solar / refrigerator analysis

[gregtbrown]

Hey campers,

I've done a lot of reading about how well the stock refrigerator would perform on the stock solar setup and I came to the conclusion that it was basically useless for going off-grid. But, in the interests of science, I decided to do some investigation on my own. I wanted to share my results as I thought someone might find it interesting.

Camper:

• 2022 Imagine 23LDE
• Solar: stock Furrion 165W single panel model# FSFP16MW (not cleaned for a while :-O )
• Refrigerator: stock Furrion 10cuft model# FCR10DCGTA (fridge/freezer was empty, and the temperature set to halfway on the dial for this test, but not turned on until the test started)
• disconnected from shore power

Test Rig:

• ESP32 single board computer (powered from an external source - not the camper)
• 2 temperature sensors
• voltage sensor (I tapped into the 12v from the connection to the usb port right below the counter)
• SD card writer (wrote a sample to a CSV file once per minute)
• small amount of programming/code to lego it all together

Conditions: mostly sunny and about 65F. I realize this is not ideal for a test, but since it's not 80deg yet in PA and I don't usually camp in the heat of the summer, it seemed like a good starting point anyway.

Here's my hacked up test setup: you can see the sensor wires going into the fridge and freezer compartments and the small screen showing the readings.


Here's the results after a day and a half:


Observations:

• The refrigerator is clearly cooled by the freezer. You can see that the freezer dropped to zero in an hour but the fridge took about 2 1/2hrs to get to 40F.
• Initially, the unit is cycling about every 45 minutes, but as it gets dark and the solar drops out, it seems to know that the voltage isn't picking back up, so it slows down, presumably to conserve power. The longest cycle is close to 6hrs. I was a bit impressed that the unit was 'smarter' than expected.
• I wish I could've tracked when the compressor actually turned on and off, but there was no easy way to do that. It would be interesting from a nerd perspective, because there seems to be some type of predictive PID-type control going on.
• The reason I believe there is something more intelligent going on than a typical on/off thermostat: Assuming that I'm correct that the cooling unit is in the freezer, which in turn cools the fridge compartment, and the fact that the thermostat is in the fridge compartment, you can see that in the first few hours, the freezer temp starts to taper off as the fridge temp settles into range. Later, when the cycle is longer, the freezer starts working harder (lower temps) to keep the fridge in range. Maybe I'm wrong, but for me, this has all the signs of a predictive controller.
• By 8am, you can see the voltage start to creep back up from the solar.
• By mid-morning, the cycle starts picking back up
• The fridge temp spike on the right, around 5:30pm, is from opening the door for about 20 minutes. Since I was only periodically spot checking the test, I had no way to tell if the fridge was actually still working, or if it had shut down in the middle of the night and just stayed cold inside. I'd read in the manual that it will shut down instead of killing the battery, but it doesn't say if it's off for good or not.
• Eventually (somewhat surprisingly) it did kick on, and I closed the door and it went on it's merry way.
• After the test, the battery meter in the storage compartment said it had 75% life left and 76aH. It also showed it was drawing ~4A while the fridge was running (6A while connected to shore power)

I realize this is not a real world test. Other factors that would definitely affect the battery:

• warmer ambient temperature
• opening/closing the refrigerator door to get 'stuff' (although having stuff in the fridge should help hold the cold)
• running the water pump
• running the furnace blower
• LED lights, probably not that much

Mostly I was just interested in how it would perform, and 'how' it performs. Honestly, I expected by morning to see that the whole thing tanked, and I was a bit nervous about letting it run overnight.

At some point I might try another test, but in warmer weather and for a longer time period.


- Greg

[Cate&Rob]

Hi Greg, Interesting post! You are the first (as far as I know) to discover that the fridge control might be smarter than an on/off temperature switch.

Rob

[Country Campers]

There is a spot in the middle of the control dial that is for "dry or off grid" camping. A different setting than just general camping. In this setting the control allows the fridge to run less to conserve battery power. It is hard to tell exactly where this is on the dial as it is in a general area on the dial, labeled but not to a direct point.

Brian

[gregtbrown]

There is a spot in the middle of the control dial that is for "dry or off grid" camping. A different setting than just general camping. In this setting the control allows the fridge to run less to conserve battery power. It is hard to tell exactly where this is on the dial as it is in a general area on the dial, labeled but not to a direct point.

Brian you are correct. In my manual and on my dial it is clearly marked as the range between 'dot' 2 and 3 of 7, or roughly between 9 and 10 o'clock. I assumed it just meant "set it low and hope for the best".
However, for my test, the dial was set above that to 12 o'clock, the "Mid" position, or 4 of 7 dots.

[Country Campers]

That is where I set mine to also, runs great there and have not had any issues.

Brian

[COReflection]

That is where I set mine to also, runs great there and have not had any issues.

Brian

I know my 16 cu ft double door Furrion has an off grid setting too but I dont like the 40* temps in that mode. Too warm. I run mine down to 32* in the fridge and 10* in the freezer. Energy use for the lower setting is not all that different than the off grid boondocking setting.

[steveluannj]

Are you sure that the change in cycle frequency is not because the air around the fridge is colder at night? I have a hard time believing the fridge is that smart. It also seems like you have the freezer temp probe close to the edge- the probe temp varies a lot more than I would expect the freezer to vary, at least where the manufacturer put the probe.
It would also be interesting to see if the results were different with some thermal mass in the fridge and freezer- that should even out the swings.

[neilstanton]

Many thanks - great info and just what I needed! I have the same trailer XLS23LDE, kept on a storage lot close nearby. I was looking to see if we can load the frig the day before leaving and run it off solar and battery until hooking up to travel next day. I just ran an overnight test, frig not loaded, and got about the same result. Started frig at 10am yesterday on setting 5 of 7, battery meter showed 100% at 2pm with frig good and cold. At 630PM with sun lower, battery showed 96%, so I lowered setting to the "off-grid" range. Still good and cold at 9am today battery at 78%. Battery is stock interstate SRM 24 81AH flooded lead acid. It was bright and sunny all day with temps in low 70's.

[neilstanton]

Initially, the unit is cycling about every 45 minutes, but as it gets dark and the solar drops out, it seems to know that the voltage isn't picking back up, so it slows down, presumably to conserve power. The longest cycle is close to 6hrs. I was a bit impressed that the unit was 'smarter' than expected.

From the graph, it looks to me like the frig internal temp setpoint is driving the compressor cycles during the daylight, while at night the freezer internal temp setpoint is driving the compressor cycle. This is indicated by the peak or max temp of each compartment for each cycle. Probably the result of ambient temp around the frig as suggested by Steveluannj. Presuming frig and freezer both have temp sensors, whichever gets "warm" vs their setting first turns the compressor on. For example, if it's 40 deg outside the frig almost doesn't need the compressor, but the freezer still needs it to maintain 10 deg F.

Likely not very high level logic, but simple on/off control with select from both frig and freezer compartments so neither go above setpoint. Simple but reliable and should work fine for this application.

[AlexPeterson]

Are you sure that the change in cycle frequency is not because the air around the fridge is colder at night? I have a hard time believing the fridge is that smart. It also seems like you have the freezer temp probe close to the edge- the probe temp varies a lot more than I would expect the freezer to vary, at least where the manufacturer put the probe.
It would also be interesting to see if the results were different with some thermal mass in the fridge and freezer- that should even out the swings.

This is good info. I would expect those swings to be largely damped out with perhaps a half full freezer and frig. The mass of a few things in each is way more than the mass of the air. It certainly might have a PID controller to accommodate the large variations in mass. Something is deciding to lengthen the cycle times, and it is not frig temperatures.

I do believe the ambient temperature explains the longer delay between runs. If one looks at the slope of the temperature rises in the frig, it can be seen that at night, the temp rises more slowly (lower slope of the "up" line). Not much else besides lower ambient temperatures can explain that.

What is unexplained is the increase in amplitude of the freezer temperature swings.