Shannon 38 - New Refrigeration

[mesquaukee]

You must have a very early copy of what I wrote. In later versions my focus has been on the differences of performance between using capillary tubes with cold plates and TX valves with evaporator plates.
In all the tests we made me measured the input amperage to all the equipment, the temperatures and pressures over several days or weeks. We used the manufactures data or their design programs for the compressors performance at those conditions.

Many cruisers fail in their initial attempts to power all their systems with solar panels and wind generator. They discover after leaving home they use more power and produce less than they thought they would. Trying to correct these problems in foreign countries tends to be expensive.
All too frequently they also discover their refrigeration unit is inadequate to deal with the heat load in the tropics. Their options are limited. Good quality insulation is almost impossible to find. The option is to increase the rate of heat removal. If increasing the ventilation of the condensing unit does not help, then modifying it is the next step. This is where I have had a lot of positive results as outlined in the current version of my book.
If modification is not an option then a 2nd install it yourself system can be imported at considerable expense or use locally available parts to build a larger system.

[mesquaukee]

Originally Posted by Richard Kollmann

mesquaukee, You are correct the real breakthrough in boat refrigeration is in the ability to adjust compressor speeds to balance system capacities for a greater SCOP. I assume your 110 volt compressor is fixed speed, what is rated capacity and COP of your compressor? If this compressor is driven by a 12 volt inverter, what is the inverter’s COP while driving compressor? How have you calculated btu capacity of condenser and evaporator to insure adequate return super cooled gas for compressor cooling?

Took me a while to dig up that info. The boat is a mess, doing a refit.
The compressor is a Tecumseh THA1340YXA, info on it is at Welcome to N A C G &Bom=TH101AR-236-A2
At the conditions we were running 14F evaporator and 105F condensing, 95F liquid we extrapolated from the published data (see above tech sheet) a 1015 BTU/hr cooling. The current measured was that drawn by the inverter, control circuitry and fans. It was 11.1 to 11.3 amps yielding a COP of 7.36 BTU/WattHr.
A Danfoss using the RS+2 design program at the same conditions at its most efficient speed would have a COP 5.86 BTU/WattHr.
The low side pressure control was later adjusted to give approximately 50 cycles per day or 2 an hour, the cut in being 20F and cut out 5F (from 20F to 10F). The compressor has now run continuously for 3 years so I would have to presume that it has received proper cooling.

The original BD3 needed 82 amphrs/day to maintain the box at 40ºF correcting for the water pump

The Tecumseh used 70 amphrs/day (amperage as drawn by the inverter and all other equipment) in normal everyday use and the box was colder was colder (freezer 20F and fridge 35F). When this was taken into account the the BD3 would use 39% more ampHrs.

A BD35 running at 3500 rpm is approximately 10% to 15% more energy efficient than the BD3. This would indicate that the 110 volt tecumseh compressor is 28% to 33% more energy efficient than the DB50.

On a further note Danfoss produces two high efficiency compressors the NTX5.2FK and the TTEE4F which if you believe their numbers are 20% to 40% more efficient than the DB50.

The low side pressure control was adjusted to give approximately 50 cycles per day or 2 an hour, the cut in being 20F and cut out 5F. The compressor has now run continuously for 3 years so I would presume that it has received proper cooling.

[chala]

Originally Posted by mesquaukee

[FONT=Calibri][SIZE=3]The evaporator should if possible totally surround your frozen food compartment. You do not want it only on two sides.

Interestingly some 35 years ago when I was involved in building bush freezers, we found out that evaporators used as partitions were the most efficients.

Originally Posted by mesquaukee

[FONT=Calibri][SIZE=3]Since you will be generating your power from the wind and solar stay away from any system that has a cold holdover plate which uses a eutectic (or any liquid) solution to store cold.

I would not give up on cold holdover but I would use cold holdover separate from the evaporator.
A normal solar regulator regulates the voltage by producing heat.
The same regulation can be achieved by increasing load on the solar array.
The refrigeration unit can easily achieve the loading by storing energy into cold holdover. This can be done simply manually or automatically using a system of reservoir and transfer pump.

[Richard Kollmann]

mesquaukee, From all the opinions and theories you have stated I do agree that the main function of a holding plate is to store energy and the higher the SCOP the greater the energy efficiency of a refrigeration unit. I would like to believe that you have found the secret to achieving higher energy efficiency by gas flow Enthalpy greater than anyone else has in the mobile refrigeration industry.

These following comments you made I believe are incorrect:
· Since you will be generating your power from the wind and solar stay away from any system that has a cold holdover plate which uses a eutectic (or any liquid) solution to store cold. They are energy hogs using as much as 2 times as much as other systems.

After twenty five years designing over 100 boat refrigeration systems using holder plates with many of these connected to 12 volt condensing units I find the daily 12 volt energy consumption much the same between conventional evaporators and eutectic plates. The performance difference between the two types of evaporators is refrigerator box’s temperature swing is greater when using a eutectic holding plate and holding plates are slower in absorbing heat than thin aluminum evaporator plates. If you ignore box temperature swing my tests show holding plates are 6 % more energy efficient than standard evaporators. On small 12 volt refrigeration the true advantage of a eutectic plate is to store alternative energy from wind and solar relieving stress on batteries extending their useful life. Danfoss now offers design information on a BD Solar 35K that can operate from 10 to 45 volts without batteries or the normal boats power grid. Excess energy of this solar unit is stored in a eutectic plate for use at night.

· You do not want one that uses a capillary tube for metering the refrigerant as once again they are energy hogs ( use 20-50% or more ampHrs)

Rate of evaporator/eutectic plate temperature change might be 20 to 50% slower but the difference in daily amp-hrs is not going to be much different than other types of refrigerant expansion devices. Thermo expansion valves do have a greater control range than capillary tubes and can help balance a poorly designed systems. In a properly designed capillary tube system it is the evaporator’s size and desired operating temperature range that determines the length and diameter of cap tubes internal orifice. There are design condition where selecting an expansion valve over a cap tube is not recommended, Examples Danfoss compressor BD3 and BD80 are both designed for low back pressure applications only and extended running at high suction pressure will over stress compressor’s electronic module.

· The capillary tube has to have the same rate of flow of refrigerant as the compressors’ rate of flow.


A capillary tube’s capacity is designed to keep refrigerant heat absorption inside evaporator and is not designed to necessarily handle max compressor capacity.



I have compared your conclusions on compressor COP against Danfoss’s application engineering data and the THE1340 compressor data and can not find anyway to agree with your assumptions do to differences in condenser gas temperatures and Tecumseh specs refer to the mass gas flow rates in their COP calculation. If you have found a way to improve the ratio you quoted between supply of energy (mass flow multiplied by enthalpy difference) to the refrigerant from the outlet of the condenser to the inlet of the compressor you should make a fortune.
You are right that mobile refrigeration units designed for the RV industry have poor performance in warm climates because their condensers are too small. US companies assembling small 12 volt icebox conversion boat refrigeration systems recognized that boats do transit from cool to very warm climates. Most of these units were designed to use air as a condenser cooling medium as air temperature was much easier to control with climate change than water temperature.

[mesquaukee]

My sole area of interest concerning the design and installation of mobile refrigeration systems is on sailboats where the desire is to use solar and wind power as the energy source. Since space is restricted for the mounting of solar panels (not to mention their expense) it would be imperative to investigate ways to reduce the energy consumption of the refrigeration system.

Theory suggests how an efficient system should be constructed, practice has proved theory correct. Now I am in the process of explaining what criteria are necessary for the most efficient system possible.
According to FrigoBoat promotional material at http://www.frigoboat.com/sysdesign/aluminumevap.html they state quite clearly that for small 12/24 volt systems cold holdover plates are not as efficient as evaporator plates.
There is a significant energy efficiency difference between a refrigerant evaporating at the temperature you wish to keep your frozen food at and the temperature it must evaporate to freeze the solution in a cold holdover plate located on one side of the frozen food.
In comparing the 2 methods of keeping food frozen we will have an immediate 20% increase in ampHrs as the freezer compartment has to be larger to contain the same amount of food. This is derived from the rule of 1 gallon of solution per cubic foot of box for a freezer and the hardware to contain and mount the tank and the rule that heat load is proportional to box size.
To this we add 28% due to the lower evaporation temperature, possibly as much as 64% or more due to the design of the system. Below is the explanation.
Evaporator plate surrounding freezer; -the refrigerant evaporates between 5 to 15 F yielding a COP of 5.9 (assuming 100F condensing, 20F subcooling, 55F total superheat).
Cold holdover plate at one end of the freezer box; -you will need a evaporation temperature of -10 to 0F which yields a COP of 4.6. See below for a definition of COP.
This results in a 28% increase in energy use provided the capillary tube is designed for these conditions which it most likely is not. More likely it is designed for -5 to 0 for a condensing temperature of 120F to compensate for changing or uncontrollable ambient conditions. This would give a COP of 3.6. This results in an increase of 64% for the same box temperature.
A capillary tube is unable to compensate for changes in ambient temperatures in an energy efficient manner. A TX valve can, its sole purpose is to make the evaporator work as hard as possible.
The problem with a capillary tube is that it can never be of a diameter and length to match the wide range of ambient conditions that are found cruising.
According to the literature for a 10F evaporator the capillary tube for a condensing temp of 130F must be 3 times longer than for a condensing temperature of 105F. Please see attachment.
A TX valves ability to adjust to the most efficient use of the evaporator reduces the compressor run time. It is true it draws more current during this shorter time of moving heat faster. To determine if this is more efficient we use COP. COP is a ratio of heat moved to energy used to move it, usually expressed as BTU/watt, perhaps if it were expressed as BTU/ampHr we would recognise its importance.
It is true that one can directly use solar power to run the refrigeration system (see Danfoss webpage listed below and pick “application description” under the Literature Tab). If you have the battery capacity it is better to charge the batteries and then use the power at night. You will have 10% to 50% more cooling than using a cold holdover plate. (loose 10-15% by storing and then recovering the energy from the batteries compared to a 28-64% loss freezing the solution in the cold plate).
Both the BD35 and BD50’s usual operating envelopes range from -22F to 22F. An evaporator operating in the range of 5F to 15F falls well inside this envelope. The only requirement Danfoss makes is to use the optional fan cooling kit for the electronic module if operating it over 3000 RPM at an evaporator temperature of 15F. This can be found in Danfoss’s Product Catalogue under the Literature tab, the last item on the page “Catalogue”. Please see http://www.danfoss.com/Products/Categories/Categories.htm?segment=RA&category=http%3a%2f%2fww w.ra.danfoss.com%2fra%2fProducts%2fProductCatalogu e.asp%3fNavigation%3dHideOnAllPages%26Footer%3dHid eonallpages%26Division%3dCC%26HL%3d1%26TopViewItem %3d74%26AppID%3d%7bc2a95dac-5e01-47df-92cd-5972d837cf1a%7d%26dyn_lang%3d
On page 3 of the above catalogue the BD35 and BD50 can be used both with capillary tube or a TEV (TX valve or thermostatic expansion valves).
On page 15 it states that the BD50 can be used for LBP/MBP/HBP applications. It is also listed for an evaporation range of -30C to 0(10)C or -22F to 50F provided the compressor is fan cooled above 32F.
As for discovering how to move more heat there is nothing new to efficiencies of scale. The larger a system is generally the more energy efficient it is. There is a balance point of amperage drawn which if too large begins to involve the Peukert Effect and if too small means using smaller less energy efficient compressors.
There are a number of 110 volt hermetic condenser systems an example of which is Grunert. They are ¼ hp and larger. They are too large for our purposes.
According to readily available material at http://www.tecumseh.com.br/ and at Danfoss (see above) the THA1340YX compressor is 30% more efficient than the BD50 by Danfoss at the same operating conditions. Accounting for the 7% to 9% loss of an inverter it is 21-23% more efficient. It also moves more than twice the amount of heat.
The Tecumseh THA1340YX compressor at 5F evaporator and 100F condenser (90F liquid, 7F superheat and 90F compressor suction) moves 839 BTU/hr at 1.1 amps (11.2 amps, 12 volts nominal) to give a COP of 7.42 BTU/watt (ASHRAE COP 4.4 BTU/watt).
The BD50 operating under the same conditions has a COP of 5.75.(ASHRAE COP 3.91 BTU/watt).
Danfoss has a compressor similar to the Tecumseh it is the NTX5.2FK. Its COP is 6.83. (ASHRAE COP 5.61 BTU/watt).

[chala]

Mesquaukee
Do you have a working system or is it still at the design stage?

[mesquaukee]

I and friends of mine have modified existing systems using the BD35. We have also constructed a couple from scratch using the Tecumseh compressor.
The results we obtained where in the ballpark of what theory would suggest.

[Richard Kollmann]

mesquaukee, I Do not question the possibility that you were able to modify a conventional standard design to make it more compatible with a particular boat’s cruising needs in the tropics. Customizing a system by selecting the correct components is what makes the difference between mobile refrigeration and conventional package refrigeration. Package refrigeration like a home refrigerator is generally inside a creature comfort area where it is rarely exposed to extreme heat or cold so there is no need to regulate the condenser cooling medium. When using theoretical percentage figures there is no way that you could have the instrumentation on your boat or other boats to confirm these very high percentages you are quoting. Performance of a refrigerator is measures by energy put into a system versus how much you get out and this is defined as Coefficient Of Performance (COP), in other words the ratio of output energy divided by impute energy. To confirm your large percentage improvements you would need actual Btu data. A Danfoss BD35 compressor with a condenser temperature of 131 degrees F and evaporator temp of 14 degrees F, running at 2000 rpm has a COP of 1.82, or if running at 3500 rpm it is less efficient with a COP of 1.68. The ability of the new generation of variable speed compressors equipped with manual or automatic speed control allows a unit to optimize COP to each refrigerator and changing climate conditions. The best COP I could find using Danfoss perimeters applied to Tecumseh THA1340 compressor was 1.64 and maybe as high as 1.74.

For those that believe a eutectic holding plate is more efficient or less efficient than a standard efficient evaporator you need to understand the reason why holding plates are useful in mobile and transport refrigeration. A eutectic holding plate serves only one purpose and that is to store surplus energy for use when impute energy for refrigeration is not available. Holdover plates are more efficient when small 12 volt compressor's capacity is slightly larger than a plate’s capacity to absorb refrigerator box’s heat and even more energy efficient when there are available sources of alternative energy from wind, solar, water generators or shore power. Technautics Cool Blue BD35 compressor and oversized condensing unit connected to a large surface area eutectic plate achieve very efficient daily energy usage in the right size box. If the heat absorbing skin area of a eutectic plate satisfies desired box temperatures it is not necessary to freeze all of a plate’s solution solid unless freezing is accomplished with free alternative energy. It is also true that when a large engine driven or large 12 volt condensing unit with capacity to freezing ¼ to 1 ½ tons of ice in one hour is connected to one, two or three average boat refrigerator/freezer, eutectic holding plates it would not be considered an efficient use of energy.

One of the side benefits of eutectic holding plates frozen by small 12 volt units from alternative energy is much longer battery life do to less battery stress. The disadvantage to eutectic plates is they do take up box’s available space; they also are very slow to pull box temperature down the first day, and make it difficult to operate a single box spillover refrigerator/freezer system.

[mesquaukee]

Precisely what I am saying a custom designed system is more efficient.

To remove the problem of precise measurements of temperatures and pressures we place the entire system in a black box . The only measurements then necessary is the temperature of the freezer and refrigerator compartments, the pressures of the low and high side and the total current drawn by the entire system. We are unconcerned with any other measurements.


An effort was made to measure the temperatures of the liquid entering the box and temperature at the inlet to the compressor which is not the actual compressor inlet temperature. They were discarded as they were unreliable. The COP numbers I quote are for identical pressures and temperatures for both the original (capillary tube) and modified system (TX valve). They were averaged from those readings.

In one modification we did of a cold plate system used a cold plate 10” by 2.5” by 16”, or 3 sqft of surface area. It was a spill over system. It used 82 amphrs/day for a box temp of 34F freezer and fridge temp of 40F. It ran almost continuously.


Modifying it gave a box temp of 18F and 35F for 47 amphrs/day with an evaporator surface area of 8 sqft. Taking in account it is colder it used 48% of the ampHrs of the original system.

The key for an efficient system is as large a surface area as possible for heat transfer, it is roughly 20 to 30 Btu/sqft. To raise this rate we need to lower the evaporator pressure decreasing efficiency. In the case above the temperature differential between the refrigerant and compartment must be quite a bit larger, at least 3 times.


This is why a lot of small 12 volt holdover plates do not work well in the tropics. Not enough surface area to absorb heat, the box temp is elevated. The thermostat keeps the compressor running at a low evaporator pressure to create a larger temperature differential in an attempt to increase the heat transfer. That is why I propose lining the entire freezer compartment with evaporator plate.

If you desire cold holdover plates use ones with large surface areas.

I do not agree with the battery stress issue for using cold holdover plates. In essence you will need to increase the solar panel area by 20-40%, or more, due to a larger box and reduced evaporator pressure. This is a problem on monohulls, where to mount them?

[mesquaukee]

There is some confusion about the COP each of us is using. The COP I am using is BTU/watt. Dividing it by 3.412 btu/watt would make it unit less, which is the COP you are using.

During the day is the time the most pressure is placed on the system; opening and closing, chilling down beverages, purchasing food and chilling it down or freezing it. With a holdover plate the refrigerator temperature rises due to the lag of the system as you mentioned. An evaporator can deal with this heat load more effectively by running more right at the time the solar panels are putting out maximum power. The temperature of the box stays more even.
To take advantage of the solar power during the day and use the cold plate to keep the box cool at night we would need to lower the thermostat setting during the day and raise it at night. This would help to reduce the temperature rise during the day. Obviously more energy would be used to do this.

[donradcliffe]

Let me say first, that I appreciate and enjoy the dialog--I'm not up to speed on the details, but it has always seemed to me that some of the claims for refrigeration systems didn't always obey the laws of thermodynamics. I think I need to read Richard's books first, but from first principles, it would seem that a holdover system would have to be less efficient than an evaporator. The maximum COP for a perfect system from thermodynamics is

COP = Tcold/(Thot-Tcold).

Now, if you want to maintain your foodbox at some minimum temperature, you will need the coolant to be at some lower temperature (Tcold). In a holdover system, you must transfer the heat from the coolant to the eutectic to the plate surface to the box, effectively insulating the coolant from the box and requiring a lower Tcold than a large evaporator. The holdover systems also cycle less often, which means that the Tcold must be even lower to maintain minimum box temperatures. The lower Tcold means a lower COP, given all else equal.

Now lets turn to the required cooling capacity. The holdover systems need bigger pumps/higher cooling capacity/higher coolant flow rates, because they only run occasionally. It would seem to me that a small pump run nearly continously would be more efficient.

Compicating this is Peukert's effect, which tends to skew the results of side-by side refrigeration systems. If your main source of energy is solar, refrigeration that draws down the batteries at night will require about 25% more energy (to recharge the batteries) than refrigeration that runs during the day (when there is an excess of solar power). Puerkert's equation also says larger motors effectively draw more amp-hours out of your batteries than smaller motors for the same amount of work.

Boats with gensets and no solar are better off with the holdover systems, because there is a big excess of available energy when the genset is on, and as long as you are committed to running it twice a day, efficiency is not important.

[witzgall]

Obmmedic;

Isotherm is selling off some excess inventory at good prices right now thru eBay. I just bought a 4203 setup from them. They have others listed as well. Look for the ebay store "Marine_refridgerators". I did not realize it was Isotherm USA doing the selling until I called them to see if I would be buying the right stuff for our needs. Since I already have the seawater intake and discharge, I hope it was a good choice for us. Is is a holding plaate with ASU model, designed for freezers up to 2.8 cu ft.

My intention is to partition off an upper-shelf to make about a 1cu ft freezer. The rest, only about 2.8 cu ft, will be fridge via spillover. The unit should be here on Friday - I'll post when I get it.

Chris

[mesquaukee]

In speaking with Randy Simpkins of Technautics he confirmed what I have noticed with most capillary tube systems used in the heat of the tropics where the condensing unit is poorly ventilated.

For a capillary tube system to work you cannot use an oversized condenser. If you do you will need more refrigerant to fill the system which will result in the possibly of refrigerant flooding the evaporator and flowing into the compressor damaging it. This limits the systems use.

I have met numerous people whose systems seem to run almost continuously. Raising the set point of the thermostat a few degrees immediately causes the compressor to cycle properly cutting the run time down. The problem is the freezer box is now around the freezing point and the fridge is 40 to 45F. This is woefully inadequate.

For most capillary tube systems in the tropics using air cooling the energy use is halved using the CoolBlue system. It uses a TX valve and a grossly oversized condenser and is charged with 26 oz of refrigerant compared to 2 to 3 oz for a capillary tube system. There is a sight glass for visual confirmation that the system is adequately charged. Due to the large size of the condenser water cooling is not required.

This I have confirmed in the modifications I have made to systems. I have obtained condensing temps of 95F with an air temp of 85F, turning off the cooling fan the temp creeps up to 100F. No need to bother with water cooling with all of its problems.

The publicized cooling capacity of a popular system using the BD50, capillary tube and an evaporator box are;

650Btu/hr for an evap temp of 25F and a condensing temp pf 130F

538 Btu/hr for an evap temp of 20 and a condensing temp of 120

331 Btu/hr for an evap temp of -4 and a condensing temp of 95

You can run this system 24 hours a day at a condensing temperature of 130F and not be able to freeze anything solid as the freezer box can never get colder than 25F. Days to make ice cubes.

To be able to solidly freeze anything you have to go with water cooling. You cannot use a keel cooler due to the evaporator flooding problem. You have to use a pump which draws a fair amount of power.

A TX valve does not suffer from the evaporator temperature being tied to the condensing temperature. The use of a TX valve as the refrigerant metering device is as close as you can come to the ideal system.

[Richard Kollmann]

mesquaukee,
Some of what you say makes sense in attempting to convince everyone that all small 12 volt refrigeration on the market today whether air or water cooled is 30 to 50% inefficient in theory compared to your design concepts. You discount the fact there might be at least one efficient system in the 100 plus icebox refrigeration conversion options available today that will be efficient in a particular boat’s wide range of climate conditions. Yes, there are units designed for the RV industry being marketed by WAECO, Isotherm and maybe others as boat refrigeration units. Your example of BD3 compressor connected to a very small eutectic plate was surely a poor unit selection for a warm climate based on your figures.

You must believe if warm conditions poorly ventilate the condenser then cool conditions will over ventilate condenser again causing poor efficiency. Did you mention to Randy at Technautics your theories about thermodynamics when using his holding plates? Such as:

· Stay away from any system that has a cold holdover plate which uses a eutectic (or any liquid) solution to store cold. They are energy hogs using as much as 2 times as much as other systems.

· With a holdover plate the refrigerator temperature rises due to the temperature lag of the system when warm food is added.

· The problem with cold holdover plates is you need;
1.) a temperature differential between the refrigerant and the solution in the cold holdover plates
2.) a temperature differential between the food and the frozen solution in the plates
3.) there is the temperature differential through the food to the side without a cold plate.

[chala]

Originally Posted by mesquaukee

For a capillary tube system to work you cannot use an oversized condenser. If you do you will need more refrigerant to fill the system which will result in the possibly of refrigerant flooding the evaporator and flowing into the compressor damaging it. This limits the systems use.

I have not experienced this so far. I must admit that I have a liquid trap.

Originally Posted by mesquaukee

Modifying it gave a box temp of 18F and 35F for 47 amphrs/day with an evaporator surface area of 8 sqft. Taking in account it is colder it used 48% of the ampHrs of the original system.

240V compressor, 24V inverter and a good box insulation get you down to 20A/hrs (24V) in the tropics, with less weight, and an eutectic “brine” system. But then how reliable is an inverter?

Originally Posted by donradcliffe

Boats with gensets and no solar are better off with the holdover systems, because there is a big excess of available energy when the genset is on, and as long as you are committed to running it twice a day, efficiency is not important.

So true even with some solar.