Electric Refrigeration
Replacement of the refrigeration is a prerequisite for removal of the diesel. The replacement will be an electrically driven compressor with a cold plate rather than simple evaporator. The amount of cooling BTUs will need to be determined for worse case ambients (tropical) to size the unit and the power draw will need to be determined.
Estimating BTU Requirements
Two relatively simple methods of estimating refrigeration needs are available. A bit of a shot in the dark is to look online for advice on estimating BTU load based on ice box size and on number of people aboard. Methods differ so this yields a range. Another basis for an estimate is measuring the thermal loss by melting ice in the ice box.
Estimates Based on Ice Box Size
Two sources of estimates were used, MyBoatsGear.com and Sea Frost.
A large (4 cubic foot) but well insulated (30R) with two people aboard is estimated at 4400 BTU/day by MyBoatsGear.com equipment reviews. A Sea Frost BDxp can produce 3.9 BTU/Wh so this estimate yields about 1100Wh per day. The MyBoatsGear.com method assumes that the number of times the ice box is openned is a function of the number of people aboard. The high estimate for a small box seems to imply that MyBoatsGear.com is assuming the ice box is used largely for beer storage and openned often.
Sea Frost does not give method or assumptions but estimates that an 8 cubic foot box will use 57 Ah per day at 12V (just under 700Wh per day).
This gives a wide range of 1100Wh/day for 4cuft to 700Wh/day for 8cuft (88-275 Wh/cuft). These estimates are quite different with the smaller box estimated as requiring over 50% more energy.
Estimates Based on Ice Melt Testing
Another method involves measuring the heat flow out of the ice box at a known ambient and estimating the heat flow at a higher ambient. The box can be assumed to be at about the freezing temperature of water. The temperature in the box is usually above that with the cold plate quite a bit colder. Using 32F as the temperature of the inside walls of the box yields slightly higher BTU flows when scaled up to higher ambient temperatures.
A first step is to measure the heat flow out of the ice box. This measurement is done by putting ice in tupperware in the box and measuring the ambient and weight of ice melted. The melt rate will be higher if the time duration between measurements is shorter due to heat lost when the box is openned to weigh the ice.
First pre-cool the box for at least an hour. Then drain off water, weigh the ice, record the time and ambient temperature (ouside the box), and put the ice back. Then periodically (each 4-8 hours) remove the ice, drain off water, weigh the ice, record the time and ambient temperature, and put the ice back. Repeat until the ice is gone. Water latent heat of fusion is about 144 BTU/pound of ice.
The test was conducted as follows. Two large tupperware containers were filled with just under 5 pounds of ice each. Each was weighed and placed in the box. The ice was removed, melt water poured off, the remaining ice weighed, and put back. This was done three times, after a one hour interval, after a second two hour interval, and after a final three hour interval. The data is summarized in the following table.
| Ice Box Melt Test 2019-10-08 - ambient 59-60°F | ||||
| Time | Weights | Total | Comment | BTU/hr |
| 8:12 | 4lb 10oz + 4lb 10oz | 9lb 4oz | start with warm box | -- |
| 9:14 | 4lb 4oz + 4ob 7oz | 8lb 11oz | lost 9 oz in 1:02 | 78.3 |
| 11:10 | 3lb 12oz + 3lb 14oz | 7lb 10oz | lost 17 oz in 1:56 | 79.1 |
| 14:10 | 2lb 15oz + 3lb 3oz | 6lb 2oz | lost 24 oz in 3:00 | 72.0 |
The ambient temperature was 59-60°F during this test. Fast ice melt was expected on the first run due to initially cooling the box but it is also possible the ice was far enough below 32°F to cover that. The third run represents a longer time period without openning the box and so is expected to gain less heat. If we use 59 as the ambient, then an estimated 181 BTU/hr is expected at 100 degree ambient.
Selection of Refrigeration System
Choice of refrigeration system was decided on the quality and flexibility of systems with cost a secondary consideration. Some of the features desired include cold plates, water cooled compressor, high efficiency, and appropriate system size for this ice box. Alder Barbour no longer offers cold plates. Isotherm (now Indel Westabo) offers too few technical specs to make any sort of informed decision though they do offer very large cold plates. Sea Frost emerges as the best of the lot with multiple systems, an assortment of cold plates, detailed specifications, and outstanding documentations and technical support.
Sea Frost provides four lines of marine DC electric refrigeration units, all using cold plates. These are BD, BDXP, BDXPX, and DC 5000. The DC 5000 is for large systems. The BDXPX uses the R404a refrigerant. The BD and BDXP use R134a. R404a freezes at a lower temperature (-30°F) than R134a (-10°F) and so is sometimes preferred for freezers and large refrigerators, but is less efficient than using R134a. The EPA is phasing out R404a systems by first restricting its use in large systems.
The Sea Frost BD is their entry level DC electric refrigeration. The BDXP has 30% more capacity but also can automatically vary the compressor speed according to load to improve efficiency.
Worst Case Power Draw and Duty Cycle in Tropics
Sea Frost provides some graphs on power draw and BTU/hr at various RPM and graphs and information on air cooling and counter flow water cooling with pump which are unfortunately provided as low resolution images. The following table attempts to capture the power draw and BTU/hr information for the BDXP.
| RPM | A @ 12V | BTU/hr | BTU/W | duty cycle | ||
| excl pump | incl pump | @ 80°F | @ 100°F | |||
| 2500 | 4.6 | 230 | 4.2 | 3.4 | 56% | 79% |
| 3100 | 5.4 | 260 | 4.0 | 3.4 | 49% | 70% |
| 3800 | 6.8 | 340 | 4.2 | 3.6 | 38% | 53% |
| 4400 | 7.7 | 370 | 4.0 | 3.5 | 35% | 49% |
The above tests use American Society of Heating and Refrigeration Engineers (ASHRE) test points which require 120°F condensing temperature be used. Sea Frost points out that "In real world usage Sea Frost performance will exceed this rate because ducted fan and/or pumped water cooling lowers the condensing temperature below these test points." Due to the lower condensing temperature at lower RPM efficiency can be expected to improve with lower RPM.
The duty cycle figures rely on the estimated 181 BTU/hr at 100°F estimate produced by ice tests. The lower fan speed result in higher duty cycle but better efficiency and quieter operation. At 2500 RPM about 1.2 kWh per day will be used according to the ASHRE figures reported by Sea Frost.
Real World Power Requirements in Tropics
The ASHRE figures are worse than a real world power draw. Sea Frost estimates that coolant temperature can be maintained at 15°F above water temperature which even in summer in the tropics (during height of hurricane season) seldom exceed 80°F in all but the Gulf of Mexico. This is well below the 120°F in the reported results. The ambient temperature can also be expected to fall well below 100°F at night. Cruising in the tropics will likely occur in the late fall to spring when temperatures are lower and hurricanes are less likely. Power draw in higher lattitudes can be expected to be far less. Based on Sea Frost sizing guidance actual power draw can be expected to be 60-70% of the reported ASHRE figures.
Power Requirements and Power Generation
The ASHRE date worst case figures yield 1.2 kWh per day with real world figures expected to be less. Assuming 330W of solar panel (one completely shaded) 1.3-1.6 kWh per day would be generated in the tropics. Power will be needed for other uses and therefore this may be too high a percentage of power generation capacity. If neither panel were shaded or one is only partially shaded power generated could be 2-3 kWh per day but this would only occur at anchor if the boat were oriented (for example with bow and stern anchors) to maximize solar generation or during the few hours where the sun is almost directly overhead.
With well over 10 kWh of useable battery capacity available, the occasional overcast day or multiple days should not be a problem. If weeks go by where half the days are overcast or if using the electric auxilliary draws down the 48V battery bank, then even this large battery capacity could be depleted.
If the daily power draw in the tropics proves too high there are options. The size of the box can be reduced and insulation further improved with more insulation and optionally a liner. Alternately a wind generator can be added to increase daily power generation. If this is very rarely problem, an ice block can be placed in the box when too much refrigeration power draw does draw the battery banks down.
Sea Frost Pricing
Below are some Sea Frost pricing.
| Sample Sea Frost Pricing | |
| Price | Item |
| $2,200 | BDXP Air-Water Cooled System With Freezer Bin & Standard Thermostat |
| $2,400 | BDXP Air-Water Cooled System With 2-custom size Evaporator Plates |
| $350 | Electronic Thermostat Thermometer with Speed Control |
| $39 | Fan Widget (each) |
The freezer bin counts as one evaporator/cold plate but is recommended only for smaller ice boxes. It may be possible to use another evaporator/cold plate with a freezer bin for a larger box. A safe estimate for planned expenses is the the system will cost $2,400 and well under $600 of extras will be needed. Professional installation could drive the cost up by an unknown amount.