Building a Custom Refridgerator

BeNimble · 10-14-2020, 02:10 AM

I am going over the top again and want to build a custom fridge.
My Plan A was this 18cf fridge I have and was going to make into a chest fridge/freezer,
but then I decided it takes up too much floor space. So Plan B I got a smaller chest freezer to make into a fridge. Now I need another freezer, so Plan C was to build a custom fridge by taking a side-by-side and remove the fridge side, the key being it includes the ice and water dispenser.
Ok, so now I'm on to what I think is the best of everything Plan D and cut a french door fridge in half. Why? It has ice/water dispenser, but the French Door fridge, the ice maker is isolated, so the fridge is still behind the door.
The Freezer door is below, and pulls out and with some mods to the door, it won't spill out the cold. Just the right size I think. I actually had to buy this French Door fridge for $50, as it doesn't work. It will also look good next to the stainless steel countertop. Is this interesting to anyone else?

Native · 10-14-2020, 03:33 AM

Interesting .... keep us posted with this mini conversion.

cadillackid · 10-14-2020, 08:36 AM

way to fancy of a french door frig.. the computer boards or inverter compressors and parts are the mpst common failure. so unless you are going to make your own cooling system for it completely then it seems like a waste if it is non working as it could be quite costly to make the original system work again. if you just got it for the cabinet and can stuff the evaporator and such into one half then go for it.. the ones i have seen the evaporator is more centered and going to be more difficult to move to one half.. the concept of a drawer freezer, door frig is neat though as i dont think ive ever seen on in an RV model before..

Booyah45828 · 10-14-2020, 08:51 AM

Sounds like a neat project.

I have a similar fridge that was given to me for the same reason, but it's an older model that is analog and uses R134a. I'm looking into replacing the compressor(I assume that's the failure) with a DC powered option. I'll just have to make sure the electrical is capable of the change, and will likely have to purchase a different heating element for the defroster and a different defrost timer.

I'm hoping to have it all done for way less then the cost of a comparably sized DC fridge.

Rucker · 10-14-2020, 10:39 AM

You raise a very interesting point that every conversion project needs to consider when deciding on a design path - what long term implications of any particular design decision may haunt us.

A simple example: I used 3/4" plywood to build my kitchen/bathroom core. Some joints are glued and screwed, and others are just screwed. That way if I need to take it apart for repair later on I'm not going to need to rip it apart.

For refrigeration, whatever you come up with that works for you, invariably it will need repair or replacement. Will you have the same appetite for a uniquely constructed fridge five years from now, especially if it breaks when you're on the road?

I've mentioned elsewhere that my fridge is a standard-sized dorm fridge driven by A/C. All of the design effort went into the electrical solution that minimizes inverter power consumption. The enclosure for the fridge is pretty standard so if it fails on the road I can stop at any applicance store and swap the unit out. Maybe need to redo some trim depending on the size of the replacement unit.

So do something radical and cool; but also consider factoring in maintenance and repair complexity.

BeNimble · 10-14-2020, 01:03 PM

The key requirement here is ice maker/water dispenser. I use ours everyday.
A standard fridge with it is huge and heavy. Plan A was to hack one into the small fridge, but I am less confident this will work properly and reliably.
The plan has always been to use the bus A/C condensor for the fridge, so it will be a custom setup. Putting the heat back into the bus seems not great approach, and making ice will mean the fridge is going to be running a lot, vs maintaining temp.

The thing about a fridge is it on 24x7, so if using an inverter it will be on all the time and incurs the overhead consumption of an inverter. A small one just for the fridge might be tolerable. The ice dispenser uses a 120vac motor.

Now I could just fix this fridge and sell it for like $400 get the cost of the Nautibus back to almost zero. I just fixed another free one that is going up for sale at $200.

BeNimble · 10-15-2020, 10:35 AM

I diagnosed this fridge and determined the compressor was running but not pumping.
Took it apart, and the reed valves are broken. It has an "inverter" compressor, which has a BLDC motor in the compressor. It takes the 120vac and rectifies it and doubles it, so it runs at 240vdc, but varies the frequency and thus rpm of the compressor.

The key point is that one can power this fridge with a simple cheap square wave inverter, it doesn't need a sine wave, as it just converts the ac into dc.
So a $30 300watt inverter will do.

Now I have been working on a 36vdc compressor motor, but really with this kind of compressor, there really is not much need.

I am going to get another free fridge today, this one is the ultra fancy Thermador $10,000 model that stopped working. I like the size, as its only 25 inches deep, vs the 'counter depth' of the French door 28 inches, and this one is tall. So if I just use the freezer side and modify it, it might be a better size.

So here is the house this fridge is coming from, $1.5 Million that is a lot even for sacramento! lol Owner said they wanted $4,000 to repair it..

Rwnielsen · 10-15-2020, 11:05 AM

It sounds like you know what you're doing, I'd go for for it. Sine wave for the electronics with a frequency controller seems to be what's existing. I'm not familiar with a square wave version but if you can make it work on the bench... I'd test run it for a spell so you don't have to mess with it on a trip. Cool idea, good luck

BeNimble · 10-15-2020, 02:23 PM

Just in case anyone cares... Here are pics of compressor insides.
The leftmost shows the std 120vac induction motor and the rotor and stator.
The middle is the one I am working on to convert to 36vdc, rewinding the stator and making a new rotor with magnets. The rightmost is the Inverter compressor, which you can see is built pretty differently. And notice the Stator is half the size of the induction motors.

joeblack5 · 10-15-2020, 03:48 PM

Great project, I would use the inverter motor and rewind that for lower voltage. Seems you made a nice cut on the compressor housing.. Maybe you can weld a new flange to it and make a O-ring groove and put it back together. In that case you can repair it forever.

I only use the danfoss BD35 compressor fridges, so have spares on the shelf.
Maybe two in parallel would have enough cooling capacity for your big fridge.

good project

Johan

Bus'n it · 10-15-2020, 04:58 PM

I have always been fascinated with refrigeration. When I was in HS, I cut open an old working Hermetic compressor, pulled out the parts and fastened them to a plywood board. That got me free cut days from the shop teacher!

Bus'n it · 10-15-2020, 04:59 PM

Quote:

Originally Posted by joeblack5

Great project, I would use the inverter motor and rewind that for lower voltage. Seems you made a nice cut on the compressor housing.. Maybe you can weld a new flange to it and make a O-ring groove and put it back together. In that case you can repair it forever.

I only use the danfoss BD35 compressor fridges, so have spares on the shelf.
Maybe two in parallel would have enough cooling capacity for your big fridge.

good project

Johan

Would you happen to know if that new Furrion Arctic 12v RV fridge uses a Danfoss compressor? I think it may be a Chinese clone. I really like that fridge for my bus but I am afraid of the reliability. Once the bus is built out the only way to get the fridge out would be through the front windshield!

BeNimble · 10-16-2020, 04:43 PM

Decided to see if I can make the Thermador freezer side into a fridge/freezer.
This fridge is a BEAST at 526lbs, even with the doors off, the contents empty, the shell was too heavy for two of us to pick up, not easy to get it on the trailer.
So I just started to dismantle it on the trailer, and have cut off the freezer side, and was still heavy, I stripped the sheet metal off the outer sides, and now it is still not lightweight, but more manageable. The sheet metal is thick and it is steel inside, not plastic like most fridges.

So next up will probably be getting it working again, enough to see if I can isolate the top portion into a fridge, not freezing cold, which is going to be a big challenge with the ice maker in the middle. The good news is this model put the compressor on top, so there is no bump into the fridge at the bottom to make room on the outside for the compressor.

cadillackid · 10-16-2020, 05:05 PM

cool stuff.. im almost wondering though that might iot be better just to use the cabinet and put in your own guts? simple guts from an older refrigerator? (or bought new) vs trying to make the oroiginal parts work in a cabinet 1/2 the size and still maintain temperature?

the idea of usingthe inverter compressor though is intruiging.. esp if you dont have to deal with an EEV.. if its still a capillary orifice .. the EEV aperture and inverter compressor speed balance is a mix of art and science and sensors that takes quite a bit to get just right..

joeblack5 · 10-16-2020, 06:27 PM

Sorry Busn it, I do not know. i have been flagged before because I thought that everybody used the Danfoss compressor and I was proved wrong.

I have a boat danfoss bd35 unit + condensor that has screw connections to the evaporator plate. If I had a project fridge then it seems that having a compressor that can be easily disconnected and replaced would be a good thing.

Danfoss, secop , nidec ..all related.

https://www.secop.com/fileadmin/user...esb610a402.pdf
Nice timeline and technical read

Johan

BeNimble · 10-16-2020, 07:21 PM

This Thermador despite being top-of-the-line has the same components as the cheap fridges I've disassembled. While the dorm fridge and the chest freezer don't even have lights or defrosters, they just have a temp switch that cycles the compressor on/off.
Anything more complicated has a microprocessor with sensors that turns things on/off.
As long as the sensors are still in place, I don't see why it would not still work. The complication here is more mechanical to isolate a fridge section where there was not one previously.

BeNimble · 10-17-2020, 10:20 PM

I got a $22 300watt inverter to see if it would power the compressor.
First I tried it on a normal one, not a chance. Then tried it on the smallest I had dorm fridge one, sometimes it would startup if I tried enough times, basically no go. If it got past starting up, it was ok when running.

So I tried it on the big French Door fridge with the inverter compressor,
and nope it doesn't work. Now I also tested the voltages to confirm 240v earlier,
and indeed I see 240v at the compressor when full speed. The fridge starts it up slow and then speeds it up, so I was thinking great, no startup motor current.
But then NOPE, the problem now is the inverter has a big inrush current to charge up two big capacitors, so when you first plugin the fridge, it draws over 5amps and shutsoff the 300w inverter. I know all about this problem with electric cars, we use "precharge resistors" to charge up the motor controller capacitors at lower voltage before turning on full battery current to the controller. I did something similiar here and added a short length of heating resistor wire, and now the fridge will power up and run with the $22 300w inverter. When running I see about 6amps at 12v input.
When I power up the ice dispenser, it jumps to 20amps at 12v, still ok with the 300w inverter.

Always interesting to waste time doing this stuff, thought I would share.

Rucker · 10-21-2020, 05:09 PM

Quote:

Originally Posted by BeNimble

I got a $22 300watt inverter to see if it would power the compressor.
First I tried it on a normal one, not a chance. Then tried it on the smallest I had dorm fridge one, sometimes it would startup if I tried enough times, basically no go. If it got past starting up, it was ok when running.

So I tried it on the big French Door fridge with the inverter compressor,
and nope it doesn't work. Now I also tested the voltages to confirm 240v earlier,
and indeed I see 240v at the compressor when full speed. The fridge starts it up slow and then speeds it up, so I was thinking great, no startup motor current.
But then NOPE, the problem now is the inverter has a big inrush current to charge up two big capacitors, so when you first plugin the fridge, it draws over 5amps and shutsoff the 300w inverter. I know all about this problem with electric cars, we use "precharge resistors" to charge up the motor controller capacitors at lower voltage before turning on full battery current to the controller. I did something similiar here and added a short length of heating resistor wire, and now the fridge will power up and run with the $22 300w inverter. When running I see about 6amps at 12v input.
When I power up the ice dispenser, it jumps to 20amps at 12v, still ok with the 300w inverter.

Always interesting to waste time doing this stuff, thought I would share.

When I saw you were thinking a 300 watt square inverter to drive an induction motor I thought you had some special JuJu to make it work. Elsewhere I posted the math on my dorm fridge inverter but this forum doesn't have very good search logic and I can't actually find the post to link to.

I love your willingness to experiment-as I've said to you before, that's the fun stuff we get to do with these conversions. I'll tag on a few comments to this thread just to round out the fridge power considerations for others watching your build and interested in the topic. Not that I'm the expert, I'm just sharing my research.

Induction motors mean inrush current, and without some way to cover that one or two second demand, you're stuck.

One rule of thumb to cover inrush current is to make sure the base capacity of the inverter can handle the inrush, not just an inverter than can handle the inrush at peak or max capacity. In my case, I needed 1200 watts for the inrush current. I picked a 1200/2400 because of this. The reasoning for this is that the inverter and the compressor motor should not run outside of their design limits in normal operation -they won't last as long, and they'll break down sooner.

Also, pure sine is always recommended over square wave for induction motors. I can't explain why as well as a Google search will tell you, but it's generally true, and more true as your inrush current approaches the design capacity of the inverter.

Inrush current might be on the nameplate of the fridge. Sometimes it's listed as LRC, or Locked Rotor Current. If it's not there and you don't have a multimeter with Max Current setting you can use 3x or 4x running current but it's best to spring for the multimeter to get the exact inrush demand.

Can you live without pure sine wave, or a smaller inverter? Sure, it might work. On the other hand, designing the power for the fridge correctly means peace of mind and fewer problems with equipment or performance down the road.

joeblack5 · 10-21-2020, 06:17 PM

Nimble is using an inverter motor, 2 phase but more likely three phase. The fridge has an integrated VFD, variable frequency drive. The first stage of the VFD has a rectifier and capacitor that converts the 120vac / 60 hz into dc voltage . Because of the slow startup and motor design the start current for the motor is low.. in general with electric motors the current is linear with the torque. So if you accelerate slow you need less torque to get to the speed. So less current..and thus no inrush current.

Since the VFD first rectifies the input voltage to DC high voltage an inexpensive square wave inverter can do the job. The VFD takes care of the frequency, phase and wave form to run the motor.

In this case the high voltage DC capacitor bank causes an inrush current unrelated to the motor. Nimble has solved that problem by charging the capacitor-s more slowly by adding a resistor in series till the capacitor bank is charged up and then bypass the resistor..before he actually starts the compressor motor... some larger VFD's have that feature build in.

All in all a very nice approach. With the result that he can run a full size fridge from a 200 watt square wave inverter.

The overall efficiency could be better. The 12 to 120 vac inverter. Then the rectifying ...last the conversion into a 2 or 3 phase voltage each have their losses. That can be reduced by using a compressor motor closer to the battery banks dc voltage.

Johan

Rucker · 10-21-2020, 06:38 PM

Quote:

Originally Posted by joeblack5

Nimble is using an inverter motor, 2 phase but more likely three phase. The fridge has an integrated VFD, variable frequency drive. The first stage of the VFD has a rectifier and capacitor that converts the 120vac / 60 hz into dc voltage . Because of the slow startup and motor design the start current for the motor is low.. in general with electric motors the current is linear with the torque. So if you accelerate slow you need less torque to get to the speed. So less current..and thus no inrush current.

Since the VFD first rectifies the input voltage to DC high voltage an inexpensive square wave inverter can do the job. The VFD takes care of the frequency, phase and wave form to run the motor.

In this case the high voltage DC capacitor bank causes an inrush current unrelated to the motor. Nimble has solved that problem by charging the capacitor-s more slowly by adding a resistor in series till the capacitor bank is charged up and then bypass the resistor..before he actually starts the compressor motor... some larger VFD's have that feature build in.

All in all a very nice approach. With the result that he can run a full size fridge from a 200 watt square wave inverter.

The overall efficiency could be better. The 12 to 120 vac inverter. Then the rectifying ...last the conversion into a 2 or 3 phase voltage each have their losses. That can be reduced by using a compressor motor closer to the battery banks dc voltage.

Johan

Thanks for the plain English version! Still, that is what I call JuJu!

I'm guessing as long as the efficiency loss is less than 20-30% it's also a good solution for overall energy consumption?

You may also like: