work very well.
Most pressure transducers can't widhstand high testing pressures when pressurising with OFN.
Is this correct?