building a incubator from a wine cooler

[gpop1]

About a month ago I told my wife that its time for a new project. My wife told me she wanted a incubator and there was a old wine cooler in the spare room. After browsing the web I couldn't find anything I liked so I decided to play with a few ideas to see if they were feasible.

my first trial was to see if the peltier (thermal cooler used to make the wine cooler cold) could be used as a heater, A peltier is not designed to be turned off and on like a bulb but that did not seem to be a major problem to fix if the peltier could produce the heat required. I wired the peltier backwards to a temp controller and with in a few minutes from cold the temp was stable between 99.4 to 100.2 which confirmed that the idea worked. Unfortunately at this point my wife is standing there with a batch off eggs 0.o.

I tried to warn her that a peltier will not last being switched on and off by a relay but all she heard was blar,blar,blar.

So time to find another wine cooler and start again. Still staying with a peltier design as this will be a 12vdc incubator that will have auto battery back up as our power is bad during the wet season.


Mk2 is a 12 bottle wine cooler. This time I used a arduino programmable micro controller. I bought a kit that included the arduino and some things to plug into it then followed the lessons on how to write code. After a few weeks I seem to have gotten the basics so its was time to start on the cooler,

The peltier is now running off pwm via a mosfet (that's just a fancy way of saying I now turn the peltier off and on thousands of times a second to control the temp). Its funny you can not turn one off and on every few minutes but if you turn it off and on really quickly then its fine.

my first attempt the temp stayed between 98 and 104.....lol

after a few days of playing it stayed between 99 and 102. The problem with a peltier is its a brick. It doesn't react to power changes as there so much stored energy that a automated control just doesn't work plus the 95cent temp sensor was a little unstable.

Upgarded to a $2 sensor and the temp was better. Deleted all the fancy code and just said 100% power until 99.4 then 50% power till 99.5 then 9% power until it starts to drop. Now the temp holds with in 0.1 of a degree. It also recovers after the door has been opened in under a minute so that's one problem solved.

will post more later.

p.s mk 1 is still running strong after one month and has done quail, turkey and pheasant eggs. If I ever get it back I will fix the heater correctly.

 

[gpop1]

arduino being wired and tested. The lcd display is taped to the computer monitor so its not in the pic

tests are on hold due to skipper getting on my desk and chewing the wires to bits





back of the cooler that wont be seen. As there was no room with the tilt tables inside it seemed a better idea to put it on the back and it should be more controllable. The idea was tested small scale with parts duck taped together, Im not sure if I need a de-humidifier yet. as I have a cold sink adding a vent to a thin aluminum tube should not be a major problem if I decide to add one




The inside of the cooler. Boy was I surprised when I realized the inside is not square. A little redesign and the tilt table is ready to slide in. Still waiting on the 0.8 rpm 12v motor to get here before final testing (will post pics later)





skipper my parrot that chewed up my project




what the cooler looked like when I got it.

 

[gpop1]

first lcd I was trying. It a 2 line 16 character which works fine but im upgrading to a 4 line 20 character as I want the home screen to display more info

below are pictures of the turner that is still a work in progress. The 3/8 dowel is being replaced with 3/8 square pvc rod, The aluminum channel still needs welding in the corners and everything needs to be made to look pretty.



over run switchs just in case it tilts to far



side that shows the 3/8 runners that engage the slots in the wine cooler



bar that links the trays together



pot that tells the arduino the angle of the trays (I need this so I can tell the trays to return to level or 5 degrees forwards so I can load and unload the trays)



turner slide inside the cooler. Thankfully its easy to get in and out as im having to check clearances for everything I add

 

[yellowherb]

Can not wait to fill this with eggs

 

[gpop1]

few more pics
this is the start of a auto vent that Im testing with a servo. If I add de-humidity to the rear I will need to make another to channel the airflow



butterfly came from a old engine.



Yay finally mk1 is empty so I can do some work on it.



a peltier assembly from mk1. I originally reversed the whole unit so the larger sink was on the inside then realized later I only had to reverse the wires. It worked well but the larger sink fan was not covered so you would occasionally hit your fingers on the fan



one of 2 temp/humidity sensors that I will be installing. Great thing with a arduno is that you can recalibrate the sensors. Planning to make the sensors check each other and If one plays up to disconnect it then alarm to tell me that it needs replacing

 

[wynn4578]

Impressive. I see now why you're going with the Arduino as opposed to the Pi. I am having trouble finding a wine cooler at the moment. When I decided to do this project there were 2 on craigslist in my area since I've decide a wine cooler is a must.....nothing.lol. I didn't even consider position control for the turners. Even going with a CPU controller my plan was still simple on - limit, wait x minutes, reverse on - limit. with a manual leveler for loading and unloading. Very impressive project you have going.

 

[gpop1]

Impressive. I see now why you're going with the Arduino as opposed to the Pi. I am having trouble finding a wine cooler at the moment. When I decided to do this project there were 2 on craigslist in my area since I've decide a wine cooler is a must.....nothing.lol. I didn't even consider position control for the turners. Even going with a CPU controller my plan was still simple on - limit, wait x minutes, reverse on - limit. with a manual leveler for loading and unloading. Very impressive project you have going.

My project started the same way as yours. I was searching the net for a cheap timer that could do 2hr then a 10-15second cycle and every timer seemed expensive. So I planned to use two cheap relay timers one in hours and one in tenths of seconds. The wife asked if the turner could hold the eggs for 48hrs before turning as most off our eggs are shipped via the post. Ok that's 3 relays. "will it stop after so many days when the eggs are ready to clamp". Ok that's 4 relays ........I like relays but now im wondering where am I going to mount all of these parts so I started to look for another option and that's where I found the arduino and all the toys that are designed to be used by it. Now I have a clock/calendar that I can reference and a timer that can work in any amount of units.

I think you still have a good plan and it should work. My plan I think will work after I iron out a few kinks

 

[Daedalus62]

Really really nice project!!!... And you have my full attention, so please keep us posted on your progress!!

I'm also planning on an Arduino controlled unit... Have all the bits 'n pieces to start learning Arduino, just can't find the time to...

Like you peltier approach... Clearly I'm not familiar with these, but I was under the impression they emit cold in one side and heat in the opposite... Why the "I wired the peltier backwards"?... Is it just an ergonomic issue?... I'm planning on a much (I hope) simpler nichrome wire coil, solid state relay controlled...

Is that a DHT11/DHT22 sensor?... How accurate/reliable are these in reading temperature and humidity at the same time?... Thoughts/experience on these?...

Clearly I still have to learn Arduino, but... Have you considered controlling egg tray turning using stepper-motors?... If this is doable with Arduino, it seems alot simpler to control the motors funtions (timing, speed, direction of turning and travel angle) all by program/code... Dispensing all extra hardware...

Sorry to be picking your knowledge so much, but this really interests me... Thanks for sharing and keep it up!

Cheers

 

[KodiakFox]

This is rad. I have a similar cooler I'm getting ready to alter although I don't plan to get quite as intricate as you have. I'm hoping I can replace some components with ones that come ready to do what I need rather than rewiring them. I'm sure it's not difficult but if I try and learn too much all at once I'll crap out!

 

[yellowherb]

Really really nice project!!!... And you have my full attention, so please keep us posted on your progress!!

I'm also planning on an Arduino controlled unit... Have all the bits 'n pieces to start learning Arduino, just can't find the time to...

Like you peltier approach... Clearly I'm not familiar with these, but I was under the impression they emit cold in one side and heat in the opposite... Why the "I wired the peltier backwards"?... Is it just an ergonomic issue?... I'm planning on a much (I hope) simpler nichrome wire coil, solid state relay controlled...

Is that a DHT11/DHT22 sensor?... How accurate/reliable are these in reading temperature and humidity at the same time?... Thoughts/experience on these?...

Clearly I still have to learn Arduino, but... Have you considered controlling egg tray turning using stepper-motors?... If this is doable with Arduino, it seems alot simpler to control the motors funtions (timing, speed, direction of turning and travel angle) all by program/code... Dispensing all extra hardware...

Sorry to be picking your knowledge so much, but this really interests me... Thanks for sharing and keep it up!

Cheers

my pc ssd crashed so im borrowing my wifes pc.

First thing to understand is a peltier is a ceramic chip. They found that when amps moved from one side of the chip to the other it had a neat affect of moving heat across the ceramic. You can not make cold air but you can move warm air which leaves you with cold air.
Now in a wine cooler the aluminum heat sinks are larger on the hot side as they are not planning to make a heater. I presumed that it was required but after more research and testing I found that the small sink was large enough to be used on the hot side. Simply reversing the way the amps flow changings the way the heat is transferred. Coolers run using pwm 12 across the peltier. Pwm is like turning the light switch on and off quickly and the light will seem to be on but dim. Its a way to keep full 12v but to change the amps. The arduino will do this at 1000's of times a second.

don't use the dht11. Its capable but unstable especially when other items are running like steppers. The dht22 seems to work fine but its slow to update which can work into your favor. Rtd pt100 and a conversion card is the best but probably overkill. Im using the dht22 and a 24hr burn in seemed to prove that it will work as required

I was going with a stepper motor when I was looking at egg turners but when I found that standard rails were to big for the cooler and interchangeable trays were cheap I changed the design to a tilt table. This causes a problem as the tray may be out of balance so the motor will need to hold the tray while it is not running. Thus I went with a low rpm worm gear motor which can not turned unless the motor is running. another problem with steppers is the lack of feedback on smaller cheap steppers. You tell it to go to a position and if it misses a step its still happy to go to where it thinks it is. With optical feedback the stepper knows its made a mistake so it auto corrects. Over sized steppers do help but you will need a H bridge to drive one. Just something to consider.

if you would like to tell me more about your design I am more than happy to help. I bought a arduino kit read 3 pages of a manual then got bored so I plugged it in and cut and pasted other peoples code or use the samples then started to modify the code to see what happened. With in a week I could code the basics. If you have a plan I can assist with the code