Oil not separating in the LP

[Superfridge]

Hey all,

System, NH3, Mobil Arctic 300 (originally on Mobil SHE 226E changed over 5 years ago), +30degC condensing, -1degC evaporating, chilled water to 2degC, thermosiphon cassette type plate heat exchanger for evaporator, plant about 20 years old, automatic venturi oil return system from PHE to compressor suction, compressor chucks oil, oil in the oil separator is very foamy.

Situation, poor heat exchange in the evaporator due to oil logging. Oil return system working normally.

I'm wondering if the oil is somehow in suspension in the ammonia, the gravity oil draining is counter flow to the thermosiphon. The LP level glasses show the ammonia to be slightly cloudy. Or perhaps the oil is impinging or adhering to the PHE plate surface. Or maybe the oil has broken down and is mixing with the ammonia? Or maybe the oil thrown from the compressor is greater than the rate of return? The ammonia may be quite wet with water. Used to be ok, only a problem in the last 2 years. Site oil analysis indicates no issues. The water side of the PHE is clean. Sorry lots of questions but not many answers.

Has anyone experienced similar? Thanks.

[RANGER1]

Not relevant

[josef]

Superfridge,
separating mineral oil is worse than synthetic oil.
Turbidity of liquid ammonia is common for Mobil 300, it can be caused by water, high discharge temperature ...
I think that if you want to put everything in good condition, you have to use Mobil SHC 226 E again, drain the old oil by hand.

My experiences and thoughts, Josef.

[Superfridge]

RANGER1 see below in red
Can you give some more details on compressor
typeno but its a pig!
oil temperature 50degC
discharge temperature 80-90degC
does it have a coalescer yes
have you contacted Mobil no, site deal with this
have you had oil test no, site deals with this
have you pumped out evaporator & drained oil, check it’s condition no, but oil was drained during shut down so the oil will separate over time
can you do ammonia moisture test yes its been done a few times and yes it is wet

Josef your experiences are invaluable, thank you. "Water and high discharge temperature's" we have both. From what I can preen out of site personal the oil was changed by "others" from synthetic to mineral due to numerous ammonia leaks on compressor bodies. The 226E is hard on O rings. A change back to synthetic could open up a can of worms!

[RANGER1]

Not related

[Tycho]

can you do ammonia moisture test [/COLOR][COLOR=#ff0000]yes its been done a few times and yes it is wet

And there's your problem.

This is going to be a long one, so buckle up

I can't explain it, but on a dry thermosyphon system, oil is not a big issue unless draining is neglected and starts filling up into the PHE.

Once you add moisture to the mix, what I have seen is exactly the same as you... heat exchange gets worse, and not really draining any oil unless it's after a longer standstill.

Of course the degrading of the ammonia because of moisture content is one thing, but when you add moisture in to the system the problem seems to become ten times as bad.

I don't have any science on it, just my experience.

I had a brand new termosyphon plant where the compressor dumped 40 liters of oil out into the system, drained it and the plant was just fine, no symptoms of loss of capacity after the oil was drained, it worked just as it should.

However, on older systems, I tend to see a deterioration in heat transfer after about 7-10 years.
And even though they are all positive pressure plants, when we do a water test or send in an oil sample, we find that the moisture content is right at the limit of where it goes from Anhydrous ammonia to becoming fertilizer.

Some of that may be from sloppy techs when replacing lost ammonia or replacing oil, but not all of it.

In earlier years, when oil was delivered in the round 20L metal containers, I was taught to always shine my flashlight down into the container to look for a grey circular blob on the bottom, which was of course moisture that had penetrated into the container if it had been stored for a long time, or in an outside warehouse, or whatever.

I don't know if this is true, but according to the guys that trained me was that the oil had taken in all the moisture it could, but being so hygroscopic it kept attracting moisture and that moisture would gather as water in the bottom, and if the blob was there, no matter if it was a penny or a pancake, to not use that container

We would complain to the supplier, but of course, we had broken the seal to inspect it so I don't think we ever got any compensation.

When the moved to the plastic containers it got a lot better, because there is probably a better seal, but after I pop the cap of one, I always shine my flashlight into the opening and stick my eye real close to try to get a look at the bottom, and maybe one in two hundred I can see a coin size of moisture, and I won't use it.


Also the ammonia itself, all my time I have received Anhydrous ammonia from a reputable company and have almost never had any problems...

Almost...

Now my connection set up is always Bottle -> Adapter -> Sightglass -> ballvalve -> hose -> ballvalve -> plant.

I want to see when the bottle is empty, and I want to be able to close off my hose, and in case it's not safe/smart to purge it on site, I want to take it somewhere where I can purge it safely .

One time, I was charging an RSW system, fresh bottle, straight from the company we always got it from, heat treated plastic seal is intact, lead wire on the protective cap is intact... all good.

Hose and everything is connected, open the bottle, see fog in the sightglass, no smell, so my connections are good, open the ball valve, and instead of seeing a grey flurry of liquid moving through I saw first gas movement and then *blop* golden honey, as in there was oil, so I closed the ball valve, closed the bottle and disconnected the adapter, sure enough it had been oil.

it was after that I made sure to have the sightglass on at all times and always monitor the first minute of ammonia going through.

The oil thing happened one time out of I don't know how many tons of ammonia I have handled, but it still made me take extra care

Same kind of sealed bottles, this has happened maybe 3-7 times in 25 years.

Same procedure, open bottle, check for leaks, open ball valve, watch for about a minute, then since you know it's going to take 15-20 minutes to empty the cylinder, go work on something else.

That's what I did all of these times.

Now we all know if there is some distance between the cylinder and the plant, fold a pigs tail at either end.
In my case I may be 2 decks down in a ship with 30 meters of hose up to the cylinder that is on shore, so I fold a pigs tail on the hose on board, when I see it stop vibrating/pulsating I know there is no more liquid coming so I have to go up top to change cylinders. I also knows it takes about 15-20 minutes to empty the standard 42Kg cylinders we have in Norway.

So new cylinder connected, flow in the sightglass, pigtail pulsating on the pier, I go back on board, takes me a few minutes to get to the compressor room, hose is not pulsating close to the liquid receiver (sometimes it doesn't at that end until the cylinder is nearing empty)... no worries, continue whatever job you are doing, look over at the hose from time to time, after 30 minutes hmmm suspicious, so head back up to the pier, sighglass is cloudy, can't see any movement, release the strap holding the cylinder in the cage, tilt it, it's still full.

Close both valves and disconnect, the sightglass is filled with ice. So that cylinder was contaminated with water.

So I have figured, in my mind, if you get Anhydrous ammonia in 42kg or whatever cylinders, you can never be really sure that you get clean Anhydrous ammonia.

Just like on OFN cylinders, it says never to drain them below 2 bars, never pull a vacuum etc.

But If I had been a Refrigeration engineer in a bind and needed somewhere to store NH3 when I was in a bind, it wouldn't have taken me 2 heartbeats to use an empty NH3 cylinder as a temporary storage

And since the manufacturer is probably at a large scare, vacuum will be done through large vacuum pumps attached to large tanks, so no drop test, just "vacuum done -> next station".


And Oil management... In my younger years on R-22 plants, I often encountered 2 shifts on a ship.
Shift one would drain oil from the compressors oil separator because the level was above the sightglass so they couldn't monitor it.
then 12 hours later the new shift would come on and put that same oil back in, because they could see the level, and probably add a little extra just to be sure.

This way of doing things on ships and other plants followed me until the older guard started retiring and younger people started taking over, with better education and after taking over from someone who had been there for 40 years they knew they weren't specialists and they started asking questions.

My advise to an engineer is "as long as you see the oil in the bottom of the sightglass but it is above the low level trip, don't add any... if it drops below the sightglass and you need to add oil, add an entire 20L drum, never ever add 1 liter, 5 liter or 10 liter and then leave the drum open for a month, use it all at once or discard the leftovers right away... Refrigeration oil is very hygroscopic, so use it or loose it

So that was a short story made very long that you shouldn't always blame the operator or the service personnel.

Sometimes they are just dealt a short hand by the fate of the refrigeration gods


This issue you are mentioning, I've seen it more often on spray chillers than on Thermosyphon chillers.

At my previous employer, I commissioned around 100+ spray chiller plants...
These were smooth tube titanium chillers, so the tubes inside the chillers were smooth, not serrated or anything.

There was one ship, where the Delta T between the suction temperature and temperature of the water was so high that it was unbelievable, the compressor suction pressure was maybe -7°C and the temperature of the water leaving the chiller was 5°C with a flow of 700m3/h.

there was just no way that computed in any simulation of chiller capacity to create gas amount to the compressor...

So we emptied the plant, removed the spray nozzle pipes so we could look into the chiller.
When i put my eyes to the opening with a flashlight, my first reaction was "wow... all these pipes are brown, shouldn't they be like shiny titanium?"

There had been other plants with the same issue leading up to this where various things had been tried, so at this stage I was sherlock holmes trying to test everything to see if I could solve the issue.

First, the flange hole was just about large enough for me to get my hand in to my elbow

First thing I did was to stick my hand in and rub and scratch along the surface with the ball of my finger and my nails, the surface didn't show any sign of even having been touched.

Next was to spray in a copious amount of brake cleaner (pure alcohol) and then reach in and rub it with a rag doused in the same brake cleaner.
It was still brown.

this baffled me a little bit, because Alcohol is both the solution and demise of so many things.

there were two nozzles on this chiller, so I took a rag dunked it in water, put my hand through one opening, looked through the other opening, then squeezed out a few drops. they hit the pipes and just fell of like they were covered in teflon.

After that I went at the pipes with a 3M polishing pad and it took me 20 minutes to get down to titanium on even the small patch I was working.

Not taking any credit, all the while I was doing that, there was a crew of professional cleaners from the oil industry rigging up.

They circulated industrial soap at 80°C for 24 hours, completely environmental friendly so they dumped it overboard, and then circulated 6 hours with fresh water to remove any soap residue...

When I looked into the chiller after they were done, the titanium pipes were shiny silver, and when I did the drop test from the rag, the water dropped down and stuck to the pipe instead of just bouncing off.

-------------------------------------------------------
end of story, on to questions :
-------------------------------------------------------

So in your system, you say poor heat exchange due to oil logging, but oil return system is working normally.

Venturi oil return back to suction, this is working as normally? how do you determine this? by seeing movement in the oil return sightglass?

I know that the Johnson Controls systems can get clogged from time to time if there is too much oil throw out, just last year a colleague drained 362 liters of oil from a JC plant that was running with a Venturi oil return...

The sight glass of the oil return was showing a nice flow, just as it should, but the delta T between the ammonia and the water was strangely high so in the end he drained 362 liters of oil from the chiller and added same amount of nh3.

of course the operators from beginning to start said "no no no, we never added any oil unless it was needed"

[al]

There's a lot of knowledge and experience in that post Tycho, always learning!

[Superfridge]

Refrigeration is a funny thing, we see stuff and just take it for granted that "it is what it is" with out knowing really what happens, we have to take educated guesses as we cannot see inside, like electricity and electron flow, we always test conventional current flow, well mostly. Sometimes i think maybe I should have gone to university and got a degree in chemical engineering or something so I could understand what is really happening, go right down the rabbit hole, which is what my post is all about.

Something is happening where the oil is not separating from the ammonia and you good folk have reassured me that what i am seeing is what i think I'm seeing and I'm not going bonkers. Degradation of the ammonia and the oil is allowing the two to mix and I think that it is fair to say that is clear now.

Yes, I know how to fix and yes RANGER1, they are cheap skates. It will be their decision.

Tyco, your experiences ring true. Sight glass? yes always. New plant with foul gas, why? Because it was in the new ammonia drum, we get what we are given, haven't seen oil yet but one day. Your question regarding the oil return, yes its returning but mostly liquid ammonia, the oil is hiding and yes oil and water can mix, google says so. We have had new oil samples analyzed for a base line and the results come back with a reading like someone spat in the sample. I'm going to read your post again. Thank you.

[mbc]

Hi

for DX systems I do this drawing and it works well

direction is different with other drawing .
Our DX in let from top. and out let with U goes to ECO.
any oil goes down and collect in U then moves to Compressor by ECO ports

drawing for the thermosiphon in NH3 and R 22- R404A ..... gases is a little different
pipe size and Circulation Rate .
hope draw and able to send .
MBC


MM.pdf