Moisture in POE or PVE oil

[Peter_1]

If POE or PVE oil has taken up moisture(compressor stood to long open for example), can it be taken out again by vacuuming it (low enough and long enough)? Anyone experienced already with this?

[monkey spanners]

I think the moisture chemically changes the POE oil, not sure with PVE.

[Grizzly]

Hi Peter.
Some may have tried?
However whenever I have encountered a moisture laden synthetic oil.
I have drained and replaced the oil, did this system not have a crankcase heater?
The absorption properties of some oils are much greater than a vacuum can remove.
Some will absorb 10% of their volume overnight if left exposed to the surrounding atmosphere.

IE the cap is left off the can.

Grizzly

[Peter_1]

Grizzly, these are scrolls in VRV machines and their's no drain on it. So removing the oil without taking out the compressor is very difficult. There's a crankcase heater which can help to vaporize the diluted moisture in the oil. After analyzing a sample of oil, we foud nmoisture in it but question remains if this could be taken out by vacuum it properly. Is the moisture somehow a prove that teh vacuum process wasn't done very well.

[al]

Earlier discussions on this suggested that the moisture combines with the oil to form a new compound which vacuuming may not remove, i'm not so sure though surely a deep enough vacuum, triple evacuation would remove a large percentage of the moisture allowing the drier to trap the rest?

al

[Brian_UK]

Interesting little article...
http://www.rses.org/assets/journal/0110_MSAC.pdf

This also seems to be common comment - "POE oil bonds with water on a molecular level, and a filter drier is the only way to remove it."

Also a report, below, makes this statement - "It has previously been reported many times that Polyolester will hydrolyze under certain conditions and Polyvinylether will not hydrolyze because of its ether structure. This can be reconfirmed through this study."

http://docs.lib.purdue.edu/cgi/viewc...5&context=icec

[mikeref]

Nope. Surface tension of various grades of oil is prohibitive Peter.

[The Viking]

Peter,
You had to open that big can that clearly stated "WORMS" on it, didn't you...

So here goes:

I once had a chat with a chemist involved in the manufacturing of POE oils and asked the very same question.
(Well, my question was more in line with "what does it mean that the new oils are hygroscopic and what impact does it have" but anyway)

The answer started something like this:
"said refrigerator oil contains as base oil components a polyol-ester oil formed by reacting a fatty acid with a polyhydric alcohol selected from Pentaerythritol (PET), trimethylolpropane (TMP) or neopentylglycol (NPG), to which are added 0.1% to 2.0% by weight of tricresylphosphate (TCP) and 0.01% to 10% by weight of an epoxy compound, wherein the epoxy compound comprises glycidyl ether or 0.01% to 10% by weight of carbodiimide"

To you and me it roughly translates to:
POE oil is made by mixing an acid with a lot of other chemicals and getting a very acidic and corrosive mixture that will eat through most metals (incl. copper and steel) if left in that state. Only by removing H2O (water) molecules from this mixture will it neutralise and become non acidic.
When they make the oil, the moisture is removed by vacuum.
Unfortunately the end result is a hygroscopic oil (it absorbs moisture) and as soon as it get any moisture in it it will become acidic.
The good news is that the process can be endlessly repeated, by exposing the oil to a deep vacuum the moisture will be removed and the oil will again become neutralised.

How deep the vacuum needs to be depends on the temperature, generally it is accepted that 500micron (1/2 Torr) will do the trick. The important thing is that the vacuum is standing, without any increase in pressure. This is why a good digital vacuum gauge with micron scale is needed. When you reached the desired vacuum level, isolate the system and turn the vacc pump off, any increase in pressure from then on indicates that there still is moisture in the oil.

Deeep breath...

Class, any questions?
Are you still awake?



.

[mikeref]

Viking. If the PPM of moisture in the oil is low enough and several days are allowed on vac, then i might accept your reply. If there is enough moisture to form a small "Puddle" of water under the oil... i would still stand by my previous post Nice breakdown in your description though.

[monkey spanners]

So you'd need 500 microns at the bottom of the sump etc, the weight of oil above would have an effect on the level of vacuum throughout the oil i think?

[mikeref]

So you'd need 500 microns at the bottom of the sump etc, the weight of oil above would have an effect on the level of vacuum throughout the oil i think?

G'day M.S. Going by your last video in "what we did today" i think there was actually OIL contamination in the compressor, rather than moisture

[monkey spanners]

G'day M.S. Going by your last video in "what we did today" i think there was actually OIL contamination in the compressor, rather than moisture

Think that one was using water as the lubricant! Would have been interesting to try and get it running again but they scrapped it.

[The Viking]

@ Mike,
In your scenario with the water the vacc gauge would rise rapidly, if you got a (deep)standing vacuum there is no more moisture left. You would be surprised how fast this can happen, it all depends on volumes and just how much water there were to start with. But you do need a decent digital vacc gauge to tell you that there isn't any rise of the pressure.

@ Monkey Spanners,
You are of course right, the weight of any fluid will have an impact on pressures.
As nature would try to keep the concentration of water molecules equal within the oil it should be enough if the surface of the mixture were at a deep vacuum, it would just take longer to boil all moisture off.
Also, 10 meter of water column = 1 Bar. Oil is slightly lighter but anyway, how deep is the oil level in a normal sump? 0.05m / 0.005Bar?


On the other hand, has anybody ever heard of a compressor failing because it had the oil changed once too often???
(Just remember that the new oil when transferred from the can to the compressor most likely will absorb moisture and therefore require a vacuum before the compressor is recommissioned, not like the good old days when you just left a bit of pressure in the crank and pumped the oil in before you restarted the system)

[Peter_1]

Thanks so far all of you. You can't believe how important all these answers and practical experience are right now ;-) The future in refrigeration business of a good colleague depends now on such a knowledge. See also my other questions in the same section. All related to the same problem.

[al]

Viking

Great reply, i've to slut some love around before being able to rep you!!

any chance you would have that in an ahem "official" publication?

Mike, wouldn't "boiling" the oil break the surface tension allowing the moisture escape?

al

[mikeref]

G,day Al. Have to say that i've not been down that road. If compressor oil is known to be contaminated with moisture, i'll find a way to change it rather than spend a lot of time on vac.
There was a job many years ago where i spent 8 days trying to remove moisture.

[al]

Very true Mike, but the design of these units defeats oil changes without literally tipping the compressor on its side!

[Fri3Oil System]

Hi!

When you have moisture in circuits containing POE oils, the contamination grows gradually with the exposition of the oil to the humidity. The problem to remove the oil, is that it gets trapped in the circuit syphons and you cannot remove it just with a deep vaccum. Moreover, in case you have water, the oil film will remain above it, avoiding this way the water evaporation.

Thing is, if you have not removed the whole content of contaminated oil, and moisture, the next charge of oil will get contaminated as well, and create a corrosive ambient in the circuit at mid term.

So, firstly, you need to totally eliminate the oils, and then remove the moisture with vaccuum and nitrogen as usual.

Regards,

Nando.