View Full Version : Where to start?

17-11-2014, 07:49 AM
Hi all, new here and was hoping to maybe get some pointers on a project I'm working on .. not sure if this is the right subcategory though? Basically I found a brand new mini split condensing unit for $30. It's fairly small at only 800W input power, but should be more than sufficient for my needs. My intentions are to use this to cool water (or some other fluid). I purchased a brazed plate heat exchanger and fitted it directly on the unit. Obviously this setup is completely different from what the unit is intended to do, so I realize there's a bit of "hackyness" going on here. This is just for my own personal use, so I'm not overly worried about whether the unit might only last a couple years running like this etc.

Basically I'm a bit lost on how I should be checking refrigerant levels on it. Being a mini split, it has an internal expansion device (just capillary in this case), so the discharge side literally just goes through the 1/4" tube right into the heat exchanger. That said, it only has a service valve on the suction side, so it's basically my only indication of what's actually going on inside the system. I noticed however that both the suction and discharge lines freeze up quite rapidly, so I'm thinking it may be low on refrigerant. I stuck my gauges on the suction service valve and measured probably around 25psi - with ~20C water running through the heat exchanger. Turned off, the system sits at about 110psi. This seems quite low, but I realize the pressures and temperatures are all dependant on the load.

I'm just trying to get my head around the proper way of measuring such a unit .. should I just connect it to the garden hose and let "ambient" temperature (28C) water flow through it and see what the pressures are at then? Or should I be regulating the water flow so that it comes out at say, 1C?

The main problem here is not having access to the high side pressures, but apparently this is common on mini splits, how do you know if they're over/undercharged?


Rob White
17-11-2014, 08:49 AM

You need to find out the duty of the original evaporator
and find out the duty of the new one. If there is a difference
then you will either see what you see, the plate heat exchanger
could be too small or if it was too big you would see high suction

On a small split the expansion device (the capillary) is on the outside
unit, that means the smaller of the two pipes is an extension to
the evaporator and is called the expansion line. This line needs to be
insulated to ensure the refrigerant does not boil off too early (before the evap)

The outside unit and the evaporator are designed to be at least 2 or 3 meters
away from each other to ensure no liquid return to the compressor. If the plate
heat exchanger is too small and if the two units are too close some of the liquid
will return down the suction pipe and cause it to frost over.

you need to size the two units correctly and then remember an AC system
is designed for air at +20 so the evap temperature will be about +5 ish.
If you want water at +1 then the evap temp will be down at about -3 to -5 ish
so that will make the suction line frost a little also.

You need to size these two correctly................................



17-11-2014, 09:02 AM
Thanks for the reply Rob. The heat exchanger states it's good for up to 5kW refrigeration capacity, though that's basically all I'm going on.

I just did a test out in the yard, running the garden hose into the heat exchanger, water entering at 28C, I regulated the water flow until it was coming out 10C cooler than it was going in (giving it a few minutes to stabilize). Using the formula: energy (kJ) = 4.19 (heat capacity of water/kg) * 2 (mass in kg) * 10 (delta temp), gives me 83.8kJ of energy. Now, it took 40 seconds to cool 2L of water, so 83.8kJ/40 seconds = 2.095kJ/s. 1kJ/s = 1kW, so that tells me I have a cooling capacity of about 2kW.

It's rated at 800W input power, so with 2.095kW cooling capacity, that'd give me a COP of about 2.6 ?.. going by that it seems it's not doing too bad? I should have also stated it's a basic R22 unit, not inverter.

Running like this, the suction and discharge lines are nice and sweaty. Dropping the flow rate much beyond this, the heat exchanger and lines do start freezing up a bit.


The Viking
17-11-2014, 11:29 AM
Hi Things,

First of all, I do agree with Rob's comments above. You need to ensure there isn't liquid flooding back to the compressor, if the suction line is too short then a loop of it will help protect your compressor.
Also the design of flows in the plate heat exchanger is important, water in at the top out at the bottom and refrigerant in at the bottom out at the top, creating a cross flow would be the norm.

Now, to answer your original question;
The only way to check the refrigerant charge in a setup like yours is by measuring the subcooling. You will need to fit an access port on the high side of the circuit, ideally on the liquid line between the condenser and the capillary tube, and the measure the difference between the saturated temp of R22 @ the pressure you get and the actual liquid temperature. When you get a value between 4K and 6K your system got the right charge.


17-11-2014, 12:33 PM
Thanks for that .. might be a bit out of reach for me though, been relying on being able to pump the unit down into it's condenser coil to work on the heat exchanger - can't really do that if I have to install a valve :(

It's a very basic unit internally:


It does indeed have a reversing valve, I've tried running it in reverse but while monitoring the temperature it was climbing a bit high for my liking. I have a feeling the heat exchanger is probably just filling with liquid - but I don't really intend to use it like that :)

Correct me if I'm wrong, but isn't that black cylinder on the compressor an accumulator? Should prevent any liquid getting back into the compressor shouldn't it?

17-11-2014, 02:13 PM
The black cylinder on the compressor is a filter not a suction accumulator

17-11-2014, 07:44 PM
Have a look behind the electrical box of the unit, there sometimes a high pressure point located thereabouts.