Frequent refrigerant charging

[desaipt2003]

We have an R134a process refrigerant system using oil flooded screw compressor. We are facing the problem of need of frequent refrigerant charging and oil charging since commissioning of the system in year 2000. Oil level falls down within 2-3 months in the oil separator. Efficiency of the system fall down in 2-3 months of operation and getting improved after refrigerant charging. Visible oil leak appers from mechanical seal of the compressor. Mechanical seal has Cast Iron-Carbon face combination. I conclude that refrigerant is also leaking along with oil from mechanical seal. I have proposed new type seal with Silicon Carbide-Carbon face combination. What the other things I should check for the recurring problem of refrigerant leak? Please give the comments.

[icecube51]

recup the refrig,and do a high presure test (30Bar) whit nitrogen.

ice

[desaipt2003]

Thanks ice,

Discharge pressure of compressor is 12 bar and we are doing the pneumatic pressure test of the system at 10 bar. We are not going to high pressure than that because of risk involved with pneumatic testing. At that pressure we are not getting leakage from the loop.

[Scramjetman]

We probably need a bit more data on pressures and temps around the system.

If you are not getting a leak with your pressure test over a couple of days, then the refrig & oil must still be in the system and something else is up.

When it is operating what are you pressures and temps, what is ambient and load temps and what is the superheat and subcooling doing. Once you have all this data, usually the fault can be narrowed down to a section of the process.

If these figures check out and it shows a loss of refrigerant, then we need to investigate the pressure test process and verify that it is being performed correctly.

[nike123]

If you are not getting a leak with your pressure test over a couple of days, then the refrig & oil must still be in the system and something else is up.

Visible oil leak appers from mechanical seal of the compressor.

.............

[bill1983]

when you say you are recharging, are you simply topping up or are you actually recovering, weighing in and out. as regards a pressure test, why not isolate the compressor and test the remainder of the system up to manufacturers spec or at least to 20 bar. make sure that any electrically operated valves are open. you can then prove the fault on the compressor shaft seal, or not. also is oil being logged in another part of the system?

[Grizzly]

We have an R134a process refrigerant system using oil flooded screw compressor. We are facing the problem of need of frequent refrigerant charging and oil charging since commissioning of the system in year 2000. Oil level falls down within 2-3 months in the oil separator. Efficiency of the system fall down in 2-3 months of operation and getting improved after refrigerant charging. Visible oil leak appers from mechanical seal of the compressor. Mechanical seal has Cast Iron-Carbon face combination. I conclude that refrigerant is also leaking along with oil from mechanical seal. I have proposed new type seal with Silicon Carbide-Carbon face combination. What the other things I should check for the recurring problem of refrigerant leak? Please give the comments.

Hi
Along with all the good advise, may I add.
Have you considered that as the compressor is in it's 9th year?
There may be just a little bit of mechanical wear within the compressor?
The type of sealing faces in the shaft seal is irrelevant in this case i think.
I may be proved wrong.
If the shaft seal is leaking because it's viscous seal is being disrupted.
Then you would loose refrigerant and your oil return and oil levels in the separator would drop.
Efficiency etc would diminish.
I suspect you have lateral play in the drive shaft of the compressor. Which through this excessive linear movement is breaking the oil seal between the 2 faces. Causing as you have noticed the s/seal to leak.

Obviously you will know the maintenance history of this compressor and are in a better position to decide as to what is causing the shaft seal to fail.
Alternatively the seal may just of worn out and is due replacement?
What do you think?
grizzly

[desaipt2003]

Thanks Grizzly,
History says that mechanical seal develops leakage after approx. 15 -30 days of replacement. Rate of leakage is not in great extent that the system can't be run. Leakage rate used to be 30-40 drops of oil per minute against acceptable limit of 1-6 drops/minute. When contacted Howden, they are telling that new age compressors are equipped with silicon carbide-carbon face combination. Compressor was replaced with new in 2007 but have the same problem. When making pneumatic pressure test of the system at 10 bar in static condition, very minor leak appers from seal. when measured with soap water, leakage observed to be one bubble in 5 seconds. As per my experience with mechanical seals, this amont of leakage shold diappeared when running with oil if seal functions properly. So I concluded it to be a problem with mechanical seal. I am ot sure if I am going in wrong direction because I am new to field of refrigeration.

[RANGER1]

Have you anylized seal after removal ie. "O" ring deterioration / shrinkage or seal face damage ,overheating ?
Does the machine run smooth and no vibration ?
Depending on type of Howden you may consider ,



Coupling misalignment as well as coupling hubs not bored out true .

Soft foot on compressor or motor can cause stress on seal , also pipework .

Motor bearings OK ?

Silica carbide is very stable and does not deform so it and antimony carbon is a good combination .

[Grizzly]

Thanks Grizzly,
History says that mechanical seal develops leakage after approx. 15 -30 days of replacement. Rate of leakage is not in great extent that the system can't be run. Leakage rate used to be 30-40 drops of oil per minute against acceptable limit of 1-6 drops/minute. When contacted Howden, they are telling that new age compressors are equipped with silicon carbide-carbon face combination. Compressor was replaced with new in 2007 but have the same problem. When making pneumatic pressure test of the system at 10 bar in static condition, very minor leak appers from seal. when measured with soap water, leakage observed to be one bubble in 5 seconds. As per my experience with mechanical seals, this amont of leakage shold diappeared when running with oil if seal functions properly. So I concluded it to be a problem with mechanical seal. I am ot sure if I am going in wrong direction because I am new to field of refrigeration.

One of our guys on the forum, Plank has had some experience of S/Seal issues on Howdens.
I will post him and see if he has something to add.
What sort of temperatures is the plant room at?
Does your equipment normally run hot?
And it does not sound as if you are going in the wrong direction.
It definitely sounds like there is an issue?
Just not sure as to what is the cause but there are lots of good guys here, someone will crack it.
Grizzly

[Plank!]

You may have a problem with the "O-ring" material.

What type of oil do you use?

[desaipt2003]

Thanks Grizzly for referring to Plank,
Material of "O-ring" is Chloroprene. O-ring is found OK generally. Bothe the seal faces use to have rubbing marks. Oil used in the system is Mobil Glygoyle 220 - Polyalkylene Glycol (PAG)

[Grizzly]

Thanks Grizzly for referring to Plank,
Material of "O-ring" is Chloroprene. O-ring is found OK generally. Bothe the seal faces use to have rubbing marks. Oil used in the system is Mobil Glygoyle 220 - Polyalkylene Glycol (PAG)

I will bow to plank on the seal as to whether that is good or bad?
But rubbing marks on the seal faces is bad.
Basically what you should see is. 2 highly polished flat faces that are unblemished. Which have a film of clean oil on them. The film of oil creates the wet seal between the 2 faces.
Which is the basis of your gas tight seal between the system and the surrounding atmosphere.
This is commonly known as a viscous seal, so you can imagine anything that disturbs that seal will affect it's efficiency.
Also the condition and type of "O" ring is relevant as Plank mentions.
What is the condition of your oil, how often is it replaced, are there filters in the oil return line.
Is the oil being carbonised after being subjected to higher system temps. Caused by low refrigerant levels.
There could be a viscous circle of faults here?
Grizzly

[Grizzly]

Credit where it is due, Ranger has mentioned it and I forgot.
But are your drive couples ever checked for alignment.
Or ever done a vibration analysis as both of these can cause issues with the s/seal.
Fit a new S/seal and when refitted, check the alignment with a Lazaro.
Incidentally what refrigerant are you running on?
And again I assume yours is a dry atmosphere within the plant room?
Grizzly

[Peter_1]

Leakage rate used to be 30-40 drops of oil per minute against acceptable limit of 1-6 drops/minute.

IS this a seave or a compressor?
Even 1 drop/minute is for me ways too much

[sterl]

How low is your normal suction pressure? And how low does it get when the circuit is running light on charge?

Is this an external oil pump package? This a variable Vi machine?

How many seals have you renewed, and how fast? Are there evidences of other leaks?

If I read all this correctly, a new machine was installed in 2007 but it developed seal leak in short order, at 30-40 drops per minute.

In addition to motor alignment and vibration monitoring: I believe you need to have an analyis done on the refrigerant. Assuming nobody cross charged anything: you might find it very acidic and thus there is tramp metals in suspension. If the lubricant has already been sampled, what was its water content?

[Magoo]

Wear marks on the shaft seal would generally indicate alignment issues, chipping of carbon rotating face is a thrusting problem. A lazer alignment is necessary. I have slaved for days with DTI's to align couplings. Not worth it, get a Lazer alignment done.
OK R-134 A is dodgy on a good day and seems to disappear from any nook and cranny, open drive shaft seals on R-134A will be an on-going problem, suggest an electronic leak detector that will spoil you day, you will have leaks everywhere.
Magoo