Refrigerator Project

This was the project: improve the old refrigerator to be more electrically efficient; have food be more accessible and I wanted ice. Motivation for the work came from a one year long cruise in 2005 from Portland to Canada then on to six months in Mexico’s Baja Peninsula. During that trip the refrigerator was the single largest user of electricity. In Mexico the refrigerator was on/running about 20 out of every 24 hour day. The compressor used about 6A while on so used 120Ah of charge. With one 80W solar panel and an 80A alternator attached to a 600AH battery bank it took many hours to replace the charge , and it didn’t make ice. It was mostly the Mexico portion that prompted the work, since the climate there taxed the system hardest. The compressor and heat exchanger (Cold Machine) were mounted in the lazerette, a closed-in space that became very warm in the late afternoon.  Since it is the temperature differential between the refrigerant and (in this case) the ambient air temperature in the lazerette the system was not terribly efficient.

l. Considering that a battery bank should not be consistently  emptied to less than 1/2 it’s capacity, i.e. 600/2=300A-h; 120A-h is about 1/2 of usable stored charge. Most  80A alternators rarely if ever puts out 80A so we needed to run the diesel literally hours each day to keep up with the energy demand.  My simplistic look at the economies of the proposed project compared the cost of using the engine (diesel, maintenance, new engine amortized over its life) and found some sort of new system to be a good investment.
My goal was to cut the refrigerator consumption in half and still have the freezer colder than the old system. At the same time I wanted to add more PV panels but that is another discussion. I wanted to improved the refrigerator three ways:

1. Design/purchase a mechanical system that was more efficient.
2. improve the insulation and
3. lastly reduce the box size to both have less space to cool and make the space more accessible.

The old refrigerator space went so far down towards the floor that I had to hold Judy by her feet while she went for the bottom of the refrigerator or we had to use a step stool. Even with food organized in a basket arrangement it was still unsatisfactory. So, in the new box, the space needed to get smaller (and less deep) and a by-product of that decision is more space is available for insulation.

Our make vs. buy decision process went something like this:

• Is it worth it to replace our refrigerator? . Yes!
• Do we want to use an independent box – just purchase a complete unit, box, insulation, and compressor or is it something we needed/want to make? I elected to make the system but this option deserves a good close look.
• Rebuild our existing system.  Insulation was old and not effective and the compressor was tired. No!
• Create a new the refrigerator system, add new counter tops. This is what we ended up doing.

I elected to put the keel cooler in first because we could combined that work with new bottom paint. Location of the keel cooler is critical since you cannot extend the refrigerant lines between the keel cooler and the compressor. So the keel cooler must be within five feet of the compressor. Once installed and bedded with 5200 I bonded the keel cooler to the boat’s electrical bonding and also ran copper foil from the cooler to the ships SSB system. Our keel cooler also has an attached zinc anode for further protection.

Next,  I attacked  the old insulation. I ended up destroying the old counter top along with the refrigerator and freezer doors. I think it would have been possible to do this non-destructively to get access to the old the insulation, but expecting to replace counter tops I didn’t put a lot of effort into saving the old tops.  I eventually removed an entire wood side to get better access. I ended up cracking some of the teak slats while removing the side but it couldn’t be helped.

1st piece of new insulation

The old insulation was an expanding foam type insulation, installed by the boat manufaturer 25 years ago, probably  a polyuerthane. The old insulation  sticks  like glue and is very tough to get out. While some of it came out in big chunks  in other places ( especially difficult to get at places) it dribbled out in small pieces. The process was long and tedous. I used a combination of tools in the removal process. A Fien tool, a few different hand saws, chisels; a good vacuum is absolutely necessary, as is a good mask.

I used three criteria for picking the new insulation:

1. Good insulating properties
2. Low moisture absorption
3. Easy to get locally.

I ended up using Dow rigid insulation because mainly because  it absorbs very little moisture over time. But also has an R value of 5 per inch. It’s an extruded polystyrene and was available at my local lumber and home improvement stores.  I ended up using both the 2 and 1-1/2 inch thicknesses to create at least 5-6 inches in all directions. Its especially important to increase thickness on the bottom and all around the freezer. More than 6-7 inches is diminishing returns.  The old refrigerator space is defined by a stainless steel shell and fits along the (compound) curved hull. If I had it to do over again, I’d fill the old space with insulation and make the whole thing rectangular. But I used the old stainless box as a starting point. I filled the drain hole, thinking that it was a place that heat would enter and thinking I would not use the drain. I think I would re-think that decision now. We did buy a wet/dry vacuum that we now use to pick out water when the freezer needs to be de-frosted.

When this picture ( above) was taken , the counter top had been removed, and the side  panel had been taken out to get access to the space. A bunch of (heavy) junk was pressing the glue on one of the 1st pieces of new insulation.

The new design defined two spaces, freezer and refrigerator separated by a piece of insulation. The evaporator was bent (you can do this or have the supplier do it) and fits into the freezer side.  A small fan, embedded in the insulation  partition between the freezer and refrigerator  blows cold air into the refrigerator space. The compressor/evaporator cools the freezer to the desired set point and a separate thermostat controls how long  the fan blows the freezer air into the refrigerator. A return air hole is also cut into the partition. I elected to use this active system, i.e. with the fan but its also possible to use a totally passive spill over system where the holes between refrigerator and freezer are engineered to the right size and no extra energy is required. I didn’t think I had a clear enough understanding of the physics so elected to use the fan.  I used two electronic thermostats made by Coastal but it’s just a possible to use electro-mechanical thermostats to both save money and energy.

Before moving on with the project this is a good time to think about any wires you may want to run from the outside into the box or boxes.

Thermostat wires and mounting should be installed with/while the  insulation is installed. Be kind of careful where you mount the thermostats sensors. They should not be touching metal and not be close to the fab or the evaporator. I ended up amking a couple of plastic holders attached to  the stainless freezer and refer walls with plastic wire ties. I also wanted light inside the box so made sure there were power wires and places to mount magnetically activated switches.  The lights are strip LED glued (or bracketed) to the upper inside lip of the cooled spaces. I used two LED strips with separate switches, one for freezer and one for refer.

At this point I’d probably recommend that you not replace the original counter tops (you did take care to not damage them when you removed them?) If you have to buy or make counter tops and refrigerator and freezer doors for your new system, it’s a huge task. There are doors available commercially but they are pretty expensive. Hinged doors are harder to make than lift out doors.

If you decide to make tops and doors there are a few considerations:

• At least two sealing layers
• Adjustable latches to make a tight fit when the doors are closed.
• make the sealing surfaces wide enough to for the gasket material you intend to use.
• Pick the gasket material ahead of time (my 1st material ended up absorbing water and then tore when it froze).
• One has to also  insure that the door swings free of the tops, not a trivial task when there are 4-6 inches of insulation hanging down.
• Or just duplicate the old doors, by making  doors that lifts up without hinges.

I made new counter tops out of Corian.  Corain is hard to get unless you are a commercial interest, so I’d recommend one of the other companies material that place less stringent rules on purchase of the raw material. Get the glue as well. I ended up purchasing the Corain of of Craigs List from a person that used to run a cabinet shop. It was inexpensive, the right color and there was enough for the project so it worked out well. The glue is purchased in the same color as the solid material, so make sure the material you have is not out-dated and therefore the glue color is no longer avaliable. I ended up buying glue multiple times , so I’d recommend getting more than you think.
Corian or the other solid top materials are  easily worked with ordinary wood working tools. Routers to smooth the edges, table saw with carbide blade to cut the top.
One side effect of working with this material (especially with the router) is the chips are very fine and pick up static as they leave the tool bit, so the materials sticks to the tools and the ground and yourself and doesn’t come off easily. Use a respirator when working with the material.

Buy enough counter material for the top and the doors. Originally I though I could cut the door hole in the top and use the scrap as the door, but I couldn’t find a saw with a thin enough blade to make a well sealed door. In the end none of the waste was really wasted because there is always the need to make gussets or other support pieces.

Make the refrigerator doors with dual sealing surfaces and gas shocks to keep the doors up at anchor or in a heavy sea. I bought some stainless door latches from Southco with adjustable up/down adjustment so I can make sure the doors fit  tight.

You can work with Corain (or other solid top material) with ordinary wood working tools

Connecting the tubing and mounting the compressor turned out to be very easy. The manufacturer puts the appropriate amount of refrigerant, R34, into the system with sealed ends that don’t leak when connecting the system together.

mounted in Lazerrette

So, after all of that how well does it all work?

I have yet to try the system in the tropics, but so far, it makes ice, runs far less often than the old system and the variable motor speed controller uses less current to boot. In northern climates (70 degrees ambient) and cold water, with the refrigerator set to 37 and the freezer to 25 it runs at about a three to one duty cycle. In other words for every 20 minutes its on its off for 60.  I’ve had the freezer down to 14 just to try it. And it makes ice.  It looks nice too.

UPDATE: Aug 2016: The refer has been operating for about 5 years now. Even works pretty well in the tropics until I pulled the boat onto the hard and lived in the boat for a few days in Panama. When the keel cooler isn’t in the water the system works but gets really hot.  I didn’t realize it had damaged the system until a few day back in the water when the compressor stopped and threw out a error code. After many hours of vacuum pumping the system is working again but still codes once a week or so. It always starts again after an error by itself after about 2 min. 

I’ve been very happy with the system , but will have to get a vacuum pump and learn how to fix the system myself (after all everything breaks eventually) because I think I can fix it better than the so called experts.

 -Ron

top with doors mounted in boat

showing the gas shocks