DIY: How to make your own 12V Fridge

[ViaVacavi]

I'm breaking my current "pseudo 12 volt fridge" discussion to this new thread from my build log ( ViaVacavi Build Log )

The purpose of this thread is to discuss DIY fridge conversions, essentially converting inexpensive and readily available mini refrigerators designed for 120 VAC use, to run off of our 12 VDC electrical systems as efficiently as possible. I would also like to gather information on energy use of popular 12 volt fridges for a comparison.  I will detail cost breakdowns, energy use and efficiency and details to do your own conversion.

Why would someone want to convert a 120 volt fridge to run on 12 volt?

• Cost.  Using all new high end parts, a converted fridge can be made for around 1/4 the cost of a similarly sized 12 volt fridge.  Using used, free and cheaper parts, a converted fridge can easily be made for less than a tenth the cost of a true 12 volt fridge of similar size.
• Layout.  Some people prefer a vertical layout vs. the standard "cooler" layout of most 12 volt fridges.
• Size.  Many mini fridges are available in the 3 to 4 cubic foot range - this is equivalent to a 90 to 120 quart 12 volt fridge in size.
• Separate Freezer.  There are many mini fridges with separate freezer compartments.  Only a few 12 volt refrigerators have this option and they are at the high end of the price range.

To be fair, there are also some drawbacks:

• A "pseudo conversion" will nearly always be at least a little less efficient than a true 12 volt fridge.
• There is a bit of added complexity.  There will be added components which could fail
• You must be at least a little familiar with wiring
• Nobody really knows anything about longevity or reliability, nor how well a converted fridge will stand up to life on the road.

Stay tuned!

 

[ViaVacavi]

Why not just get an inverter, and run your 120 volt fridge off of it from your battery bank?

This certainly will work, and many people have done it.  However, it is the least efficient option.
Inverters are not 100% efficient, so power will be lost during the conversion from 12 volts DC to 120 volts AC.  This power loss is present whether the compressor in the fridge is running or not.  On most fridges, the compressor only has to run roughly 25% of the time (or even less), so it makes no sense to suffer this power loss 24 hours a day.  On a positive note, no special skills are needed and you just plug the refrigerator into the inverter at one end, and plug the inverter to a 12 volt socket at the other end.

Here is an example of the power needs for such a setup. 
This example will assume a fridge that uses 60 watts while running, where the compressor runs 25% of the time.  This will also assume that the inverter consumes 0.5 amps just being plugged in. (Of course in the real world, there are many factors that will influence efficiency)

• Running 6 hours per day, the compressor will consume 30 amp-hours from your battery bank (60 watts divided by 12 volts = 5 amps, multiplied by 6 hours equals 30 amp-hours)
• Running 24 hours per day, the inverter will consume 12 amp-hours from your battery bank (0.5 amps multiplied by 24 hours equals 12 amp-hours)
• This adds up to 42 amp-hours per day to run the fridge

 

[ViaVacavi]

Conversion Plan

I still like to call this a "pseudo-12 volt conversion", as it will still use the 12o volt compressor of the refrigerator.
The plan is to convert the control system to 12 volts, and use that to turn an inverter on and off ONLY when the refrigerator calls for the compressor to run.  To explain this a little better, I've made a couple simple diagrams of a standard mini fridge

[img=494x640]http://i827.photobucket.com/albums/zz195/julieswiger/ViaVacavi/120v fridge_zpsiqg5ljc7.jpg[/img]
Above is a typical diagram of a simple mini fridge.  As you can see, the thermostat of the fridge is a simple switch mechanism that closes to turn on the 120 volt compressor whenever the internal temperature goes above a certain value.  Most refrigerators have a "coldness" dial that allows this temperature value to be changed.


[img=495x640]http://i827.photobucket.com/albums/...avi/12v fridge_zpsmuztnprr.jpg~original[/img]
Above is the proposed conversion, which I will complete on my own mini-fridge.  While I changed the color of the wires, you should notice that the main control loop is pretty much the same, except that it is now being run on 12 volts DC.  Since the thermostat is a simple switch, it really doesn't care whether the electricity is AC or DC - it will simply open or close as designed based on temperature.
When the thermostat closes, it will energize a simple relay, which will send 12 volts to a small and efficient inverter, which will then run the compressor.  Aside from that, the light bulb may also need changed to a 12 volt bulb, but most incandescent bulbs work fine on DC. Another thing to point out is that the switching mechanism in the thermostat has a limited current capacity, and in the case of my mini-fridge handles about 0.5 amps.  Since this will equate to about 5 amps now on a DC circuit if we were to try to use the thermostat switch to directly power the inverter, we use a simple automotive style relay to do the switching to avoid damaging the thermostat.

The advantage to this conversion over just plugging the fridge into an inverter, is that the parasitic losses from the inverter will ONLY be present when the compressor is actually running, which should be 6 hours or less per day.

For an example (and still theoretical), now consider the following:

• Running 6 hours per day, the compressor will consume 30 amp-hours from your battery bank (60 watts divided by 12 volts = 5 amps, multiplied by 6 hours equals 30 amp-hours)
• Running only 6 hours per day now, the inverter will consume 3 amp-hours from your battery bank (0.5 amps multiplied by 6 hours equals 3 amp-hours)
• This adds up to 33 amp-hours per day to run the fridge.  This is nearly 30% more efficient than just plugging the fridge into an inverter.  It is also getting very close to numbers reported by users of true 12 volt fridges of the same size.

 

[ViaVacavi]

Starting Measurements Before Conversion

We've had our mini-fridge for about a week and a half now, and it has performed flawlessly on the 120 volt shore power available at the campground.  We have been monitoring power usage using a kill a watt meter to get a starting point so we can compare energy usage after the conversion.

Here are some facts about this fridge:

• This is a Magic Chef 3.1 CuFt. Refrigerator/Freezer combo
• The unit draws exactly 60 watts while the compressor is running as reported by the kill a watt meter
• The compressor currently runs anywhere from 9 to 15 minutes per hour depending on ambient temperature

[img=640x480]http://i827.photobucket.com/albums/...hoto-359157907_zpscbmgbayr.jpg~original[/img]
Power usage has been very consistent day to day.  From the above picture, you can see that the power usage had been measured for 81 hours and 15 minutes (81.25 hours in decimal)

[img=640x480]http://i827.photobucket.com/albums/...3_kindlephoto-359250344_zpsfvbswfcg.jpg[/img]
During this 81.25 hours, the fridge consumed 1320 watt-hours of power.
1320/81.25=~16.25 watt-hours used each hour. 
16.25 * 24 = 390 watt-hours used per day
If this were on a 12 volt system at 100% efficiency, that would equate to about 32.5 amp-hours per day usage out of a 12 volt battery bank, which wouldn't bad at all for a refrigerator this size.

 

[ViaVacavi]

For this test conversion, I picked the following inverter:
https://www.amazon.com/gp/product/B00IXOWPQM/ref=oh_aui_detailpage_o04_s00?ie=UTF8&psc=1

(note to admin: please feel free to replace this link with Bob's own affiliate link and delete this text)
The inverter cost me $101, and is rated at 120 watts. It is a Samlex PST-120-12 and is a pure sine wave inverter.  Please note that a much cheaper modified sine wave inverter could be used, but I want this to be a best-case scenario test.

Here are the reasons why I chose this inverter:

• Very high quality and reputable manufacturer
• 2 year warranty
• Nice and small size
• Very efficient
• Pure sine wave
• Power rating closely matches what is needed for the fridge.
• Has high quality Anderson power pole connectors

[img=640x480]http://i827.photobucket.com/albums/...avi/IMG_20160615_131358_zps1h4knek6.jpg[/img]
Here is the inverter right after being unboxed.  I placed a pen near it for size comparison.

[img=640x480]http://i827.photobucket.com/albums/...avi/IMG_20160615_131413_zpsthjm2j8q.jpg[/img]
The power plug is very high quality, with the power pole connectors on one side and a standard 12 volt plug on the other.  For permanent installation, it will be easy to cut the 12v plug off of the end and hard wire to a 12 volt source.


Tomorrow I will do the actual conversion and take pictures during the process.

 

[ViaVacavi]

Have you tried starting an inverter with a motor connected directly like that? Might need a time delay relay between them to let the inverter charge up before applying the load. I have no experience with this though.

Guy

No I have not, but I already have a time delay relay should I need it.  Part of the testing is going to be how the inverter handles starting up loaded, and use the time-delay relay if necessary.  Here is a link to the time-delay relay I will be using if it is determined that it's necessary:
https://www.amazon.com/gp/product/B0126WT5QO/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

 

[ahh_me2]

I plugged in my kill a watt meter onto my small 120 volt fridge today, and I will run it for 4 days or 96 hrs.
I will then put that same meter on to my Engel 12 volt compressor fridge for the same amount of time, as it is currently hooked up to 120 volts.
Then I will run the Engel off of 12 volts, and use one of my RC watt meters for the 96 hours and get the results based on 12 volt input.

I'll report back when done.

 

[Matlock]

Subscribed.......

 

[ViaVacavi]

I plugged in my kill a watt meter onto my small 120 volt fridge today, and I will run it for 4 days or 96 hrs.
I will then put that same meter on to my Engel 12 volt compressor fridge for the same amount of time, as it is currently hooked up to 120 volts.
Then I will run the Engel off of 12 volts, and use one of my RC watt meters for the 96 hours and get the results based on 12 volt input.

I'll report back when done.

Thanks!

What are the capacities of the two fridges?

I'm thinking we need a metric that takes capacity into consideration, as a smaller refrigerator is understandably going to use less power - both due to a smaller compressor and less surface area to exchange heat with the surroundings. How about watt-hours/day/CuFt? That will allow reasonable comparisons between different capacity refrigerators. Thoughts?

 

[jimindenver]

My thoughts

It takes some gear to hook up a compressor and such. Gauges, vacuum pump, brazing torch, and that's it you can just switch out the Compressor. Got the gear and knowledge and it might be a fun project for a donor mini fridge. I've seen 12v compressor kits for around $200 at some point.

My second thought is that by the time you get to the size 3.2 ft two door mini fridge, the 12v's can pull some power too. The Grape 12v, 5 ft upright is a pig. For a bit more than most two door minis the Edgestar 3.2 mini pulls .64a at 120v. Yes you need a inverter but for $200 more my Tripp-lite industrial inverter will run a lot more than the fridge and shuts itself off when not needed. That's the fridge and inverter for much less than a bigger 12v unit.

You will need to add insulation as the 12v's are better insulated. Even at that size they will be more efficient just for being a chest unit and may still use less power. The thing is by the time you have a system that can run a fridge of that size plus the rest of your van, the difference between the fridges and even the inverter losses are not going to matter. It should take days for the difference to make a noticeable impact on a bank big enough to dependably run either fridge.

 

[ViaVacavi]

My thoughts

It takes some gear to hook up a compressor and such. Gauges, vacuum pump, brazing torch, and that's it you can just switch out the Compressor. Got the gear and knowledge and it might be a fun project for a donor mini fridge. I've seen 12v compressor kits for around $200 at some point.

I've thought about this approach, but it isn't DIY friendly for 99% of people. I work with an HVAC tech that could evacuate the system, but I'm aiming more for a reasonably easy DIY solution. Definitely would be fun though

My second thought is that by the time you get to the size 3.2 ft two door mini fridge, the 12v's can pull some power too. The Grape 12v, 5 ft upright is a pig. For a bit more than most two door minis the Edgestar 3.2 mini pulls .64a at 120v. Yes you need a inverter but for $200 more my Tripp-lite industrial inverter will run a lot more than the fridge and shuts itself off when not needed. That's the fridge and inverter for much less than a bigger 12v unit.

I agree. I think that we are going to find that there really isn't a huge difference in efficiency between 12v and 120v fridges, with insulation probably being one of the bigger determining factors. The idea behind the dedicated inverter for the fridge is so that parasitic losses can be cut 100% when the compressor isn't running.

You will need to add insulation as the 12v's are better insulated. Even at that size they will be more efficient just for being a chest unit and may still use less power. The thing is by the time you have a system that can run a fridge of that size plus the rest of your van, the difference between the fridges and even the inverter losses are not going to matter. It should take days for the difference to make a noticeable impact on a bank big enough to dependably run either fridge.

Insulation is planned, but I want to do it after the first round of tests so that a comparison can be made before/after insulation
I agree, that the difference is going to be pretty minimal on any well designed 12 volt system - which just lends credence to the idea that doing a conversion like this can save a substantial amount of money over buying a 12 volt compressor fridge of similar size. I could easily put together a conversion like I'm proposing in this thread for under $200 with all new parts, and about half that if I shop around and use a cheaper inverter. To get a true 12 volt refrigerator of the same capacity easily goes into the $800-$1000 range. I'm hoping that I can publish actual figures so that others can judge for themselves which option may fit their needs.

 

[jimindenver]

Were I to convert anything, it would be a mini freezer. They are already better insulated. Also with the mini fridges you have to look for the draw. The same style two door pulls .8a if Galance and up to 1.5a with other brands. The edgestar is alone at .64a.

 

[ViaVacavi]

Were I to convert anything, it would be a mini freezer. They are already better insulated. Also with the mini fridges you have to look for the draw. The same style two door pulls .8a if Galance and up to 1.5a with other brands. The edgestar is alone at .64a.

The MagicChef 3.1 CuFt Refrigerator/Freezer combo I have draws between 0.48 and 0.51 amps @ 120VAC while running.  I was actually surprised with how low the draw was.  I'm sure there is a lot of variation between brands and depending on the age of the unit

 

[jimindenver]

That's great. Do you have a model number and what the tag on the back says the draw is?

 

[ccbreder]

I agree that brand of inverter is good quality. Did that small inverter handle the in-surge of current to start the compressor?

 

[ZoNiE]

I'm keeping an eye on this one too. I have a friend who did something similar, but converted an oddball Norcold AC/DC fridge which had an AC "swing motor" compressor and 1970's era inverter. It worked well, but was an old unit. He removed the old inverter and wired in a current tech, albiet MSW inverter.

I really want to be able to swap out my gas fridge and just leave it on all the time so I can store my rig closer to where I use it. I have 390W solar and 225Ah of Battery. The prices for units like the Norcold DE0061 are just too high. I could do the compressor swap on a low cost magic chef, but then I have an oddball again. I'd rather just do the electrical conversion.

Thanks for your very detailed Posts. I like your paper cabinets.

 

[Telcobilly]

Subscribed

 

[ViaVacavi]

Sorry - I've been away for a bit and been super busy with other areas of my build but wanted to update this post.

Over the last week I've done a ton of experiments.  Here are my current findings:

I have 3 inverters:

1. 120 watt Samlex pure sine wave
2. 300 watt Bestek modfied sine wave
3. 1000 watt Bestek modified sine wave (this wil be the "main" power inverter for our build for a simple 120v system on the road

I also got a dedicated 12v power meter to run some tests.
[img=640x480]http://i827.photobucket.com/albums/...avi/IMG_20160621_162235_zpssbwjrbco.jpg[/img]
The 120 watt Samlex draws 2.9 watts in standby state. Sadly, it won't even think about starting the fridge compressor.

[img=640x480]http://i827.photobucket.com/albums/...avi/IMG_20160621_162344_zpswwq4mfmb.jpg[/img]
The 300 watt Bestek uses 3.5 watts in the standby state.  It came close, but can't quite start the compressor.

[img=640x480]http://i827.photobucket.com/albums/...avi/IMG_20160621_162751_zpsojqwcwyr.jpg[/img]
The Bestek 1000 watt uses 8.2 watts in its standby state.  It very easily starts the compressor.


From the above, it's easy to see that the larger the inverter, the more power it will take to even have powered on.  There are also two facts that are readily apparent from my tests:

1. It takes a large inverter to start the compressor
2. Once the compresor is running, it takes very little power to sustain running of the motor (about 60 wats, or about 5 amps on a 12 volt system)

So here is the plan moving forward.  My next test will be using the 1000 watt inverter to start the compressor, and then switching over to a small, more efficient inverter to sustain runing fo the compresor.  Since my build is already going to use the 1000 watt inverter, this isn't realy an added component.  I will just have a remote way to trigger the inverter to come on just long enough to start the compresor, then switch to a small inverter to continue runing the compressor once it's started to keep power loss at a minimum.

Of course there is a balancing act here:
- Having both inverters live to the same line at the same time would be bad - at least one would most definitely blow, especially if the phase angles of each were 180 degrees apart.
- Too long of a pause during switching will allow the compresor motor to stop, rendering the entire technique useless
- Time will tell how much the compressor motor will like being switched live between two sources that will likely be at different phase angles.  In the wind industry, we do this all the time to coax an induction turbine in sync with the grid with no issues - but it's hard to say how such a small motor will react.  That surely won't stop me from experimenting to find out though!

This week I will build the switching mechanism and give it a try.  I'll report back with findings, good or bad.

 

[SternWake]

Thanks for the details. I love it when people have measuring tools and use them and share their observations.

Dig the DIY experimental mindset too.

If you could gather some data on how much the fridge consumes over a time period would be a good figure to have. Such as mine( vitrifrigo c51is) runs usually no more than 20 minutes per hour, UNless of course i am opening the door a lot or just put a bunch of warm stuff inside. When full and when empty, the duty cycle is much different too.

I put one of these on the DC input to measure WH and AH consumed:

https://www.amazon.com/GT-Power-Ana...8&qid=1466702297&sr=8-1&keywords=gt+power+130

Evacuating heat from the condenser is where some of the biggest gains in efficiency are to be had. My Danfoss/secop has a radiator style condenser with an upgraded 120MM computer fan attached( Noctua NF-f12). Rather than just let this fan circulate air it itself has heated, I made mine pull cool filtered air from the floor below and the air pushed ( not pulled) through the condenser has no chance of being recycled through condenser. That and extra insulation and basically the only time I worry about my fridge, is when it needs a defrosting after a period of time with lots of humidity.

The more ice on the evaporator plate, the less efficient the fridge is, as the ice is insulating the cold plate, preventing it from extracting heat from the interior. My last defrosting was overduw and the duty cycle and interior temps dropped considerably once I rid the evap plate of all the Ice build up.

I also have a 0.03 amp 40mm 12v computer fan inside which greatly speed the cool down of items placed within and evens out internal temps nicely. My cheap american swill beer is at this moment, 33.5f.

 

[ccbreder]

Yes we like experiments and back up data. I would suggest a condenser based soft start for your compressor and then just use one inverter. The mariners use such soft starts for their air conditioners. Using two inverters to power one device will prove to be complicated. If the inverter is only on while the compressor is running, does it make that much difference what the efficiency of the inverter?