Refrigerator help

[Captain Bill]

I have a 8 year old Nova Kool RFS7501 that started having cooling problems. The symptoms were, the refrigerator compartment would only cool to 61f, the freezer compartment was 24f. The evaporator plate in the refrigerator compartment was covered in condensation but not frosted while the main feed line into it was covered with frost. The compressor was running pretty hot. I talked to Nova Kool support and they concurred that the unit was probably low on R134a and should be checked for leaks and serviced by a professional.



Trying to find a refrigerator repair person to work on a boat fridge on the boat was pretty hard but I finally found one. They came out, and put in some R134a and declared all was good and if that didn't solve the problem I would need a new compressor, charged me $293 for a shot of refrigerant and left.



The unit has been running now for 22 hours in free air and empty so it's not trying to cool anything off but the refrigerator itself. That is it has not been put back in its enclosure so there are no restrictions on cooling air. While they don't appear to have solved the problem, what they did managed to change the problem. The evaporator in the freezer has stabilized at 29f measured by a digital thermometer in contact with the evaporator. The Refrigerator compartment has stabilized at 49f. These are remote thermometers so I am not opening the door to get readings. The evaporator in the refrigerator compartment is now completely frosted. The refrigerant return line at the compressor is covered in ice. The compressor itself is slightly cool to the touch instead of slightly warm as it was when the system was new. The amp draw seems to be up about an amp as well.



After watching several videos on problems with refrigerator I believe that the unit is showing symptoms of being overcharged. I hesitate to call these guys back as I suspect they are likely to tell me I need a new compressor. Given what they charged me for 40 minutes worth of work and a bit of R134a I suspect a compressor replacement would probably exceed the price of a new refrigerator.



While I have a basic understanding of the principles of refrigeration it is the one system on the boat that I have basically no experience with regarding service or repair. My question to the refrigeration experts out there is, am I on the right track with my overcharge diagnosis, or am I really going to need a new compressor?



Thanks for any advice you can provide.

[Sailmonkey]

It certainly sounds overcharged, bear in mind though the extremely small charge in these things is fiddly at best. Venting a “pffft” of gas could be the fix, or there could be something else going on.

[Richard Kollmann]

Poor cooling before tech could have been poor condenser cooling air flow bad fan or clogged fins.

Frost on line returning to compressor now indicates too much refrigerant. I suggest the following turn unit off and open box to allow system to return to ambient temperature before restarting. If units amp draw exceeds 6.5 amps running there is too much refrigerant. It control module prevents compressor from starting now there is way too much refrigerant.

If you need to remove refrigerant let out only two seconds at a time with unit turned off. Then when restarted and running after 10 minutes amperage should be below 6.5 amps. Final adjustment hours later line returning to compressor will only have a few inches of frost outside refrigerated box and evaporator will have frost covering 85% of evaporator surface area.

[OzeLouie]

Hi Captain, as stated above the tech has altered your problem from no refrigerant gas to grossly overcharged and to suggest that the compressor may need changing simply confirms the poor work / advice. Your compressor is OK. Being on the job for 40 minutes it is most unlikely that the proper remedy has been performed and simply adding gas will most likely make things worse in the long term.. Like kicking the can down the road!!
The ONLY reason a fridge system requires gas and that is because it has a GAS LEAK! and the only long term remedy is the standard refrigeration industry practice and as this article describes:

Refrigerant gas charge:

A common cause of gas leak is from couplings but more often from corrosion (pitting) of aluminium evaporator tubing or damage to the soft aluminium evaporator (cold plate) Unfortunately this failure often leads to entire system contamination requiring replacement.

If a system is running but the evaporator or eutectic plate is not or only partly frosting, it is most likely that the system has lost refrigerant gas. Usually loss of gas causes low or high power consumption rate. (High if air contaminated, evacuation and new filter essential as a minimum)

If a system needs gas it can only be because it has a gas leak, a leak that must be identified and repaired. (If the gas leak was in a wet area like the evaporator, contact for further instruction.) After repair the filter dryer should be replaced and the system evacuated preferably for several hours before adding fresh refrigerant gas.

For proper evacuation, if possible connect both your gauge suction line (blue) and high side (red) to evacuate system. Once evacuation is complete, add refrigerant until there is an above zero PSIG gauge reading, close off red at gauge manifold, remove red high side hose charge from unit and continue to slowly top up system via suction (Blue) now with the unit running. Only use pure refrigerant gas, usually R134a. When recharging, run the unit and only slowly allow refrigerant in sufficient to maintain a 12 - 14 PSIG suction. . Once this pressure is maintained without addition, close off and leave running. Fine tune later once temperature lowers and settles.
If overcharged some purging is required. Best to purge in 3 to 4 second increments 15 minutes apart with the system running. Observe the suction pipe and you will notice that eventually the sweating / frosting will subside and you will have a cool but dry suction. That is the objective.

Do NOT use cans of R134a that contain dyes or sealants like those used on auto air conditioners.

Filter Dryer:

Unfortunately most small refrigeration systems are fitted with cheap 'spun copper dryers'. These items usually have no felt filter pad unlike the proper filter dryer type 032.

Copper spun dryers only have a screen. The screen doesn't collect the fine gunk that rubs off the desiccant beads or sludge that can form as a result of contaminated oil, eventually to possibly cause capillary or TX valve restriction.

If repairing a leaking system we suggest fitting a proper type 032 or 052 flared filter dryer mounted vertically, exit downwards. If this can't be fitted in place of the old copper type, then simply leave the copper one in place and fit the proper filter dryer as well.


Conclusion:
Sadly Captain your system has issues. It most likely still has a gas leak, could be or be becoming contaminated. The only long term remedy is to have it repaired properly by a competent person or replace it. I would suggest you turn it off and let it warm then check for leaks using soapy water (dish wash stuff is good) specially check the evaporator and its aluminium pipe to copper join area. You will need good light, and eyesight as the leak may only indicate as pin prick sized bubbles. If the evaporator and its pipe run have a leak (as I suspect) it would be worth replacing the evaporator and its lines and have the system re-gassed as described here by a competent tech.

Note: if replacing the evap, check that the area where the aluminium and copper pipes join is covered and sealed to prevent future corrosion.

Cheers Louie

[Richard Kollmann]

OzeLouie's recommended industry standard is not applicable in Captain Bill's situation in a remote area of the world with limited expertise and lack of manufacturing equipment. Another point is it would be a mistake to service a Nova Cool RLS 7501 and Danfoss BD50 compressor to a low side pressure as high as 12 to 14 psi. Normal stabilized low pressure on this unit is 3 to 8 psi depending on ambient temperatures.

The confusing part of original post is compressor reported as hot, if to hot to keep hand on would not point to low on refrigerant but knowing amp draw of system would confirm loss of refrigerant. Now after adjusting frost level or load amperage of system by removing a small amount of refrigerant the leak rate and be checked my monitoring frost weekly. Before pushing unit back in enclosure check servicing cap for leaks all other connections are metal to metal sealed.

[Captain Bill]

We just completed Ozelouie's suggestion of a soapy water search for leaks twice, the second time using a magnifying glass. While I agree that there must be a leak, it must be really slow. Nova Kool support called me hast evening and suggested I bleed off a bit of refrigerant. Which I did in three stages. Unfortunately I now seem to have overdone it. It seemed stable last evening but this morning it was showing signs of low refrigerant again. I'm going to buy a set of gauges and see if I can manage to do this properly instead of just guessing. I used the soapy water on the valve to make sure it wasn't leaking and got no bubbles. I am assuming that I didn't wait long enough to make sure it was stable last night.



Just an FYI the condenser fins were surprising clean in my view. There were a few minor areas with some spiderweb looking material and a really small amount of cat hair from our cat. Frankly I expected it to be much worse. It is really difficult to access the back of the enclosure. I have about 90 square inches of vent at the bottom rear of the enclosure and 120 inches at the top front, I'm way above spec on ventilation.



I am worried that the fan is getting a bit old. We lived on the boat for 6 years full time mostly in the tropics where the thing almost never cycled off. For two years after that it has been running approximately 1/3 of the time. These muffin fans are not all that long lived and are usually rated between 30,000 and 70,000 hours. I figure this fan has about 60,000 hours on it, so I'm planning to replace it.



Last night when I thought we were making progress I had the following temperature readings. The evaporator in the freezer was -1f and the refrigerator evaporator at 6F according to the IR thermometer. The refrigerant return line stopped condensing and the case of the compressor is at 104f instead of cool to the touch. I think these numbers sound close to what one would expect.



One other thing, this unit has a bd35, not a BD50.

[OzeLouie]

Originally Posted by Captain Bill

We just completed Ozelouie's suggestion of a soapy water search for leaks twice, the second time using a magnifying glass. While I agree that there must be a leak, it must be really slow. Nova Kool support called me hast evening and suggested I bleed off a bit of refrigerant. Which I did in three stages. Unfortunately I now seem to have overdone it. It seemed stable last evening but this morning it was showing signs of low refrigerant again. I'm going to buy a set of gauges and see if I can manage to do this properly instead of just guessing. I used the soapy water on the valve to make sure it wasn't leaking and got no bubbles. I am assuming that I didn't wait long enough to make sure it was stable last night.



Just an FYI the condenser fins were surprising clean in my view. There were a few minor areas with some spiderweb looking material and a really small amount of cat hair from our cat. Frankly I expected it to be much worse. It is really difficult to access the back of the enclosure. I have about 90 square inches of vent at the bottom rear of the enclosure and 120 inches at the top front, I'm way above spec on ventilation.



I am worried that the fan is getting a bit old. We lived on the boat for 6 years full time mostly in the tropics where the thing almost never cycled off. For two years after that it has been running approximately 1/3 of the time. These muffin fans are not all that long lived and are usually rated between 30,000 and 70,000 hours. I figure this fan has about 60,000 hours on it, so I'm planning to replace it.



Last night when I thought we were making progress I had the following temperature readings. [SIZE=2]The evaporator in the freezer was -1f and the refrigerator evaporator at 6F according to the IR thermometer. The refrigerant return line stopped condensing and the case of the compressor is at 104f instead of cool to the touch. I think these numbers sound close to what one would expect.

[SIZE=2]One other thing, this unit has a bd35, not a BD50.

Good Bill you are on the right track but still have a gas leak to find!!!

Suggest checking the aluminium evaporator tubing (which may be hard to get at) as that is a leak prone area (pitting... see sketch below)
If you can locate the leak you can then determine what is best to proceed to a permanent fix. If the evap has pin holes, you could purchase a replacement one and have a competent tech fit if for you following the correct repair / re-gas procedure I posted earlier and you should have a permanent long term fix.

Just one thing: If the evap has to be changed, make sure the replacement has the aluminium to copper join area is sealed air tight to prevent future pitting.

[Richard Kollmann]

Captain Bill you know now you took the wrong track when you did not follow my instructions of a two second release of refrigerant. This system only contains from 50 to 80 grams by weight of R134a pure refrigerant. The correct volume can only be provided the manufacturer of that system. Plus or minus 25 grams has a major effect on performance. The technician had to remove all refrigerant remaining in system before adding the correct volume or topping off existing volume by monitoring current draw and low side refrigerant pressure. The problem with low pressure fine turning is low pressure is not stable except in a time window 10 to 20 minutes after a warm system is started.

You had an opportunity after technician over charged unit to find or confirm an actual volume of refrigerant and the leak rate in days, weeks or months. Now that you shot yourself in the foot you need to fill system with refrigerant till evaporator develops a frosted surface or amp draw of unit is stabilized 5.5 amps now that we know compressor is a BD35.

I find that the servicing cap seal is the first place to look for leaks on this type unit and leaks that take up to a year to show up requires a great deal of time to find. Because this is not an ice box conversion unit if there is really a leak in evaporator or in its copper to aluminum tube connection it is not repairable or can evaporator be replaced.

[Captain Bill]

Originally Posted by Richard Kollmann

Captain Bill you know now you took the wrong track when you did not follow my instructions of a two second release of refrigerant. This system only contains from 50 to 80 grams by weight of R134a pure refrigerant. The correct volume can only be provided the manufacturer of that system. Plus or minus 25 grams has a major effect on performance. The technician had to remove all refrigerant remaining in system before adding the correct volume or topping off existing volume by monitoring current draw and low side refrigerant pressure. The problem with low pressure fine turning is low pressure is not stable except in a time window 10 to 20 minutes after a warm system is started.

You had an opportunity after technician over charged unit to find or confirm an actual volume of refrigerant and the leak rate in days, weeks or months. Now that you shot yourself in the foot you need to fill system with refrigerant till evaporator develops a frosted surface or amp draw of unit is stabilized 5.5 amps now that we know compressor is a BD35.

I find that the servicing cap seal is the first place to look for leaks on this type unit and leaks that take up to a year to show up requires a great deal of time to find. Because this is not an ice box conversion unit if there is really a leak in evaporator or in its copper to aluminum tube connection it is not repairable or can evaporator be replaced.

It would have been nice to wait weeks or months but I want to use the thing and as it sat with a 50 degree fridge and a barely freezing freezer it was pretty well useless. I did about 5 seconds the first time and the second an third was about two seconds. It seemed to me I had hit the sweet spot before I went to bed, but alas in the morning it was clear that I had let too much out. Either that or I didn't get the valve shut tight enough and it bled refrigerant slowly over night. All of the tubes on this unit are soldered so the "pro" had to add a service port. It was the clamp on type which punctured the copper service tube. The valve itself is opened and closed with an Allen wrench. The cap does not appear to be anything more than a dust cover. There is no gasket as far as I can see and the threads on the cap seem too roughly cut to seal.



My vacuum pump and gauge set should arrive Weds. and I'll pump it out and add the proper amount of refrigerant. The sticker on the door says 5 oz's (140 grams) and I have confirmed that with Nova Kool. I also have acquired a clamp on digital ammeter that does DC so I'll be able to accurately monitor the current draw. My old one only did AC.


I suspect the leak is in the condenser tubing. The lower 4 tubes are quite rusted as they are nearest to the air intake for the enclosure (isn't salt air a wonderful thing). Though I could not see the leak with the soapy water test it is the only area that looks compromised at all. I may be able to remove the lower mounting screws and immerse it in a tray of water if I can find one deep enough.



I also tested my fan today and it's only drawing 150ma. Not too bad for as many hours as it has on it.

[Captain Bill]

I got my vacuum pump and gauge set last night and went out to test my fridge this morning. I put on the gauge and the pressure inside without anything running was Zero gauge. Just to make sure it wasn't a faulty gauge I remove the line and opened the valve and nothing came out of the port. Clearly the gauge was reporting correctly and all of the refrigerant had leaked out since I last worked on it even though I had tightened the valve as tight as I dared. My thought was, ok, my leak is now much bigger than it was. I then hooked up the vacuum pump and pumped it down to -30 gauge. I know it's probably not exactly -30 but the scale between 0 and -30 is only about .25 inches so not a lot of resolution. It took less than a minute to get down to -30. I closed the valve on the gauge but left the gauge attached and the valve on the service port open. The pressure has been sitting at -30 for an hour and a half now and as far as I can tell the needle has not moved at all. At this point I am guessing that the valve in the tap is bad and the refrigerant has slowly leaked out of the service port. I'm going to leave the vacuum on it for a couple of more hours to make sure it doesn't leak somewhere else. I did not see the tech clean the service tube before putting on the tap so I guess it's possible there's a small piece of dirt or something that is preventing the valve from sealing. Since it doesn't seem to be leaking with the valve open I am guessing it is not sealing around the puncture, not the valve stem. If anyone has some good advice on replacing the tap I'm open to suggestions. My plan at the moment is to "close" the valve to the point that the needle is sticking out enough that I can line it up with the hole and then put the bolts in. I will scotch bright the tube before I do that. If I'm on the wrong track with this please tell me.

[Richard Kollmann]

It sounds like now if there was a leak before and now the self piercing valve is also leaking it would be a mistake to connect vacuum pump adding to the contamination.

The valve tech installed usually will have a one time use seal.

I would recommend not running compressor until both leaks are corrected.
Connect refrigerant bottle and purge air from gauge set add refrigerant pressure with compressor off and bottle right side up for gas vapor only till pressure in system is equal to bottle pressure then close off refrigerant flow.

Leak testing is to be done with a 50/50% mix of liquid dish washing soap and water. Apply mixture with a one inch paint brush agitating liquid on suspected area till it resembles shaving cream foam in order to break down surface tension between leak and metal. For slow leaks like one ounce per several weeks watch area for 5 minutes. Leak at the new servicing valve and blue hose connections need to also be checked.

I am still wondering why this unit needed the new servicing valve sense for 24 years all these systems required a servicing valve to reclaim refrigerant.

Good luck you will need it.

[Captain Bill]

Originally Posted by Richard Kollmann

It sounds like now if there was a leak before and now the self piercing valve is also leaking it would be a mistake to connect vacuum pump adding to the contamination.

The valve tech installed usually will have a one time use seal.

I would recommend not running compressor until both leaks are corrected.
Connect refrigerant bottle and purge air from gauge set add refrigerant pressure with compressor off and bottle right side up for gas vapor only till pressure in system is equal to bottle pressure then close off refrigerant flow.

Leak testing is to be done with a 50/50% mix of liquid dish washing soap and water. Apply mixture with a one inch paint brush agitating liquid on suspected area till it resembles shaving cream foam in order to break down surface tension between leak and metal. For slow leaks like one ounce per several weeks watch area for 5 minutes. Leak at the new servicing valve and blue hose connections need to also be checked.

I am still wondering why this unit needed the new servicing valve sense for 24 years all these systems required a servicing valve to reclaim refrigerant.

Good luck you will need it.

This is a Nova Kool which is made in Canada, but there clearly no servicing valves in the installation. I have not turned on the compressor. When I saw that there was no pressure in the system I tried pulling a vacuum to see If I could find the leak. It has been five hours and the gauge still hasn't moved. I've got plenty of R134a so I can pressurize the system as you suggest. The tap that they put on does not have a schrader valve, just an open port. The "pro" let refrigerant out of the system a couple of times by unscrewing the needle with an allen wrench. This allen screw appears to be the only way to open and close the valve. I have been looking on line and the one on my unit seems to look like a cheap style which is available on Amazon for less than 5 dollars. The ones with schrader valves are more in the $30 range. Do I understand correctly that these cheap valves can only be opened and closed once or that you cannot reuse the valve once it has been removed?

[pyrrano]

The piercing valve is for temporary use only to check pressures, you need to braze in a service port (suction/return side) in order to evacuate the system. Proper vacuum is 200-500 microns but since you don't have a micron gauge a I'd pump it for an hour make sure there's moisture in the system.
If you decide to keep the piercing valve and it vacuum holds at 30" charge the system by pressure not by weight. Charge 134a vapor (can tap on top) into suction (return) line through your gauge till it stops rising, shut gauge off, turn compressor on, most likely pressure will come down to 0 or even into vacuum, keep adding vapor in small bursts till you get 10-12 psi for refrigerator 0 psi for freezer.
Suction /return line should be cold to the touch and make sure it's insulated. Check the pressure again when it's close to set point, suction pressure will get lower as it's cooling down. Keep checking amp draw, it should not get higher than the factory recommendation.
Make sure the condenser is clean and has proper airflow.
Check all 3 pipes after a few mins compressor run. Compressor to condenser should be hot, condenser to cold plate should be warm, cold plate to compressor should be cold.

[Captain Bill]

Originally Posted by pyrrano

The piercing valve is for temporary use only to check pressures, you need to braze in a service port (suction/return side) in order to evacuate the system. Proper vacuum is 200-500 microns but since you don't have a micron gauge a I'd pump it for an hour make sure there's moisture in the system.
If you decide to keep the piercing valve and it vacuum holds at 30" charge the system by pressure not by weight. Charge 134a vapor (can tap on top) into suction (return) line through your gauge till it stops rising, shut gauge off, turn compressor on, most likely pressure will come down to 0 or even into vacuum, keep adding vapor in small bursts till you get 10-12 psi for refrigerator 0 psi for freezer.
Suction /return line should be cold to the touch and make sure it's insulated. Check the pressure again when it's close to set point, suction pressure will get lower as it's cooling down. Keep checking amp draw, it should not get higher than the factory recommendation.
Make sure the condenser is clean and has proper airflow.
Check all 3 pipes after a few mins compressor run. Compressor to condenser should be hot, condenser to cold plate should be warm, cold plate to compressor should be cold.

Thanks, I think I've made a little progress on the diagnosis after following Richard's advice to add a bit of R134 to the system. While I haven't actually found the leak yet, I have identified the probable area. After letting it sit two hours, not running with about 40 PSI in it, I stuck the halogen sniffer in the refrigerator side and got no reaction. When I put it in the freezer side It screamed at me for about 5 seconds and returned to normal, which is a short beep every second or two. I'm assuming that it leaked enough into the freezer side that it set off the detector until the concentration got below detectable levels because the door was slightly open. It's getting dark here now so I will search for the leak in the morning when I have some decent light. The sniffer arrived from Amazon today which is why I had not used it before. I had to take the doors off to get to the mounting screws and the Pros didn't put them back on before sniffing for leaks. I guess the leak is so small that without the doors closed it never achieved detectable concentrations. They did put the doors back on to charge and run the unit, but did not sniff for leaks after doing so.

[pyrrano]

Yeah, there is usually a leak when the system is low on charge. Some leaks can be fixed with 2 component epoxy putty but brazing is the best way. It's it super small, topping off the system every few months might save you a lot of $.