fridge pressure release valve opened - why?

[danielb]

Hi all, hoping for some insight in to a weird problem.
I have a fridge/freezer of 80L driven by a York (CCI) 210 compressor on the engine. Had it recharged the other day after I rebuilt the compressor with a new shaft seal and valve plate gaskets.
Unit was charged with R406 and mineral oil, as well as fitting a new dryer and transfer valve. A pressure test and use of gas sniffer device (esp around my compressor) showed no leaks.
I ran the system for an hour and a half the day after regassing, but on my second use a week later the system ran for a couple of minutes before the pressure release valve (mechanical) opened and dumped out most of the seriously expensive gas....

The technician first suggested the condenser was overheating, and was the water cooling getting to it? I'd touched the condenser soon after operation and it was not overly warm. The engine driven water pump is pumping at full capacity, definitely no problem there.
I got a call back from the tech who then suggested it was air drawn in to system after last 1.5 hr run via shaft seal when pump sump went in to vacuum. I think he said this happens at shut down or when evaporator freezes??
Seemed convenient that the shaft seal I fitted (with greatest possible care and to the manual) is possibly to blame...
Are there other scenarios that can cause this outcome? Overgassing, under or over oiling of system?

[Richard Kollmann]

Engine driven water cooled refrigeration requires a high pressure switch set at around 200 PSI to open clutch on compressor at slightly higher than normal liquid pressure. To prevent explosive damage a relief valve or blow out plug is provided in high pressure liquid refrigerant side of system.

Extremely high pressure most likely was caused by:

Too much refrigerant added to system.

Blended refrigerant lost one of the blend gases that is used to reduced pressure of major gas R22 caused by the way it was serviced. Freon R22 alone causes much higher pressure.

Air a non condensable can be drawn into compressor by operating system at suction pressure to in a vacuum.

Interruption to good water cooling.


Comments:
Question why were you using R406a instead of a conversion ester oil and 134a?

Check relief valve for leaks as they do not always reseat properly and leak after a release.

Install a high pressure switch to prevent future problems as you were lucky this time that the Butane and mineral oil in this 406a blend did not start a fire.

[danielb]

Hi Richard,
Thanks for the reply - hoped you might be online.
My system only has a relief valve, but the clutch control sounds like a good idea.
Can overcharging the system cause a 250-300 PSI RV to open?

With regard to comment "Air a non condensable can be drawn into compressor by operating system at suction pressure to in a vacuum." how/when/where does this 'normally' happen?

What are the chances that a shaft seal that passes pressure testing then allows vacuum (at presumably lower PSI) to allow air in to system? Are you aware of Yorks doing this?

R406a and mineral oil were already used, and did not want to flush system for POE. Is there an advantage to going to 134a?

Thanks.

[danielb]

Hi Richard,
Another question;
It's correct that the York shaft seal requires the sump oil for lubrication, right?
To the best of my knowledge the technician who charged the system did not add any oil to the sump via the side plugs, only via the suction line. I had discussed this with him the previous day, saying I would reinstall the compressor without oil as I only had POE, not mineral oil. Seems likely he forgot or doesn't understand Yorks and as a result my seal got smoked on the first run, then sucked in air after being shut down.
Plausible?

[Richard Kollmann]

I have no idea of what kind of relief pressure valve is in your system. Normal high pressure on that kind of system would be with R12 refrigerant 100 to 130 psi. The newer type compressors when designed for car A/C use a non adjustable relief valve set at between 300 and 400 psi is my guess.

Unless air is mistakenly added during or before servicing I would not think you could get low pressure side low enough to reach a vacuum low enough to suck in air. Is there oil behind clutch on compressor? If you think seal is bad now pressurize system without running it and see if seal leaks. These shaft seals are tight enough that during deep dehydrating vacuum of one atmosphere 14.7 psi (31 inches vacuum) it will not let air in. And you are not going to see this low of vacuum from compressor.

If system worked well with 406a and mineral oil I am not suggesting the change to 134a and ester oil as you have enough trouble already.

I am not sure what seal is in your compressor a spring loaded carbon plate to stainless washer or a rubber V seal but they both need a film of refrigerant oil to achieve a tight seal. To guarantee a tight seal these automobile compressors need short term rotation every few weeks. YOUR QUESTION “my seal got smoked on the first run, then sucked in air after being shut down. Plausible?” NO I do not think so.
Again pressurize system and see if seal holds pressure for several hours without running compressor. The first question, is there oil in compressor? Depending on size of York compressor and if up right or on its side make a dip stick and check its oil level. I would not run system again without a high pressure electrical switch installed and set to 200 max with a 40 psi differantial.

[danielb]

Hi Richard,
My system was inspected by the technician today, and we confirmed that indeed the shaft seal (carbon, with spring) was now leaking (having passed pressure and vacuum testing originally). Also would not hold 30" vacuum - slumped to 24 in about 10-15 minutes. He pressure tested with nitrogen gas, and (only) on my insistence we checked the oil level in the compressor after degassing the system. Initially he said he would need to take it to the work shop to do this so he could drain it to measure what came out..... I pointed out that the service manual details an exact and easy method of measuring via dipstick, in situ.
The dipstick came back up with a tiny drop at the end, and was wet only a few mm up. The manual specifies 25-37mm in the sump (vertical install).
The tech wrote this off to oil migrating in to the system, and normal behaviour as it'd return.....wow.

When the system was charged with oil, it was done so via the suction line. The manufacturer CCI has since got back to me and said this is definitely not a method they condone, and could result in damage to compressor. A large local automotive/industrial outfit made the same comment.
I'm having them assess the compressor, but unsure if their findings can clarify cause of seal failure, though under lubrication is obviously looking likely in my eyes.
I can't understand where the oil went though. When the RV went off, only maybe a couple of table spoons worth came out.

Where would you suggest going from here?

[Richard Kollmann]

Not being familiar with your system’s design I will not be able to state the amount of oil this system needs. What I can do is give you some general idea of normal conditions from information you have posted.

The ER210 compressor is a 10 cu inch compressor and is sold for automobile A/C systems normally where it used with 14 (413 mil ) fluid ounces of oil. Because in the marine refrigeration operation there maybe unknown oil traps in holding plates, liquid receiver, filter/dryer and long lines additional compressor oil is required. For my engine drive designs I added 3 extra ounces on single holding plate systems.

Adding oil through low side service port or liquid refrigerant can damage the read valves but not the oil seal. Best way to add additional oil is to lay compressor on its side and use fill plug. After installing compressor or adding oil turn compressor center shaft over at least 10 revolutions by hand to insure there is no liquid in one of the cylinders.

It is very unlikely that the missing oil from compressor’s crankcase is in the system. And if oil is lost in boat 10 to 15 ounces would make a major mess.

Carbon spring loaded shaft seals need oil film on their surface to maintain a tight refrigerant seal. This is another reason compressor is turned by hand before engaging clutch. This warning comes with all carbon seals DO NO TOUCH CARBON OR STAINLESS WASHER WITH FINGERS and use refrigerant oil to lubricate seal on installation.

The disadvantage of carbon seals is they leak if not kept wet with oil.
The advantage of carbon seals is they can many times resurface themselves by running for compressor for a period of time.

Recommendation

I would lay compressor on its side without disconnecting lines. Open drain plug and remove all oil that drains out and refill with 14 ounces of oil.
Set compressor upright with belt on but not tight turn center shaft through with palm of hand 10 times. Add one pound of refrigerant and have belt only tight enough to turn compressor with clutch engaged.

Start engine and after it is warm lower RPM to idle and engage compressor clutch. If belt slips tighten it just enough to keep compressor turning. If compressor clutch is clicking off and on add just enough refrigerant to keep clutch engaged.
After system has run for at least 30 minutes turn it off. And let it set more than 12 hours. When you stopped system there would have been a degree of cooling in plates but system is still low on refrigerant this means the next day if there is any leakage it will show up in less cooling when running engine or if system is protected by a low pressure switch clutch will cycle on and off.


Final step after satisfied with no leaks is to:

• Add a high pressure switch set to less than 200 psi protecting system. There were two reasons why the belt tension was left loose 1. To prevent what caused the excessive pressure that opened relief valve before or liquid lock damage to compressor if a surge of liquid returns to compressor. In either case you want belt to slip without another expensive problem.
• Dehydrate system with a good refrigerant vacuum pump.
• Re-tension drive belt and add correct amount of refrigerant.