Bizarre FrigoBoat Behaviour !

[s/v Jedi]

Just want to start this thread so I can update it as events unfold :-)

We have these Frigoboat systems with Danfoss BD50 compressor, keel-cooler condensers and flat evaporators. They have been running without any maintenance worries for 6 years now and we got used to not having to deal with refrigeration problems. I have sold my gauge set, vacuum pump etc. years ago.

So we were a bit in shock when one box didn't really cool anymore. The compressor runs but the evaporator just has condensation, no ice. No sound of moving refrigerant either, so our first diagnosis was a blockage.

Thinking back, this must have been the moment where it all seemed reasonable but from where on it all becomes bizarre.

Frigoboat comes with resealable connectors which have O-rings. You get a whole bunch extra which indicates you're supposed to replace them now and then, although there's no mention on intervals etc. But, it's reasonable that 6 years is too long. I see some oil residue around them so they must leak a bit which means moist air can enter the system too so an ice blockage is likely.

However, Richard Kohlman, our CF refrigeration expert , has repeatedly warned about the keelcoolers so I had some worries, like dirt coming out of them and blocking the lines. Frigoboat does not put a filter in front of the evaporators to prevent dirt to get to the capillary tube.

So, we start monitor first to see what happens. We find that before the box temperature gets to 10 degrees Celsius, the circulation starts again and the box is pulled down to 5 deg C which is when the thermostat shuts the compressor down. This does not indicate blockage by ice at all because it would freeze again way before the box is down to 5 deg C, keeping the plate just above freezing all the time.

And the next cycle... it works normally again and some time later it starts all over.

We decide to test with the heat gun. We wait until it happens again and point the heatgun onto the evaporator. Refrigerant start flowing immediately. A minute later it stops again and every time we put the heatgun on it starts again: moisture freezing and blocking !!

But for whole periods it just works fine. Could it be blockage from both ice and something else?

After reading some info, I decide to tap the line where it reduces to the small diameter of the capillary tube. It does not start flow so we also use the heatgun and heat the whole plate thoroughly this time while I tap around where the cap tube starts. We also put the system to manual speed and the highest speed setting. My hope was to flush all that blocks out through and from the evaporator so that it has to pass the small filter/drier just before entering the compressor again. The compressor stops with 3-flashes LED meaning overload. This sounds okay as we're heating the evaporator. I switch the speed control to automatic again and the compressor tries to start every minute or so until all is cooled down enough to succeed. It quickly pulls the box down to 5 deg C and reduces speed to the middle setting. Next cycle it pulls it down very quickly and the controller chooses the lowest speed setting.

This is where it has been sitting for days now, lowest speed setting and quickly pull the box down every cycle without any trouble. It didn't work this efficiently for a long time !

to be continued...
Nick.

[s/v Jedi]

The system is still working perfectly now, with auto-speed selecting the lowest speed at all times.

I have come up with this explanation: oil contaminated with moisture has been accumulating in the evaporator until there was enough to freeze into a total blockage. Before that, it created a partial blockage which reduced the efficiency so that the compressor speed control selected a higher speed than normal. Our procedure with the heatgun and highest speed flushed the oil/moisture out of the evaporator back into the compressor.

At this time I don't think I also have solid matter causing blockages. But I am aware that there's too much moisture in the system to be handled by the tiny drier/filter. I can also expect oil to accumulate in the evaporator again, plus the O-rings still leak requiring maintenance.

I see 2 possibilities to remove the moisture: 1) evacuate while using the heatgun to help this like Kohlmann has described many times or 2) install an extra, bigger, filter/drier between the condenser and evaporator. I guess when option 1 is chosen it's best to also include option 2.

I think I will need to prepare for the quick fittings to be inoperable after this much time, loosing the charge when I disconnect a coupling. I am still considering to buy a new vacuum pump plus extra filter/driers.

It seems I'm the only one with this kind of problem :-)

cheers,
Nick.

[Richard Kollmann]

Logical approach to Bizarre Frigoboat’s behavior is to follow standard process of elimination using nondestructive testing means first.

Compressor runs at times with little cooling, no frost only condensation on evaporator. This then eliminates all electrical items, boat’s wiring, Danfoss’s electrical control module, thermostat and its wiring, and even electrical problems inside compressor.

If problem is not electrical then it must be a refrigerant flow problem.

There are times when evaporator cooling is good, assuming although not reported that frost covering of evaporator’s surface area exceeds 80%. This would eliminate major loss of refrigerant as cause of poor cooling.



If evaporator frost is present and cooling is good then later poor cooling with only condensate on plate, then interrupted refrigerant flow is cause of problem. Possible causes of temporary refrigerant flow are:

• Mechanical problems not likely on this BD compressor as electronic control module would stop compressor.
• Compressor reed valve failure again not likely as ounce a valve fails cooling is gone for ever.
• Moisture in system will produce the conditions reported if there is one or two more drops of water mixed in refrigerant that could not be absorbed by filter/dryer’s desiccant material. As liquid moisture inters capillary tube it freezes blocking flow. When flow is blocked temperature rises to above freezing allowing flow to beginning again. This on and off freezing cycle will produce an evaporator temperature just above freezing causing condensation to form on evaporator instead of frost. At some point moisture will slowly move through cap tube and freeze without blocking flow in suction accumulator section of evaporator allowing refrigerant flow until evaporator gets warm again. Then moisture can again move through condensing unit to block flow in cap tube again. If this system can produce frost on more than 90% of evaporators surface and is six years old with no one tampering with refrigerant moisture should not be causing blockage.
• It is unlikely that moisture can leak through O ring couplings and mix with refrigerant if refrigerant levels maintain full evaporator frosting. It would take a vacuum in suction side of system to draw moisture in.
• On older systems with Freon and mineral oil it was possible over time for wax or sludge to restrict or block cap tube but this is not going to allow flow off and on as reported.
• Possible liquid refrigerant and oil backed up in keel cooler do to cold seawater or refrigerant velocity is to low to empty cooler. Conventionally designed condensers allow gas to inter at top and with gravity’s help liquid refrigerant mixed with oil exits at the bottom. Warming up evaporator may increase liquid refrigerant velocity if problem is do to cold seawater
• Solids or sludge mechanical blockage of capillary tube devices is unusual on systems that are free of moisture and have filter dryer correctly installed. If Ester oil and 134a refrigerant react the same as Freon and mineral oil moisture added to the mix acid sludge will form that can block cap tube..
• Large solid piece or pieces of material small enough to pass through unfiltered tubing but too large to pass through cap tube’s small orifice will restrict flow. This type blockage condition will and has caused temporary tube blockage on Frigoboat Keel cooler systems, the temporary solution is to stop unit and tap on liquid line till item falls off cap tube inlet and slides back in liquid line. Permanent fix is to replace complete evaporator assembly. Evaporator replacement should include installation of new larger filter dryer in liquid line where it should have been in the first place to prevent future cap tube blockages.

As far as the LED troubleshooting light indicating a three flash amperage overload signal I can not explain except if compressor stopped and quickly tried to restart while under a load.

Nick, hopefully the problem with your Frigoboat keel cooler model was an isolated case on capillary tube blockage and now blocking item has moved downstream where it will be captured by the incorrectly installed filter.
If problem does reoccur do not over react and start replacing compressors or evaporators as others have done on this system. If it turns out there is a loss of refrigerant do not forget to check keel cooler inside boat for leaks.

[CarlF]

I once had a frigoboat problem that is somewhat similar to what you describe - but it seemed to be set off (rarely) by greater than usual heeling. The folks at US Frigoboat suggested that contaminates sometimes intermittently clog the tiny soldered junction of the capillary tube. They suggested bending the tubing so as to change the angle of this solder joint - as I remember to make it more vertical. (this was bending the main tube that supports the capillary not the capillary tube itself). They said that if this failed the refrigerant should be purged and refilled to get rid of the contaminant.

I tried this and the problem didn't return - of course, it might just be karma

Carl

[s/v Jedi]

Ah there's Richard to the rescue

The reason I am 99% sure it's moisture is that 1 second of heat gun to the evaporator causes flow again, then frost followed by blockage. We can repeat this again and again...

There is oil residue around the quick connectors so I know they are leaking. The loss is minute because 100% of the evaporator is frozen in between blockage but some frost on the suction line has gone over time. This seems in line with the small leaks around the O-rings.

I will remove the floor board above the keel coolers to check for leaks there too. Will use the soapy leak detector.

The water here is Panamanian tropical hot. The 3-flashes on the LED was caused by heating the evaporator with the heatgun... which would lead to overpressure (overload) I think, much like starting the system with a very hot plate/box.

On the Italian website of Frigoboat I see manuals that do show a decent filter/drier in the right place. Got too expensive probably.

About the small filter/drier that sits in the suction line just beore entering the compressor: it's a small copper thing... I wonder if heating it while evacuating the system would regenerate the desiccant in it? If it comes to evacuating, I will try, although it's hard or impossible to know if it worked.

The keel coolers are the low point of course so it might well be that oil got accumulated in there. The box with the trouble has that big Frigoboat box evaporator and I think the suction line is also connected on top of it. The other 2 boxes have a flat plate formed to line 3 walls and both connections are on top too, so all these have a low point in there.

CarlF: yes it makes sense to have that part of the line where the capillary tube connects vertical. But mine isn't and the copper has become hard enough to fear cracks when bending again.

thanks guys!
Nick.

[s/v Jedi]

Finally I received the gauge set and vacuum pump. I had already bought some R-134a locally so I was ready to do the job.

I started with disconnecting all three quick-connect fittings and replacing the rubber O-rings there. They were indeed hardened and malformed/squared. I used the spare O-rings that came with the compressor units. The quick-connect fittings worked as advertised. This installation is 6 years old and I was a bit worried that a fitting would malfunction, causing the charge to escape.

When all this was put together again, I evacuated the system with the vacuum pump for 2.5 hours while repeatedly using a heat-gun to heat up the whole system/circuit. I put extra heat on the little filter/drier (hoping to get the moisture out/regenerate it). After 2.5 hours I stopped the pump, moved the yellow hose to a R134a tank and let it equalize pressure. I then closed the red valve on the gauge manifold, closed the valve to the R134a tank and started the compressor by connecting the thermostat again. About 25% of the evaporator got frost and I kept adding R134a on the suction side bit by bit until the whole evaporator got frost. As the system got colder I needed to add more refrigerant now and then. I got to see a little condensation on the return line but no frost.

After a couple of cycles (it's a fridge that switches compressor on at 7 deg. C and off again at 5 deg C.) all looked good so I removed gauge set and I'm now looking at a system that works as good as it did when new 6 years ago!! The automatic speed controller has already lowered compressor speed to lowest setting and it runs about 1/3rd of the time (this is a 7 cu.ft. box in the Panamanian tropical heat!)

I paid $79 for the gauge set with hoses that have little valves at the ends which is nice to keep it free from contaminants while working with them. The vacuum pump cost $89. I bought both new on Amazon.com. I used almost 1 small can/tank of R134a for the complete system: compressor, keel cooler and the big box-evaporator (I have no pipe-extension(s) between evaporator and compressor). These cans cost about $10 each here in Panama. I have two more systems so I'll go ahead and replace O-rings for those too and check their gauge readings/top them up a bit.

What I like about this is that I knew absolutely nothing about this when we started cruising in 2003... but the old Glacier Bay system forced me to study it quickly... which lead me to replace it with this Frigoboat setup... which worked without any maintenance for 6 years... and now that it needed attention, I still know how to deal with it and fixed it quick and without trouble. I also received a new (spare) flat evaporator and a bigger filter/dryer so I can even replace an evaporator if needed.

In case that anybody who reads this is about to install a new system: specify that you want a good sized filter/drier installed right in front of the capillary tube of the evaporator. You don't get these with the Frigoboat systems. They sell a version with quick-connect fittings (pre-charged with R134a etc.) for $169.- or something which is so expensive. I would have it soldered in the return-line of the keel-cooler/condenser, just in front of it's quick-connect fitting, for a mere fraction of that price.

I also recommend to change the rubber O-rings of the quick-connect fittings every 5 years. Do this when you have vacuum pump, gauge-set and R134a near-by. Best to do this with a check-up of the system.

ciao!
Nick.

[Mirar]

Thanks for taking the time to post this info Nick. I am about to install two of these same units in my new alloy cruiser I am building. I plan on putting the keelcoolers into tanks welded to the inside skin and filled with de-ionized water to hopefully stave off any electroysis between the bronze keel cooler and the aluminum hull. I will take the keel coolers into the local refrigeration shop and get a better filter/drier installed.
cheers, greg

[s/v Jedi]

Originally Posted by Mirar

I plan on putting the keelcoolers into tanks welded to the inside skin and filled with de-ionized water to hopefully stave off any electroysis between the bronze keel cooler and the aluminum hull.

I recommend you use engine coolant and also isolate the keel-cooler from the tank.

ciao!
Nick.

[seashine]

As I'm about to start building our fridge I'd just like to say thanks to the Jedi and of course also to Cap'n Dick the famous refrigerateer :-)

[hpeer]

Well, speaking of FrigoBoat weird problems.....

Ours is about four years old. Lightly used. Installed by PO.

The main issue is that if I charge the system it seems to keep a charge (for a couple of days) until I turn it off. Then it looses the charge over a day or so.

I have recharged the sysetm a couple of times all with the same result.

I changed the O rings and the Scharder vavle with no change in operation.

[ShaunJ]

Sounds like you have a leak, you can buy freon with dye in it. Next time you charge it, use the stuff with dye in it and get a UV light and yellow glasses. Run it for a while and then check the entire system. Shine the UV light on every inch of the system and look thru the yellow glasses, it will be very evident when you find the leak.

[s/v Jedi]

If you loose the charge that quick, you can use soap bubble technique. I have read too many warnings about not using dye in these systems...

p.s. my system is still working the same so the repair was 100%

ciao!
Nick.

[ShaunJ]

Jedi has a good point, soap will work also. Most manufactures do not recommend using dye, however sometimes it is needed.

[Richard Kollmann]

Hpeer, The condition you describe will result when there is a high pressure leak in suction side of system. When unit is keeping evaporator cold even when turned off the low pressure side of system rarely has more than 20 pounds of pressure in it. When running pressure in low side is less than 8 psi.When system is off for several hours suction pressure in compressor and evaporator will rise to match pressure ratio of there increased temperatures.

Leak testing with a one inch paint brush and a mixture of 50/50% liquid soap and water is best way to find leak. With compressor off add refrigerant (gas only) no liquid till pressure in system is equal to bottle pressure. Using paint brush agitate soap solution on suspected leak area to break down surface tension and form a shaving cream texture. This method will find leaks as small as 1/2 ounce per year.

Suspected leak areas on low pressure side of a Frigoboat unit first would be poor sealing cap on compressor refrigerant service fitting. This cap's seal is important as all valve cores leak.

Second place to check for leak is Frigoboat's line connector on evaporator return line to compressor.

If your unit has an aluminum evaporator there will be a joint in aluminum return line coming out of evaporator where line changes to copper check this joint for a leak. All of these checks must be done with compressor off.

Final check is the most time consuming testing complete surface area of evaporator one square inch at a time.

[hpeer]

Thanks Richard. I will give that a go.