View Full Version : ERN2920 freezes okay then stops after compressor change

29-04-2017, 02:55 PM
Hi all,

I am an amateur doing this with limited equipment, so sorry if I cant give exact pressures/temperatures etc.

I replaced the compressor in an R600a electrolux ERN2920. Have tested all the joints with pressure, no bubbles in the soap, and vacuumed the system, and it holds the vacuum for over an hour without any sign of leaks.

I vacuumed it out for 2 hours, weighed in the 74g of R600a and the evaporator cooled to -20 degrees. I assumed all was good, but 20mins later it started warming. The compressor was still running okay. The discharge from the compressor had gone from hot to tepid.

Suspecting moisture (I don't have Nitrogen to purge it) I vacuumed it out for 10 hours. (I don't have a way to measure the level of vacuum). I also warmed the various parts with a hairdryer as that seemed a good idea at the time. (But was probably pointless!) I recharged it, and got the same result. ie. It cools well, then after 20-40mins it stops cooling, and the discharge pipe goes from hot to tepid. This time I measured the suction tube pressure throughout the test. It remained at around 8psi when working and after it had stopped working.

I still suspect moisture possibly jamming the compressor valves? But that seems a random event, where this I can reliably reproduce on every test. It works for a while, so I am thinking that its not a full blockage, so maybe moisture is forming ice and is blocking the cap tube? I have not replaced the drier filter, but since it works for a while how can that be the issue. Its also awkward to get at and shield from the torch. I still suspect moisture forming ice, but before forking out 50 quid on nitrogen I though I would ask the experts!

I have argon gas, and I let that run through the system when changing the compressor, no oil or other muck came out. I let it bubble into water. The water was clean after. The flow was steady.

Any help is appreciated :)


29-04-2017, 09:04 PM
Has the system got hot gas defrost? If so it sounds like a problem with the hot gas defrost.

01-05-2017, 12:40 AM
Thanks for the reply. I am not sure what gas defrost is. The spec in the owners manual says "Automatic defrost in fridge" (not sure what that is either!). But if this was a defrost mode, why would the compressor be still running?

The Viking
01-05-2017, 01:24 AM
To ensure your system is free of moisture your are better of spending your 50 on a vacuum gauge than on OFN.

Your pressure reading of 8psi, how reliable is that? The reason I'm asking is because the saturated temperature of R600a at 8psi is 0 degrees C... Either the pressure reading is out or the temperature couldn't have reached -20. To reach -20 the R600a would have to run at around 5psi BELOW atmospheric. (Which makes R600a a questionable choice for the application).

So, you have used refrigerant from a cylinder marked R600a and not patio-gas or other butane products?
If this is the case and considering that this system will run at or below atmospheric pressure, the most likely scenario is that you got a leak somewhere on the suction side of the refrigeration cycle.
One easy test to carry out, that doesn't rely on bubbles, is to pressurise your system with OFN (or Argon will do in a pinch) to around 80psi and leave it for 2-3 days - if it still got the same pressure at the end as you had when you started then it is unlikely that you will have any leaks.


01-05-2017, 02:26 AM
Thanks for the reply.

The charge was done with R600a from a reputable supplier. Not patio gas! lol!

The 8psi was at the suction tube. Maybe its not very accurate. I am using a charging manifold and that gauge had a scale of 0-350psi, so I would not expect it to be too far out.

If there was a leak, wouldn't it fail and stay nonworking? Maybe I didn't point this out, but if I leave is off for 20 mins then turn it back on, it starts cooling again. (Then, after the evaporator gets down to -20 or so, it will then stop cooling and warm up. The compressor remains running and the output pipe is no longer hot.) I have not left it long in this state in case something is getting damaged.