Trailer boat - conversion of an custom ice chest to electric fridge
- Thread starter walt
- Start date
Walt, could you integrate a support into the Bimini frame by going back to the pushpit frame. I did that on my Hunter and was very happy with the function and look. The panel sat a couple of inches above the Bimini. It didn't seem to add much windage and in your case would lower the weight.
Bob
Bob
Thanks for the feedback and ideas! I normally dont even have the bimini on the boat during the winter season (Lake Havasu) and already have something in place for the trailering issue. With trailering, you will have a situation where the mast is all the way forward to the bow pulpit and another situation where the mast has been moved back to the base with the mast resting in the crutch and ready to be raised.
What I did was to make my crutch adjustable in height (keep in mind its still a prototype..). The panel will have nothing in the back to cause any partial shading - only the sail and boom which you cant do anything about. In the lowest crutch position, I just want the top of the crutch mostly below the panel so it doesnt shade. I have a very high position that I will use for raising the mast. The crutch needs to be high enough to that the mast clears the panel on the forward edge. I have a third position for trailering on the road. I dont trust that pin that allows for the adjustment to be robust enough for going down the road with the wieght of the mast on it so I still need to work something out. I like the crutch on a vertical pole as this is the lightest way to carry the load of the mast.
I did end up keeping all the panel mounting hardware to about 15 pounds total and I think its plenty strong. The two masts were shaped pine and then wrapped with carbon fiber cloth and glued with epoxy (free from a friend who was cleaning out his garage). The 160 watt panel weighs 26 pounds so I added a total of 41 pounds for the panel and mount.
What I did was to make my crutch adjustable in height (keep in mind its still a prototype..). The panel will have nothing in the back to cause any partial shading - only the sail and boom which you cant do anything about. In the lowest crutch position, I just want the top of the crutch mostly below the panel so it doesnt shade. I have a very high position that I will use for raising the mast. The crutch needs to be high enough to that the mast clears the panel on the forward edge. I have a third position for trailering on the road. I dont trust that pin that allows for the adjustment to be robust enough for going down the road with the wieght of the mast on it so I still need to work something out. I like the crutch on a vertical pole as this is the lightest way to carry the load of the mast.
I did end up keeping all the panel mounting hardware to about 15 pounds total and I think its plenty strong. The two masts were shaped pine and then wrapped with carbon fiber cloth and glued with epoxy (free from a friend who was cleaning out his garage). The 160 watt panel weighs 26 pounds so I added a total of 41 pounds for the panel and mount.
I have installed a Victron Blue Solar 75/15 MPPT charge controller and I guess now commited so have the Isotherm GE80
SX with Isec on order.
A suggestion was made early in this thread by jeepbluetj regarding using the solar charge controller load output and the Victron has an interesting mode for the load output.
------------------------------------------
From the manual..
When a solar charge controller is not able to recharge the battery to its full capacity within one day, the results is often that the battery will be continually cycled between a partially charged state and the end of discharge state (my note.. pretty sure this could easily happen with for the solar powered marina application where the fridge is on all the time). This mode of operation (no regular full recharge) will destroy a lead acid battery within weeks or months.
BatteryLife algorithm (a jumper setting for controller load output)
The BatteryLife algorithm will monitor the state of charge of the battery and day by day slightly increase the load disconnected level until absorption voltage is reached. From that point onwards the load disconnect level will be modulated so that absorption voltage is reached about once every week. The BatteryLIfe algorithm will substantially increase service life of the battery when compared to conventional fixed voltage load disconnect.
-------------------------------------
I also have that solid state relay on order but may initially wire things up using this BatteryLife load output.
SX with Isec on order.
A suggestion was made early in this thread by jeepbluetj regarding using the solar charge controller load output and the Victron has an interesting mode for the load output.
------------------------------------------
From the manual..
When a solar charge controller is not able to recharge the battery to its full capacity within one day, the results is often that the battery will be continually cycled between a partially charged state and the end of discharge state (my note.. pretty sure this could easily happen with for the solar powered marina application where the fridge is on all the time). This mode of operation (no regular full recharge) will destroy a lead acid battery within weeks or months.
BatteryLife algorithm (a jumper setting for controller load output)
The BatteryLife algorithm will monitor the state of charge of the battery and day by day slightly increase the load disconnected level until absorption voltage is reached. From that point onwards the load disconnect level will be modulated so that absorption voltage is reached about once every week. The BatteryLIfe algorithm will substantially increase service life of the battery when compared to conventional fixed voltage load disconnect.
-------------------------------------
I also have that solid state relay on order but may initially wire things up using this BatteryLife load output.
Sort of interesting.. I hooked up the panel today and had the mast crutch all the way as high as it would go as shown in one of the pictures above. There are also all sorts of lines etc going across the panel besides the shade of the mast crutch (same spot as in the pictures) and I was getting 3.8 amps into the battery. I then moved the crutch to its lowest position so it no longer contributes to the shade and the current jumped up to 5.7 amps. That is an increase of 50%! This is about 3 hours before sunset and about two months before the shortest day of the year.
Update.. I have been using the solar panel and fridge now for a string of days and have the fridge power turning off at dusk and back on at dawn. The solid state relay I had picked (details at the bottom) seems to work well for this application (details on request).
I have been posting pictures of the solar panel in the Macgregor owners area since that is specific to the boat but here is a link to a picture taken a couple days ago.
I have a Linklite battery monitor which I believe to be accurate. I can not tell how much power the fridge is using during the day because I am measuring current into the battery so what I see is a mix of solar and loads. However, I have been able to measure what the fridge uses at night as its the only load on the battery.
This data is all taken in Lake Havasu Az where the current days are mid 80's F and the nights are low 60'sF. The nights are about 13.5 hours.
Before I had the solid state relay set up, the fridge was using about 9 amp hours over night. I actually thought this was fairly good and likely due to the ISEC controller in the fridge.
I then hooked the solid state relay up and have had to work some bugs out of the control that only powers the fridge when the sun is shining but what happens is that at dusk, the fridge turns off. At dawn, the fridge turns back on and since the temp has drifted up during the night, it starts off running hard. The spec for this fridge says 2.5 amps but I have seen over 5 amp at startup. The fridge will run constantly for maybe an hour and the charge monitor measured an accumulated charge of around a mere 2.5 amp hours if the sun is up.
We had one very cloudy day and the battery got down to a max of 12 amp hours. At the end of the day even though it was cloudy, the panel had somehow fully charged the battery - which was surprising.
So on a even sort of sunny day and when Im not using the boat, the battery only goes down about 2.5 amp hours and then soon that is all put back in. For the rest of the day, the 160 watt panel easily runs the fridge and fully charges the battery back to float. This should be really easy on the batteries. The down side of this is that during the night, the fridge temp creeps up to about 48F. Ive had some milk in the fridge and its staying very fresh. Drinks get tapped into later in the day and of course are always nice and cold then.
Im not posting this for everyone (since most wont even read this far) but for even one person who finds this useful, the schematic is below (solid state relay details at the end). The original setup I used had some minor oscillation in the turn on and turn off right at dusk and dawn so I changed things to the two resistor setup which adds hysterisis. What happens is actually simple. At night the output of the solid state relay (- side) is off so is at zero volts. As soon as the sun comes up, the solar panel starts to output current through the 1.3K ohm resistor and when about 3 ma is reached, the solid state relay turns on and the output (- side) goes to 12 volts. This is fed back to the input through the 12K ohm resistor immediately pulling the input a little higher getting the input out of the noise trigger region. Turn off is just the opposite. Ive been watching the turn on and turn off (meaning Im at the boat often for dawn and dusk lately) and the little bit of positive feedback is doing a good job of snapping the power on and off to the fridge. When we are boat camping, I will turn on the bypass switch and the fridge will run full time.
Picture below - compressor. I have had to cut things up a little to get good circulation but it stays fairly cool now when things are running.
Overall.. it seems to work nicely. I can leave the fridge on all the time with no shore power and it seems to be very easy on the batteries only putting them through a very shallow discharge in the morning with lots of time with the batteries fully charged.
Was it worth it not to buy ice.. Not even remotely so.. But.. I have had fun.
Solid State relay
TOOGOO(R) SSR-100 DD Solid State Module Solid-state Relay DC-DC 100A 3-32V DC/5-60V DC
https://www.amazon.com/TOOGOO-SSR-100-Solid-Module-Solid-state/dp/B073XKF5BN/ref=sr_1_1?ie=UTF8&qid=1507645198&sr=8-1&keywords=SSR-100+DD+Solid+State+Module+Solid-state+Relay+DC-DC+100A+3-32V+DC/5-60V+DC
I have been posting pictures of the solar panel in the Macgregor owners area since that is specific to the boat but here is a link to a picture taken a couple days ago.
I have a Linklite battery monitor which I believe to be accurate. I can not tell how much power the fridge is using during the day because I am measuring current into the battery so what I see is a mix of solar and loads. However, I have been able to measure what the fridge uses at night as its the only load on the battery.
This data is all taken in Lake Havasu Az where the current days are mid 80's F and the nights are low 60'sF. The nights are about 13.5 hours.
Before I had the solid state relay set up, the fridge was using about 9 amp hours over night. I actually thought this was fairly good and likely due to the ISEC controller in the fridge.
I then hooked the solid state relay up and have had to work some bugs out of the control that only powers the fridge when the sun is shining but what happens is that at dusk, the fridge turns off. At dawn, the fridge turns back on and since the temp has drifted up during the night, it starts off running hard. The spec for this fridge says 2.5 amps but I have seen over 5 amp at startup. The fridge will run constantly for maybe an hour and the charge monitor measured an accumulated charge of around a mere 2.5 amp hours if the sun is up.
We had one very cloudy day and the battery got down to a max of 12 amp hours. At the end of the day even though it was cloudy, the panel had somehow fully charged the battery - which was surprising.
So on a even sort of sunny day and when Im not using the boat, the battery only goes down about 2.5 amp hours and then soon that is all put back in. For the rest of the day, the 160 watt panel easily runs the fridge and fully charges the battery back to float. This should be really easy on the batteries. The down side of this is that during the night, the fridge temp creeps up to about 48F. Ive had some milk in the fridge and its staying very fresh. Drinks get tapped into later in the day and of course are always nice and cold then.
Im not posting this for everyone (since most wont even read this far) but for even one person who finds this useful, the schematic is below (solid state relay details at the end). The original setup I used had some minor oscillation in the turn on and turn off right at dusk and dawn so I changed things to the two resistor setup which adds hysterisis. What happens is actually simple. At night the output of the solid state relay (- side) is off so is at zero volts. As soon as the sun comes up, the solar panel starts to output current through the 1.3K ohm resistor and when about 3 ma is reached, the solid state relay turns on and the output (- side) goes to 12 volts. This is fed back to the input through the 12K ohm resistor immediately pulling the input a little higher getting the input out of the noise trigger region. Turn off is just the opposite. Ive been watching the turn on and turn off (meaning Im at the boat often for dawn and dusk lately) and the little bit of positive feedback is doing a good job of snapping the power on and off to the fridge. When we are boat camping, I will turn on the bypass switch and the fridge will run full time.
Picture below - compressor. I have had to cut things up a little to get good circulation but it stays fairly cool now when things are running.
Overall.. it seems to work nicely. I can leave the fridge on all the time with no shore power and it seems to be very easy on the batteries only putting them through a very shallow discharge in the morning with lots of time with the batteries fully charged.
Was it worth it not to buy ice.. Not even remotely so.. But.. I have had fun.
Solid State relay
TOOGOO(R) SSR-100 DD Solid State Module Solid-state Relay DC-DC 100A 3-32V DC/5-60V DC
https://www.amazon.com/TOOGOO-SSR-100-Solid-Module-Solid-state/dp/B073XKF5BN/ref=sr_1_1?ie=UTF8&qid=1507645198&sr=8-1&keywords=SSR-100+DD+Solid+State+Module+Solid-state+Relay+DC-DC+100A+3-32V+DC/5-60V+DC
Last edited:
Nice install Walt. I think you will love it the next long trip you take.
We've been using the 63 quart Edgestar for quite a few years now in the Mac and I've tried a number of different tactics running the fridge. I use to run it down to about 37 during the day to take advantage of the panels like you are doing and then turn it off at night. Below 37 on the digital display and it would freeze some items at times. The fridge is normally packed full, especially at the beginning of a trip so there isn't good circulation within the box. Also with the box stuffed full there is a lot of thermal mass there to cool down. With it off all night it would climb at times to about 45 by morning. Then it would take a long time to cool all of that mass back down to 37 again.
If the box only has a few items like what you have shown it cools quickly. Also this box doesn't have the same size compressor as what you are using and what I'm using in the fridge that I built for the Endeavour...
http://purplesagetradingpost.com/sumner/endeavour-inside-mods/inside-index.html
.... so I don't think it can respond as quickly to cooling a large load.
It has a digital display with buttons to quickly change the temp setting so now I put it at 37 in the morning and about 42-43 before going to bed and let it kick on if it needs to during the night as I don't want the contents to get warmer than that.
We were on Lake Powell recently for two weeks and spent two nights in the location above where we only had about 2 hrs of sun on the boat as the rest of the time it was shaded by the canyon wall. I worried a little about that but the batteries never got anywhere near 50% so it was no problem. In the Bahamas I was in the first tropical storm of the year and had 3+ days of clouds and still had no problems with the batteries but have a lot of solar on the boat so even with the clouds was getting some input from the panels.
We've used the home-built fridge in the Endeavour for over 8 months while working on the boat in Florida and it also has a small freezer section and both are usually full. It cycles off and on day and night but never runs for very long compared to the Edgestar that cycles less frequently but runs longer when it comes on. The one in the Endeavour is almost twice the size of the portable in the Mac but uses quite a bit less electricity since it has probably over twice the insulation as the portable.
I'm considering getting the Isotherm controller for it but might wait to see how it and the batteries perform once we are on the water and only charging with the panels (480 watts). I'll put the four 6 volt batteries in when we return this time and run only on the panels even when in the yard.
I'll be interested in your updates when you take a trip and have the fridge full. Thanks for the info you've posted as it has been interesting,
Sumner
===================================================================
1300 miles to The Bahamas and Back in the Mac...
Endeavour 37 Mods...
MacGregor 26-S Mods...
Mac Trips to Utah, Idaho, Wyoming, Canada, Florida, Bahamas
We've been using the 63 quart Edgestar for quite a few years now in the Mac and I've tried a number of different tactics running the fridge. I use to run it down to about 37 during the day to take advantage of the panels like you are doing and then turn it off at night. Below 37 on the digital display and it would freeze some items at times. The fridge is normally packed full, especially at the beginning of a trip so there isn't good circulation within the box. Also with the box stuffed full there is a lot of thermal mass there to cool down. With it off all night it would climb at times to about 45 by morning. Then it would take a long time to cool all of that mass back down to 37 again.
If the box only has a few items like what you have shown it cools quickly. Also this box doesn't have the same size compressor as what you are using and what I'm using in the fridge that I built for the Endeavour...
http://purplesagetradingpost.com/sumner/endeavour-inside-mods/inside-index.html
.... so I don't think it can respond as quickly to cooling a large load.
It has a digital display with buttons to quickly change the temp setting so now I put it at 37 in the morning and about 42-43 before going to bed and let it kick on if it needs to during the night as I don't want the contents to get warmer than that.
We were on Lake Powell recently for two weeks and spent two nights in the location above where we only had about 2 hrs of sun on the boat as the rest of the time it was shaded by the canyon wall. I worried a little about that but the batteries never got anywhere near 50% so it was no problem. In the Bahamas I was in the first tropical storm of the year and had 3+ days of clouds and still had no problems with the batteries but have a lot of solar on the boat so even with the clouds was getting some input from the panels.
We've used the home-built fridge in the Endeavour for over 8 months while working on the boat in Florida and it also has a small freezer section and both are usually full. It cycles off and on day and night but never runs for very long compared to the Edgestar that cycles less frequently but runs longer when it comes on. The one in the Endeavour is almost twice the size of the portable in the Mac but uses quite a bit less electricity since it has probably over twice the insulation as the portable.
I'm considering getting the Isotherm controller for it but might wait to see how it and the batteries perform once we are on the water and only charging with the panels (480 watts). I'll put the four 6 volt batteries in when we return this time and run only on the panels even when in the yard.
I'll be interested in your updates when you take a trip and have the fridge full. Thanks for the info you've posted as it has been interesting,
Sumner
===================================================================
1300 miles to The Bahamas and Back in the Mac...
Endeavour 37 Mods...
MacGregor 26-S Mods...
Mac Trips to Utah, Idaho, Wyoming, Canada, Florida, Bahamas
Last edited:
Sumner, thanks for the data points.. The fridge is a big power load.. It really cuts the margins for both energy storage and generation that I have been used to. Solar power generation can vary a huge amount so you either need an aux power source, lots of solar panel and battery capacity or be willing to still buy ice once in a while. Sometimes you are probably working the batteries hard with fairly deep cycles. My experience with the 160 watt panel is limited but so far it seems to have a fair margin over what I need. But.. Im sure there will still be times where I would need to buy a bag of ice. 160 watts is about the largest size I could fit in that particular location sticking to a single panel and not interfering with the backstay.
Tested the DC power to the fridge shut off today at sunset. I took the picture below just moments before the solid state relay shut the power off to the fridge. I was watching the voltage to the input of the solid state relay. The spec says the relay needs 3 volts to turn on. As the sun went down, the solar output got lower and lower. When the relay input voltage got down to 3.4 volts, the output voltage to the fridge turned off and the hysteresis circuit kicked in immediately dropping the relay input voltage to 2.8 volts which is well below the threshold. At least that part of things seemed to work well.
Tested the DC power to the fridge shut off today at sunset. I took the picture below just moments before the solid state relay shut the power off to the fridge. I was watching the voltage to the input of the solid state relay. The spec says the relay needs 3 volts to turn on. As the sun went down, the solar output got lower and lower. When the relay input voltage got down to 3.4 volts, the output voltage to the fridge turned off and the hysteresis circuit kicked in immediately dropping the relay input voltage to 2.8 volts which is well below the threshold. At least that part of things seemed to work well.
Last edited:
Nice thread going here, folks. I was looking at that relay schematic, but I think my crew is into the icebox too much after dark to turn it off at sundown- which is 5 P.M. today. I do turn the fridge off when I hit the sack, and back on in the morning when I shut off the anchor light. My compressor is under the sink, next to the fridge, so I leave the cabinet door open to let the heat vent away from the fridge, and that seems to be a big help.
FYI, I forgot to mention that the solid state relay described above has the input circuit isolated from the output circuit. The input circuit is hooked only to the solar panel so this should work with any type of charge controller, even types that dont bias the solar negative to ground.
I had some specific applications in mind for this. When we take multi day trips where we are on the boat all the time, I probably will just leave the fridge on all the time, maybe dial back the temp at night like Sumner described. I think we will have plenty of solar for the fridge plus our other needs and the Isec does seem like it goes easy on the power at night - ie, automatically dialing back the temp a little instead of needing to do this manually. We are going to take a two night camping trip soon and I will also just leave the power on all the time for this.
The main reason for the auto turn off at night is that at the moment, Im thinking I can leave the fridge on at the marina for four month's. At this marina, I do not have reliable access to shore power. This is smack during the middle of winter (Arizona) so the nights are long but the night temps also drop. When I go down for an afternoon sail or just a bike ride to the marina, the chest and contents will be cold.. Four months of significant cycling would probably take some sort of toll on the batteries but with the auto shutoff, I think the batteries will hardly be involved. They would just see a few amp hour drain first thing in the morning and then spend the rest of the day fully charged.
At the moment I still would not leave the fridge on with the auto night shut off if Im away from the marina for any length of time. Power will get completely shut off then.
I had one more data point with the 160 watt panel. Took the boat out sailing for about five hours. Auto fridge circuit was on (fridge powered), I had a car stereo playing, tiller pilot, chart plotter, VHF radio. The battery monitor only sees what goes into or out of the battery so I dont know what excess power I had but I definitely had excess where the solar charge controller cut back as the battery stayed fully charged all day.
I had some specific applications in mind for this. When we take multi day trips where we are on the boat all the time, I probably will just leave the fridge on all the time, maybe dial back the temp at night like Sumner described. I think we will have plenty of solar for the fridge plus our other needs and the Isec does seem like it goes easy on the power at night - ie, automatically dialing back the temp a little instead of needing to do this manually. We are going to take a two night camping trip soon and I will also just leave the power on all the time for this.
The main reason for the auto turn off at night is that at the moment, Im thinking I can leave the fridge on at the marina for four month's. At this marina, I do not have reliable access to shore power. This is smack during the middle of winter (Arizona) so the nights are long but the night temps also drop. When I go down for an afternoon sail or just a bike ride to the marina, the chest and contents will be cold.. Four months of significant cycling would probably take some sort of toll on the batteries but with the auto shutoff, I think the batteries will hardly be involved. They would just see a few amp hour drain first thing in the morning and then spend the rest of the day fully charged.
At the moment I still would not leave the fridge on with the auto night shut off if Im away from the marina for any length of time. Power will get completely shut off then.
I had one more data point with the 160 watt panel. Took the boat out sailing for about five hours. Auto fridge circuit was on (fridge powered), I had a car stereo playing, tiller pilot, chart plotter, VHF radio. The battery monitor only sees what goes into or out of the battery so I dont know what excess power I had but I definitely had excess where the solar charge controller cut back as the battery stayed fully charged all day.
Last edited:
Thanks for the "Cold" beer, from this refrigerator, the last sunset I saw you. Go ahead and call me what you call the dog.
I have a feeling you aren't going to need to buy much ice with the 160 watt panel. We had 180 watts when we went to Florida the first time and ....
... I think we ran the 12 volt gen-set about 3-5 hours total in 7 1/2 weeks on the water and we sat a lot of those days so the outboard wasn't putting anything into the batteries. Our load was heavier, as besides the Edgestar, I used the CPAP machine every night and we had a lot of computer time also along with the other usual items.
Since I've added two more batteries and now have 560 watts total I've never run the gen-set again. I'm tempted to remove it but just as soon as I do that the charge controller or something else might fail
.I think that is a great idea for the use of the 'auto turn off'
,Sumner
===========================================================================
1300 miles to The Bahamas and Back in the Mac...
Endeavour 37 Mods...
MacGregor 26-S Mods...
Mac Trips to Utah, Idaho, Wyoming, Canada, Florida, Bahamas
I have no plans at all to do this because I think they are obnoxious.. but if I thought I needed additional aux power (just to avoid ever buying a bag of ice), it would be either the Honda 1000 or 2000 gensets and a 110VAC to 12 volt high current battery charger. You can charge standard batteries at .2C in bulk and for my two golf cart batteries, this is 40 amps. At 12 volts, that is around 500 watts and even with power factor loss, the 1000 watt Honda genset (110VAC into the charger) can supply this. Seems you were only getting a max of about 250 watts out of the lawn mower engine setup (er.. I have s) and its likely heavier with only very basic regulation.
Besides being able to best charge the batteries, you would also have 110VAC for other things like a microwave, air conditioner, coffee maker.. hair dryer, curling iron (???).
Not going to happen for me.. this boat is questionably large enough just for the 160 watt panel. But if I ever become a "prepper" (sounding more plausible all the time), that is what I would do.
Besides being able to best charge the batteries, you would also have 110VAC for other things like a microwave, air conditioner, coffee maker.. hair dryer, curling iron (???).
Not going to happen for me.. this boat is questionably large enough just for the 160 watt panel. But if I ever become a "prepper" (sounding more plausible all the time), that is what I would do.
Last edited:
It has a 60 amp car alternator with the Honda engine. I have a Honda 2000 and they are about the same weight and not much difference in sound. I wasn't suggesting this as an alternative just noting that I did have to use some backup with the 180 watts of solar that I originally had.
I never had to run it for more than about 20 minutes at at time (approximately 9-12 times in 7 1/2 weeks). It would put out about 25 amps (all the batteries would accept at their current charge state) and when that dropped to about 12 I'd shut it off and let the solar finish the charge on the batteries. It takes up less space than my Honda 2000 would if I tried to take it on the boat and everything on the boat is 12 volts.
I also don't like running a gen-set of any kind or the engine for charging, thus the addition of more solar on both boats. My needs are different than a lot of people as I like to be out for weeks/months at a time on anchor and in situations where there is no ice available,
Sumner
==============================================================
1300 miles to The Bahamas and Back in the Mac...
Endeavour 37 Mods...
MacGregor 26-S Mods...
Mac Trips to Utah, Idaho, Wyoming, Canada, Florida, Bahamas
The marina solar powered fridge has been on all the time for about three weeks now and would seem to be working "almost" perfect. I modified the night turn off a little (new diagram) to improve the "snap" in the power turn on and off and on a typical day (Lake Havasu Az in November - one month before the shortest day of the year, days 75 to 85F, nights 55 to 60F), the batteries end the day fully charged and in float mode. The fridge power is shut off at night and when the power comes on in the morning, the fridge immediately comes on. With a variety of junk in the ice chest (mostly cans and a milk jug), the battery charge drops to between 2 to 4 amp hours (1 to 2 percent of new capacity) before the 160 watt panel begins to make that up plus run the fridge. I think this setup has been very easy on the batteries.
During the night, the fridge temp will creep up into the mid to high 40's when the power is off. This seems to have pretty much no impact on how long the milk is lasting (still good for weeks) but what does happen is that if the fridge is not kept clean, that temp cycling results in more smell. I had a small milk spill (dont know how.. but the boat gets sailed during this time) which is how I noticed this issue. If the fridge is kept clean, it all works wonderfully and the drinks are always nice and cold.
A recent thread on SBO https://forums.sailboatowners.com/i...-pcm-to-the-icebox-refer.187101/#post-1416306 discussed phase change material for use in the fridge and this is very interesting to me for this solar powered on all the time fridge project.
http://store.puretemp.com/blockvesl/ This block material is the one interesting to me as it looks fairly durable and the size is about perfect for my chest (about 13 inch by 10 inch by 1.5 inch thick). It would sit up againts one wall of the ice chest and be easilly removable for cleaning. One of these freezes at 39F and that would seem to be a good temp to insure that it freezes during the day under normal fridge operation but would absorb energy at night holding things at 39 F for as long as possible. I will post the numbers if someone want to check but I calculated that the single PCM block at 39F stores the same amount of "cold" as 2.36 gallons of water changing by 4 degrees C (water is always a liquid).
I dont think the PCM would reduce the morning amp hour drain on the battery very much (which is currently only 1 to 2 percent of capacity - ie, very small) but I think it would reduce the night time temperature increase. These PCM blocks are only $20 and I will order one when I can stomach the $18 shipping and tax charge.
160 watt solar is still seeming like significant over kill for this application but that is a good thing as you want some margin. The worst I have seen so far is an outing with one somewhat cloudy day and one very cloudy day and a campout where the battery got down -25 amp hours. The next somewhat sunny day cleared that up and ran the fridge.
The circuit that I have been using for a while now is in the first image. Second image shows a aluminum guard around the cold plate to keep junk in the ice box from damaging anything.
Note, the resistor values are chosen for the solid state relay in post #26 in this thread.
During the night, the fridge temp will creep up into the mid to high 40's when the power is off. This seems to have pretty much no impact on how long the milk is lasting (still good for weeks) but what does happen is that if the fridge is not kept clean, that temp cycling results in more smell. I had a small milk spill (dont know how.. but the boat gets sailed during this time) which is how I noticed this issue. If the fridge is kept clean, it all works wonderfully and the drinks are always nice and cold.
A recent thread on SBO https://forums.sailboatowners.com/i...-pcm-to-the-icebox-refer.187101/#post-1416306 discussed phase change material for use in the fridge and this is very interesting to me for this solar powered on all the time fridge project.
http://store.puretemp.com/blockvesl/ This block material is the one interesting to me as it looks fairly durable and the size is about perfect for my chest (about 13 inch by 10 inch by 1.5 inch thick). It would sit up againts one wall of the ice chest and be easilly removable for cleaning. One of these freezes at 39F and that would seem to be a good temp to insure that it freezes during the day under normal fridge operation but would absorb energy at night holding things at 39 F for as long as possible. I will post the numbers if someone want to check but I calculated that the single PCM block at 39F stores the same amount of "cold" as 2.36 gallons of water changing by 4 degrees C (water is always a liquid).
I dont think the PCM would reduce the morning amp hour drain on the battery very much (which is currently only 1 to 2 percent of capacity - ie, very small) but I think it would reduce the night time temperature increase. These PCM blocks are only $20 and I will order one when I can stomach the $18 shipping and tax charge.
160 watt solar is still seeming like significant over kill for this application but that is a good thing as you want some margin. The worst I have seen so far is an outing with one somewhat cloudy day and one very cloudy day and a campout where the battery got down -25 amp hours. The next somewhat sunny day cleared that up and ran the fridge.
The circuit that I have been using for a while now is in the first image. Second image shows a aluminum guard around the cold plate to keep junk in the ice box from damaging anything.
Note, the resistor values are chosen for the solid state relay in post #26 in this thread.
Last edited:
Before I forget this data point..
On January 15, the high temp here (Lake Havasu) was 73F and the low temp was 50F. At sunset the day before, I put a blanket over the solar panel, switched the fridge control to always on and noted the amp hour reading (down a couple amp hours as I used some power while setting this all up). The fridge was the only load on the battery and the solar panel was disabled. Fridge temp was set to 40F.
I came back 24 hours later after the boat just sat in the marina and I was very surprised to find that the fridge had used just 9.5 amp hours. I believe this is an accurate measurement (Linklite battery monitor measuring in and out to the battery which I have checked for accuracy in the past).
This is a much lower energy use number than I have heard in the past. I barely remember some class I had about a million years ago but I believe the heat flow through an insulator is proportional to the temperature difference across the insulator. For my data point, I think it was simply because of the overall lower ambient temps. At the bottom of this post is what was inside the fridge for this data point. If the temp inside the fridge is held constant (as was the case here), the mass inside the fridge should really have little effect on the energy flow through the insulator walls. I think the main thing that lots of stuff in the fridge would do is limit internal air flow which actually should be a good thing. I dont think full or empty would have had much affect on how much power was used. It mainly the quality of the insulation and the temperature difference inside to outside.
I will have to repeat this same test later in the year (maybe several times) just to see how ambient temp affects things (which of course it will - but how much).
Finally.. one other sort of interesting thing. I have this solar charge controller Victron Blue Solar 75/15 MPPT charge controller and it has a feature that accounts for the charging algorithm being repeated on a daily basis. The length of time the controller will spend in absorption (14.4 volts) is dependent on the battery voltage the controller measures first thing in the morning (ie, just before the solar panel starts to produce power). If the battery voltage is above 12.6 volts, absorption only last 1 hour. If the battery voltage is between 12.2 to 12.6 volts, the absorption will last 2 hours. In my system where the fridge automatically shuts off at night, the fridge will turn on a fair amount of time before the charge controller turns on and when the fridge turns on, the load will drop the battery voltage to about 12.5 volts. So even though the batteries end the day fully charged (pretty much every day), the controller selects the 2 hour absorption time because of the fridge turning on.
I really dont think this matters at all.. but sort of an interesting artifact of having the auto fridge turn on and off with the solar panel input.
The fridge has been on constantly (except for one week where we were gone) for around 11 weeks now. I think its overall been very easy on the batteries and at least for this particular application, the 160 watt solar is significantly more than I actually need. But.. having margin is not a bad thing.
On January 15, the high temp here (Lake Havasu) was 73F and the low temp was 50F. At sunset the day before, I put a blanket over the solar panel, switched the fridge control to always on and noted the amp hour reading (down a couple amp hours as I used some power while setting this all up). The fridge was the only load on the battery and the solar panel was disabled. Fridge temp was set to 40F.
I came back 24 hours later after the boat just sat in the marina and I was very surprised to find that the fridge had used just 9.5 amp hours. I believe this is an accurate measurement (Linklite battery monitor measuring in and out to the battery which I have checked for accuracy in the past).
This is a much lower energy use number than I have heard in the past. I barely remember some class I had about a million years ago but I believe the heat flow through an insulator is proportional to the temperature difference across the insulator. For my data point, I think it was simply because of the overall lower ambient temps. At the bottom of this post is what was inside the fridge for this data point. If the temp inside the fridge is held constant (as was the case here), the mass inside the fridge should really have little effect on the energy flow through the insulator walls. I think the main thing that lots of stuff in the fridge would do is limit internal air flow which actually should be a good thing. I dont think full or empty would have had much affect on how much power was used. It mainly the quality of the insulation and the temperature difference inside to outside.
I will have to repeat this same test later in the year (maybe several times) just to see how ambient temp affects things (which of course it will - but how much).
Finally.. one other sort of interesting thing. I have this solar charge controller Victron Blue Solar 75/15 MPPT charge controller and it has a feature that accounts for the charging algorithm being repeated on a daily basis. The length of time the controller will spend in absorption (14.4 volts) is dependent on the battery voltage the controller measures first thing in the morning (ie, just before the solar panel starts to produce power). If the battery voltage is above 12.6 volts, absorption only last 1 hour. If the battery voltage is between 12.2 to 12.6 volts, the absorption will last 2 hours. In my system where the fridge automatically shuts off at night, the fridge will turn on a fair amount of time before the charge controller turns on and when the fridge turns on, the load will drop the battery voltage to about 12.5 volts. So even though the batteries end the day fully charged (pretty much every day), the controller selects the 2 hour absorption time because of the fridge turning on.
I really dont think this matters at all.. but sort of an interesting artifact of having the auto fridge turn on and off with the solar panel input.
The fridge has been on constantly (except for one week where we were gone) for around 11 weeks now. I think its overall been very easy on the batteries and at least for this particular application, the 160 watt solar is significantly more than I actually need. But.. having margin is not a bad thing.
Last edited:
I took another data point on the power consumption of this electric ice chest now that it has warmed up considerably here. Remember back in mid January (Lake Havasu Arizona), I had measured the 24 hours consumption and the fridge only used 9.5 amp hours.
I just did the same test again April 3 here in Lake Havasu during an unusually warm spell which had 24 hour low temp of 63F and a high temp of 91F and of course significantly longer days compared to nights. Fridge temp was set to 40F and contents was a gallon of milk and six pack of can drinks. Batteries were fully charged and at 5:30 PM, I removed all charging sources and the only load was the fridge. A Xantrex linklite battery monitor kept track of amp hours. I disabled the night time shut off circuit so the fridge ran the entire 24 hour period.
For this data point, the amp hours used in 24 hours was 28.1 amp hours.
In theory (look up R value definition to verify if you wish), the energy loss is proportional to the difference in temp across the insulation. Since the fridge inside temp was set to 40F, you would expect to double the energy loss going from an average external temp of 60F (delta = 20F) to 80F (delta = 40F). And you would expect the energy input to the fridge to be somewhat proportional to the energy loss. So.. not too surprising that the power used went way up on a fairly warm day.
Solar output varies a huge amount with all of these variables: time of year, latitude (insolation), cloud cover, temperature, flat or angled, type of controller, etc. The only solar predictor that I somewhat like is a unitless equation:
Solar panel rated power * N = amp hours/per 24 hour period
I have measured N at about .31 once and generally just go by that N varies from maybe .2 to .5.
For my particular case where I am powering this fridge with a 160 watt solar panel and for the amp hours I just measured, the mininum N that I need is
28.1 amphours / 160 watt rated panel = .175 This is under my assumed range of .2 to .5 so I think I have a good chance of the fridge being sustained by the batteries even on a fairly warm day and with the solar output on the lower end because of conditions. I will find out later this summer but expect to still have excess power. The only bad thing is that the closer you are to barely having enough generated power is that it works the batteries harder so they dont last as long.
FYI, at the bottom of the cold side is a lexan platform that has standoff that are just PVC pipe rings glued on with contact cement.
The compressor side has a lot of ventilation. Top and bottom louvered vents on the front, the whole end is open with a screen mesh and the top cover has vent holes on the back and end sides.
Aluminum guard keeps items in the fridge from damaging the cold plate while sailing.. Stuff definitely gets thrown around (beer bottle projectiles). Temp measuring detector shown in the lower right side of the pic.
I just did the same test again April 3 here in Lake Havasu during an unusually warm spell which had 24 hour low temp of 63F and a high temp of 91F and of course significantly longer days compared to nights. Fridge temp was set to 40F and contents was a gallon of milk and six pack of can drinks. Batteries were fully charged and at 5:30 PM, I removed all charging sources and the only load was the fridge. A Xantrex linklite battery monitor kept track of amp hours. I disabled the night time shut off circuit so the fridge ran the entire 24 hour period.
For this data point, the amp hours used in 24 hours was 28.1 amp hours.
In theory (look up R value definition to verify if you wish), the energy loss is proportional to the difference in temp across the insulation. Since the fridge inside temp was set to 40F, you would expect to double the energy loss going from an average external temp of 60F (delta = 20F) to 80F (delta = 40F). And you would expect the energy input to the fridge to be somewhat proportional to the energy loss. So.. not too surprising that the power used went way up on a fairly warm day.
Solar output varies a huge amount with all of these variables: time of year, latitude (insolation), cloud cover, temperature, flat or angled, type of controller, etc. The only solar predictor that I somewhat like is a unitless equation:
Solar panel rated power * N = amp hours/per 24 hour period
I have measured N at about .31 once and generally just go by that N varies from maybe .2 to .5.
For my particular case where I am powering this fridge with a 160 watt solar panel and for the amp hours I just measured, the mininum N that I need is
28.1 amphours / 160 watt rated panel = .175 This is under my assumed range of .2 to .5 so I think I have a good chance of the fridge being sustained by the batteries even on a fairly warm day and with the solar output on the lower end because of conditions. I will find out later this summer but expect to still have excess power. The only bad thing is that the closer you are to barely having enough generated power is that it works the batteries harder so they dont last as long.
FYI, at the bottom of the cold side is a lexan platform that has standoff that are just PVC pipe rings glued on with contact cement.
The compressor side has a lot of ventilation. Top and bottom louvered vents on the front, the whole end is open with a screen mesh and the top cover has vent holes on the back and end sides.
Aluminum guard keeps items in the fridge from damaging the cold plate while sailing.. Stuff definitely gets thrown around (beer bottle projectiles). Temp measuring detector shown in the lower right side of the pic.
Last edited:
Forgot to add that the fridge ran for just over 18 weeks almost continuously this last winter powered only from a 160 watt solar panel in a slip (short period where it was off in Jan while out of town). Power shut off at sunset, turned back on at sunrise. I think it all worked just fine and it was no doubt nice, wife approved. Maybe drank a little too much beer because it was so nice and had to stock mostly Coors to keep the consumption down.
Cooling beer..
At the beginning of this project I was trying to size a solar panel for the electric fridge and had been watching forum postings to try and get an idea. Numbers varied all over the place as it depends on volume, insulation and ambient temp. I tried to measure what my fridge did and to summarize, my fridge is set at about 40F (mostly, it has the ISEC conroller) and on a day in Jan (this is in Arizona) with a varition of 50F to 75F and longer nights, I only used 9 amp hours in a 24 hour period. But a few months later when the days were getting longer and the temp ranged from 63F to 91F, the fridge used 3X the energy at 28.1 amp hours.
So.. there are two things that require power in a fridge ignoring cool lost by opening and closing the door. One is just the heat loss through the insulation and the other is cooling stuff added to the fridge.
The main thing that I ever added was naturally beer. So I calculated approximately how many amp hours would be required if a 16 oz beer at ambient was added to an ice chest. The calculation (with a couple assumptions) says it takes around 1 amp hour to cool one 16 oz beer from 70F to 40F.
Here is the calc.
Assume that beer has the same specific heat as water in the liquid phase at 4.19 KJ/KgC
Q (energy in KJ) = specific heat * mass in Kg * delta temp Celsius
Mass of a 16 oz beer is .4723 Kg
Temperature change from 70F to 40F is delta 16.67 C
Throw all that into the equation gives 32.99 KJ required to cool the 16oz beer.
Convert KJ to Watt hours , it takes 9.137 watt hours to cool the beer.
At 12 volts, that is .7614 amp hours
I assumed a fridge cooling inefficiency is 1.5X and that is how I arrived a 1.14 amp hours.
To round things off a little.. it takes about 1 amp hour to cool a 16 oz beer from 70F to 40F
At the beginning of this project I was trying to size a solar panel for the electric fridge and had been watching forum postings to try and get an idea. Numbers varied all over the place as it depends on volume, insulation and ambient temp. I tried to measure what my fridge did and to summarize, my fridge is set at about 40F (mostly, it has the ISEC conroller) and on a day in Jan (this is in Arizona) with a varition of 50F to 75F and longer nights, I only used 9 amp hours in a 24 hour period. But a few months later when the days were getting longer and the temp ranged from 63F to 91F, the fridge used 3X the energy at 28.1 amp hours.
So.. there are two things that require power in a fridge ignoring cool lost by opening and closing the door. One is just the heat loss through the insulation and the other is cooling stuff added to the fridge.
The main thing that I ever added was naturally beer. So I calculated approximately how many amp hours would be required if a 16 oz beer at ambient was added to an ice chest. The calculation (with a couple assumptions) says it takes around 1 amp hour to cool one 16 oz beer from 70F to 40F.
Here is the calc.
Assume that beer has the same specific heat as water in the liquid phase at 4.19 KJ/KgC
Q (energy in KJ) = specific heat * mass in Kg * delta temp Celsius
Mass of a 16 oz beer is .4723 Kg
Temperature change from 70F to 40F is delta 16.67 C
Throw all that into the equation gives 32.99 KJ required to cool the 16oz beer.
Convert KJ to Watt hours , it takes 9.137 watt hours to cool the beer.
At 12 volts, that is .7614 amp hours
I assumed a fridge cooling inefficiency is 1.5X and that is how I arrived a 1.14 amp hours.
To round things off a little.. it takes about 1 amp hour to cool a 16 oz beer from 70F to 40F
Last edited: