Looking to covert to 12v refridgeration

[Richard Kollmann]

Originally Posted by RaymondR

The rebuild of the refrigerator has now commenced. I have decided to persist with the eutectic tanks and use a calcium chloride/water eutectic fluid.

The plan is to raise the tanks above the base of the box and use computer fans to draw warmer air from the top of the box and circulate it across the bottom of the tanks and backwards and forwards under and over metal trays.

One of the things I discovered which would account for the poor performance of the old refrigerator was that the evaporator plates were piped backwards with the liquid refrigerant being pumped into the tops of the plates and a consequent very small refrigerant charge.

Eutectic plates are a form of evaporator and evaporators need to be mounted as high in box as possible. Performance of eutectic plates depend on a number of design factors, Surface area of evaporator and its relationship distance to tanks exterior surface. Calcium Chloride salt is a poor choice unless it is neutralized against corrosion. Eutectic plate energy storage is only practical if there is sufficient surplus of alternative energy or compressor output is excessive for evaporator inside eutectic plate.

You are correct refrigerant volume flow and the area of super heat coverage within evaporator tubing is important. To achieve proper refrigerant flow your system will need a receiver and a Thermostatic controlled TXV to keep correct evaporator super heat while slowly lowering temperature of solution well below its eutectic temperature.

Heat extracted from refrigerated box area must reach the cold evaporator after passing through plate and eutectic solution and be absorbed inside evaporator tubing. Heat is only absorbed by refrigerant during it phase change from liquid to a vapor. If your plates are designed deferring liquid gravity and normal freezing of plate from bottom up refrigerant may never freeze solution in top of plates. Most advertised eutectic plates sold today are really cycling plates too small and require compressor to run even at night.

[RaymondR]

Hi Richard.

The bottom located eutectic plate scheme is now gone and they will now be fitted on the long side of the box. The problem is that the existing box is very irregular in shape making it very difficult to measure and I found I needed about 1/4" more length to get the tanks to line up properly. Unfortunately attempting to implement fan forced air circulation in the side tank scheme will be much more difficult. However there is room to provide an air circulation path around the back of the tanks.

My primary reference since I started messing about with boat refrigeration has been Nigel Calder's Refrigeration for Pleasure Vessels. In Chapter 4 he discusses cold plates and recommends against using antifreeze solutions as eutectic fluids due to their freezing characteristics. Unfortunately the true eutectic fluids he addresses are all salt and water solutions. However having provided a fairly detailed reasons why antifreeze solutions are deficient eutectic fluids he then proceeds to list them in his strength/weight table. I intend running some tests when I install the cooling coils in the tanks to try to gauge the corrosion potential and may decide on another solution if they provide bad results.

My accidental eutectic tank experiment on my previous boat proved very successful in vastly reducing battery voltage cycling and consequent extension of battery life and I found I also had a 2-3 day holdover capacity with the time the ice took to melt. This new installation hopes to emulate the former experience.

When I bought the boat it had insufficient solar panels for a cruising boat which spends long periods away from the dock. This, coupled with the very large ice box they had converted to a refrigerator and the poor installation kept the condensing unit running about 50 minutes per hour. Since I bought the boat I have more than doubled the amount of panels and now have sufficient power available to run two condensing units for much of the day. I am expecting to see 50-60 amps out of the new panel farm and believe that with the reduction of amp hour draw during the night hours the batteries will require far less charging.

I have also been pondering on schemes to exploit the speed variation capabilities of the Danfoss BD50 compressors on the condensing units. I realize that increasing the speed decreases the efficiency and life and that cooling of the compressors will be required. I would appreciate any input you might provide on this subject and whether the TXV valve would be able to manage the refrigerant flow variations of say a two compressor speed arrangement would cause.

I always read your posts and have learned a lot about boat refrigeration from them. Regards RaymondR

[Coolerking]

What are the dimensions of the plate you have, and the interior volume of the box itself?
We have many hundreds of plates in service on all manner of vessels.
They all are CaCL true eutectic plates.

[Richard Kollmann]

RaymondR, Refrigerator Design Concept

The bottom located eutectic plate scheme is now gone and they will now be fitted on the long side of the box. The problem is that the existing box is very irregular in shape making it very difficult to measure and I found I needed about 1/4" more length to get the tanks to line up properly. Unfortunately attempting to implement fan forced air circulation in the side tank scheme will be much more difficult. However there is room to provide an air circulation path around the back of the tanks.

Comment
A single eutectic plate is commonly used where separate box temperature areas are not required. Surface area of eutectic plates will determine volume of heat moving into eutectic solution per hour. Natural air tumbling of heat rising and cold air descending is what normalizes box overall temperature. Spacing plates away from walls increases air contact with plates. Small fan that increase air tumbling across plates can reduce overall box temperature by six degrees.


My primary reference since I started messing about with boat refrigeration has been Nigel Calder's Refrigeration for Pleasure Vessels. In Chapter 4 he discusses cold plates and recommends against using antifreeze solutions as eutectic fluids due to their freezing characteristics. Unfortunately the true eutectic fluids he addresses are all salt and water solutions. However having provided a fairly detailed reasons why antifreeze solutions are deficient eutectic fluids he then proceeds to list them in his strength/weight table. I intend running some tests when I install the cooling coils in the tanks to try to gauge the corrosion potential and may decide on another solution if they provide bad results.

Comment
Calder’s book has good information on components but not those experienced with many mobile refrigeration industry units. He did assemble a eutectic system as shown in this book but did not use this design on his next boat. His use of salt brine and water described how much chemistry was required by trucking dole industry to prevent interior failures do to corrosion. Eutectic plates sold with Salt Brine eutectic have its fill plug welded shut and expansion air space inside plate contains only dry Nitrogen.

Food Grade Propylene Glycol and water is the most commonly used eutectic liquid solution as it is non toxic and non corrosive. Propylene glycol and water eutectics have two identifiable problems, glycol will separate from water over time and it temperature rise over time is 15% faster than brine and water. Separation Glycol/water in boat eutectic plates is not normally of concern do to boats normal dynamic motion. Gradual rising of eutectic freeze point not seen with frozen water is troublesome in selecting true holding plates. Tests show that a Glycol base plate must contain 15% more solution than Brine solutions plates to perform equally. Most installers of undersized plates do not expect them to be used as holdover plates and use them as thermostat controlled cycling plates.

Water freezing and melting although is not a true eutectic is a perfect ice source of stored energy. The temperature of water ice solution remains at 32 degrees F as long as there is ice in plate tank. Unfortunately +32 degrees F eutectic ice is not going to produce adequate refrigerator box temperatures from +33 to +40 F . Freezer box temperatures are best at zero degrees F (-18C). Too insure freezer solution is completely frozen fast enough evaporator coil inside eutectic liquid needs to be 10 to 20 degrees below its eutectic freeze point.


My accidental eutectic tank experiment on my previous boat proved very successful in vastly reducing battery voltage cycling and consequent extension of battery life and I found I also had a 2-3 day holdover capacity with the time the ice took to melt. This new installation hopes to emulate the former experience.


Comment
There are a number of variables that will determine how long a refrigerated box will remain at your desired box temperatures. When Designing 24 hour holdover plates for tropical weather I use one gallon of solution for each cu ft of freezer space and ½ gallon for each cu ft of refrigerator space. Plates designed for non tropical weather plate solutions volumes can be cut in half.



When I bought the boat it had insufficient solar panels for a cruising boat which spends long periods away from the dock. This, coupled with the very large ice box they had converted to a refrigerator and the poor installation kept the condensing unit running about 50 minutes per hour. Since I bought the boat I have more than doubled the amount of panels and now have sufficient power available to run two condensing units for much of the day. I am expecting to see 50-60 amps out of the new panel farm and believe that with the reduction of amp hour draw during the night hours the batteries will require far less charging.

Comments
A combination of Wind generator and Solar panels is the best alternative energy but you still need a large house battery bank. If solar is only source of extra energy I recommend panels rated at 100 watts per cu ft of refrigerator box and 200 rated watts per cu ft of freezer box if cruising in tropical climates. Live aboard boaters now days are carrying Honda portable generators with 100 Amp charger to handle redundancy problems.


I have also been pondering on schemes to exploit the speed variation capabilities of the Danfoss BD50 compressors on the condensing units. I realize that increasing the speed decreases the efficiency and life and that cooling of the compressors will be required. I would appreciate any input you might provide on this subject and whether the TXV valve would be able to manage the refrigerant flow variations of say a two compressor speed arrangement would cause.

Comments
Normally with eutectic plates you will want maximum refrigerant flow from compressors when there is a charging current above 13.6 volts. Maximum refrigerant flow for BD50 compressor is defined onDanfoss Spec sheets. The size of TXV orifice and valve manual adjustment will keep compressor within design limits.
When a charging current above 13.6 the compressor will need a 1500 ohm resistor in thermostat wire. When voltage is below 13.6 a relay can be used to switch 1500 ohms to best resistance COP. You may want to use a low cost battery combiner to switch 1500 ohm resistor out of circuit when charging current is low and an electronic thermostat with adjustable differential.

[RaymondR]

Thanks again for your advice Richard.

I have had to reduce the size of the tanks from 500 mm by 350 mm by 150 mm to 100 mm to get them to fit with a shroud 20mm clearance all round the tanks so air circulation around the back of the tanks should be good. However the reduction in volume will reduce the hold over time.

Pondering further on the TX Valves and mindful of a previous post of yours where you ran three plates with individual TX Valves in parallel I am now wondering whether one could use a TX Valve for each compressor speed with a solenoid valve to switch in the second
Valve along with a relay to switch in the resister circuit for the compressor speed increase. The valves are not expensive and having a second TX Valve would allow system tuning for each compressor speed mode.

[9cx5q9cq]

I worked with both of them at great length. I really liked Oz Pete's system. Very slick. In the end I opted to go with 2 separate Nova Kool systems . About the same as OzPete's system which is a single system with spillover. I liked the simplicity of the thin plate. I like that I have 2 separate systems that can be either fridge or freezer. If one stops... no big deal. Frustrating but not a big deal.

[Richard Kollmann]

RaymondR I have to agree with 9cx5 about the total separation of having two completely independent systems that are maintainable and repairable any where in the cruising world without special parts.

I failed to mention before that evaporator coil tubing diameter inside your eutectic plates must be no larger than the low pressure fitting on compressor to insure oil returning flow velocity. Sounds like your design concept is for a simgle temperature box either freezer or refrigerator. Two eutectic plate's performance is much improved with two condensing units, two receivers and two TXVs.

Doing the math and staying within basis proven design practices will insure the success of your refrigeration unit.

[RaymondR]

I will have two separate systems with a condensing unit for each eutectic tank. What I am exploring is whether or not I could implement both a low speed compressor setting for periods of low Amp availability and a second of high availability. There are large variations in the power available between morning, middle of the day, and afternoon.

My apprehension of TX valve operation is that the orifice sets the back pressure at the maximum allowable refrigerant flow rate to ensure state change in the condenser and the valve acts to reduce it to keep from slugging into the compressor.

The potential problem would appear to be could a single metering orifice provide sufficient flow at both the high and low compressor speeds?

[Richard Kollmann]

Originally Posted by RaymondR

I will have two separate systems with a condensing unit for each eutectic tank. What I am exploring is whether or not I could implement both a low speed compressor setting for periods of low Amp availability and a second of high availability. There are large variations in the power available between morning, middle of the day, and afternoon.

My apprehension of TX valve operation is that the orifice sets the back pressure at the maximum allowable refrigerant flow rate to ensure state change in the condenser and the valve acts to reduce it to keep from slugging into the compressor.

The potential problem would appear to be could a single metering orifice provide sufficient flow at both the high and low compressor speeds?

In each of your condensing units it will be the TXV that controls refrigerant flow through evaporator. The TXV temperature range and orifice you select is based on max flow for your system design. My thinking based basest on Danfoss BD50 spec sheet, you will want a valve with a maximum full open orifice range flow of 1300 Btu and a automatic temperature range from +40 F to -40 F for a Danfoss BD50 compressor. When starting compressor for the first time with warm evaporator and correct amount of refrigerant in receiver tank flow through TXV may exceed 1300 Btu. As Txv temperature sense tube clamped to outlet line of evaporator detects cold temperature refrigerant flow and low pressure/temperature through evaporator will continue to decrease to 800 operating Spec maximum Btu and keep decreasing until thermostat stops compressor. Compressor speed is not involved in TXV adjusting refrigerant flow. Running compressor at low speed only causes compressor to run longer. The daily energy needed to freeze eutectic plates solid can be reduced only through slower compressor speeds after plate is half frozen. A 1500 ohm compressor speed resistor when a charging current is available and no resistor when running off storage battery. The automatic speed controls available for Danfoss compressors are not designed for energy storage.

Fixed orifices are used in some air conditioning units and capillary tube systems where a narrow evaporator temperature range is required. Industry experience has proven full range +80 to – 30 degrees F is the needed range for refrigerant temperature flow control devices for boat refrigerator holdover eutectic plates.

[RaymondR]

Originally Posted by Coolerking

What are the dimensions of the plate you have, and the interior volume of the box itself?
We have many hundreds of plates in service on all manner of vessels.
They all are CaCL true eutectic plates.

Hi Coolerking.

The design of the box and the number of plates is still a work in progress but I'll provide the dimensions when I freeze the design.

Do you include any corrosion inhibitor in the CaCl eutectic solution? I'm wondering whether a small addition of NaOH to raise the Ph into the basic range might be helpful.

[Coolerking]

Our plates are stainless. Copper tube is bent and placed into box, then back plate is spot, then seam welded to seal, the tube exits are soldered shut, plate is leak checked, we then compress the sides of the plate, fill with whatever temp solution we are using, seal and then unclamp it.
To this date we have not lost one that I am aware of, we have thousands of them in service over a period of at least 25 years.
We are building some big ones now, I'll try and post some pics up.
We don't add anything other than the pure eutectic mix.

[RaymondR]

Originally Posted by Coolerking

Our plates are stainless. Copper tube is bent and placed into box, then back plate is spot, then seam welded to seal, the tube exits are soldered shut, plate is leak checked, we then compress the sides of the plate, fill with whatever temp solution we are using, seal and then unclamp it.
To this date we have not lost one that I am aware of, we have thousands of them in service over a period of at least 25 years.
We are building some big ones now, I'll try and post some pics up.
We don't add anything other than the pure eutectic mix.

That's good information Coolerking and very reassuring to have and I will fabricate my tanks the same way, thank you, RaymondR.

The new box is almost fabricated and ready for epoxy sealing and fibre glassing of the inner surfaces. The image illustrates the section containing the eutectic tanks. It's been a bit trying having to make it in sections which will fit through the existing hatch but I'm slowly getting there.

[Cgstock]

Just built a stainless box for an Ozefridge system. Have not connected it yet as other projects came up, but the quality of the unit looks great.

https://www.facebook.com/sailingknoc...5929734268695/

[RaymondR]

Hi again Richard,

For me TXV valves appear to be a whole new area of expertise I need to acquire and it's proving to be fairly heavy going.

The first obstacle was to set in my mind that a TXV valve is not a temperature control device it is a flow control one and now that I appear to have achieved that mindset I'm finding the going a little easier.

I'm still working my way through your post #39 and as usual the going is being obstructed by the damned metric system and unit conversion needs. Refrigeration tons and btu's appear to have a connection to the real world which appears to be sometime missing with the metric system. However I have now discovered that if one downloads Danfoss information from a US source it provides specifications in both American and metric units so hopefully this will help.

Again, thanks a lot for your assistance.

[Richard Kollmann]

Originally Posted by RaymondR

Hi again Richard,

For me TXV valves appear to be a whole new area of expertise I need to acquire and it's proving to be fairly heavy going.

The first obstacle was to set in my mind that a TXV valve is not a temperature control device it is a flow control one and now that I appear to have achieved that mindset I'm finding the going a little easier.

I'm still working my way through your post #39 and as usual the going is being obstructed by the damned metric system and unit conversion needs. Refrigeration tons and btu's appear to have a connection to the real world which appears to be sometime missing with the metric system. However I have now discovered that if one downloads Danfoss information from a US source it provides specifications in both American and metric units so hopefully this will help.

Again, thanks a lot for your assistance.

There are three different refrigerant flow control devices used in small boat refrigeration systems:
1. Evaporator temperature manually adjustable low pressure regulators that are designed to maintain a fixed operating temperature/ pressure throughout compressor running cycle. This Evaporator Pressure Regulator (EPR) can be set to maintain a fixed evaporator pressure and refrigerant flow. The EPR valve is also used on boats with two separate temperature refrigerator and freezer evaporator plates. SeaFrost is the only company in the US that uses the EPR valve on small Btu compressor design. Do to the pressure drop caused by length of refrigerant lines and number of line angle changes adjustment of the EPR valve will be different on every boat it is installed in.
2. Capillary tube refrigerant flow control device Advantages are, They are less expensive, No moving parts and does not require maintenance, Its orifice is always open to flow so pressure differential is bled off when compressor stops reducing starting load on next cycle of thermostat, A good choice on small hermetically sealed compressors with standard thin plate evaporators. Disadvantages of Capillary tube flow control devices are: They cannot respond to ambient air, water temperature changes. A twenty degree F change can vary refrigerant charge resulting in 25 gram of refrigerant gain or loss on small 12 volt BD compressors will affect systems performance. Capillary tube system’s refrigerant charge volume is critical and is normally preset by the system manufacturer based on standard day temperatures of 70 degrees F. The assumption in these system designs is normal operating compressor temperatures will not cause oil or component failures. Clean refrigerant is very important on cap tube systems which means the normal refrigerant filter/dryer needs to be in a convenient location just before liquid refrigerant reaches beginning of Cap tube. Large Btu cap tubes have large orifices that can be unplugged but small Btu tubes must be replaced when restricted by solid particles. A major disadvantage of capillary tube flow control devices is they are not adjustable so evaporator superheat depends on system temperature and refrigerant volume changes. These cap tube systems are never used today on eutectic energy storing holding plate evaporators do to their limited control range. Evaporator Cap tube range for a refrigerator temperature control range is +5 to +20 degrees F while range for a freezer is -10 to +5 degrees F.

3. Thermo Expansion Valve (TXV/TEV) refrigerant flow control advantages, are it is designed to adjust flow of refrigerant into evaporator so that only vapor exits evaporator. While maintaining heat absorbing liquid phase change to vapor, commonly referred to as super heat, the TXV maximizes refrigerant liquid flow into evaporator. The automatic control of TXV orifice opening and closing adjustments are the pressure of pressure changes in the small tube from TXV and the other end clamped to outlet tube of evaporator. The TXV valves used on refrigeration needs to have internal pressure equalization so next compressor start up is at low torque. Because the TXV regulates the volume of refrigerant circulating in the systems loop a refrigerant storage tank defined as receiver is required to accommodate excess liquid as valve closes down reducing evaporator lower temperatures.