Peltier Lagering

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

LayMeister

Well-Known Member
Joined
Jun 27, 2008
Messages
322
Reaction score
1
Location
Annan
Hi,

I've been thinking about building an insulated box, cooled by a Peltier device, to use for lagering or even just controlling primary fermentation temperature. I have been looking for a device to use and almost bought a 5 gallon Coleman cooler today for $100 so that I could rip it out.

Then I started thinking about how many watts of cooling I really needed. The Coleman was about 50 W, but it said it could cool the inside (empty) to 40 degrees F below ambient in 3 hours. The assumption here is that whatever you put in should already be cold for it to work.

Then I saw a bunch of devices on E-bay with ratings up to 545W! And only $20, but it would require a 40A 12V power supply.

So I decided to do some basic calculations. I have a attached an image of my excel spreadsheet for you all to have a look at. If my calculations are correct then steady state load for a full 23l glass carboy in a 4" thick Styrofoam box is only about 10 W. (Looks like the Coleman cooler should work). However, even with 80W it takes about 10 hours to cool from 30C to 0C.

Given these results, I think I'll go for 168W device for $6.95 on Ebay

Comments?

Paul

PS If you are interested in a copy of the excel file, let me know.

Peltier.jpg
 
Would work nicely I think, just remember to have sufficeant cooling for the pelt. If you get to many watts (I think when using a pelt on a computer 200watts) you need to start using water to cool them. Lower watages have heatsinks and fans.

I have a phase change hooked up to my computer now, using propane gas I can get to about -50C head temps. I could easily lager in my bathtub if I put the head in there and had some sort of circulating propeller.
 
I always thought this was an awesome idea, but I think Yuri had limited success with it.

https://www.homebrewtalk.com/showthread.php?t=27524&highlight=peltier

not to discourage you though, I'd love for this to work so I could lager.

I saw the thread about Yuri's endeavor. I think he has two problems:

1) The tank has a large surface area as compared to the volume. This means a lot of heat can flow in through the stainless keg. (The optimal shape is actually a sphere which minimizes the ratio of surface area to volume.) In addition, water is a poor heat conductor. This means it is difficult for the heat to flow from the top and bottom of the keg to the cooling probe in the middle.

2) He didn't insulate the keg. In my calculations, it takes much more power to cool the carboy than to keep it cool. However, if I assume a insulation thickness of .25" (I don't know the thermal resistance of Stainless) then it takes 175W to keep it cool; at .125" it takes 350W! (My assumptions don't fit his configuration so there is some significant error margin here). In any case I think that he needs a lot more powerful Peltier device if he doesn't want to use insulation.

I plan to construct a box with the cooler at the bottom. There will be a 16" diameter, 1/4" thick, plate of aluminum inside the box between the carboy and the cold side of the Peltier. The hot side will have an air cooled heatsink to begin with. If this isn't sufficient then I'll change to a water block and cooling water controlled by a solenoid so that I don't have to leave the water running all the time. I expect that once the carboy is cooled that it will run less than 5% of the time.

Thanks for the encouragement,

Paul
 
Yuri has his mounted on a keg, I assume the kegs are made from Aluminum a good metal for transfering heat but not the best and probably not good enough for a large keg. I think what he would have a problem with is one cold spot and the rest of the keg staying at about the same temp. If it were to work on a keg you would need to use some sort of foam insulation, I would also put the peltier on a copper block and solder some copper heatpipes around the keg.

Also with that little fan and heatsink I doubt that pelt is strong enough to even have the capcity to cool the keg, he would need a larger pelt or use more than one. I think to get best results from a pelt it would need to be utilized in a pool of water, somehow soldering some copper heatpipes on to it that went into the tub where something circulated water through them(a pond pump or something)

Yuris idea was good but I think it had flaws from the start. If I were to do it I would get a CPU waterblock like this: http://www.sidewindercomputers.com/thxwcpuwablf.html and have it mounted to the hotside of the pelt for cooling. This waterblock would have a returning loop that passes through the block, into a small radiator, into a small reservoir, through the pump and back through the block. This is your cooling unit for the pelt. it can be contained in a very small space(smaller than your computer case) So box this setup up, then take one of these: http://www.sidewindercomputers.com/thhrplforso7.html and use thermal epoxy to attach it to your cold plate on the pelt. Mount this piece in your tub and use a pond/fountain pump to move water over it.

Now keep in mind a pelt uses lots and lots of electricity and it will be close to water so make sure this pelt is waterproofed in some way(it shouldn't be touching water but just in case) and make sure it has short protection. I also havent studied this setup to find out if the heatsink would make the whole heatsink cold or not or if it would peter out at the base. Im sure if you use a strong enough pelt it would be great(def. good enough to get to lager temps). So in essence you have a small condenced version of a wort chiller running in your tub of water that your beer is in. I would use as small of a tub as possible such as a trash can so you have as little volume to flow over your heatsing as possible but still enough water to flow around your entire beer. There would be a trick in finding a good balance for water volume to pelt size. To much water your pelt runs for a long long time before it cools down to the proper temp, to little your pelt runs all day maintaining the temp(and you loose temp stability). Im thinking 10 gallons but this is variable of course.

So your looking at a decent price tag, probably closer to 200 bucks for the whole rig but you now have a Portable fermentation cooler(wich means lagering in the dead of summer in nearly any house), and even a wort chiller.

I say go for it Paul, it can work. We use these things on computers all the time I don't see why we can't use it on a beer. Keep us posted!
 
I attempted to make a peltier cooled box and ran into a few problems. The major one was removing the heat from the hot side of the peltier before it could transfer to the cold side. The issue with the higher wattage peltiers is that it gets so hot on the hot side that it takes alot to cool it. I bought a pretty large heatsink and a high rpm - large radius fan to cool the hot side. I was never able to get a frost on the cold side like I was able to with a lower wattage peltier. In fact I burnt the peltier out because it got too hot.

I settled with an ice probe (fish tank cooler) and a corny. With that combo I can get -20 degrees ambient.
 
Yuri has his mounted on a keg, I assume the kegs are made from Aluminum a good metal for transfering heat but not the best and probably not good enough for a large keg. I think what he would have a problem with is one cold spot and the rest of the keg staying at about the same temp. If it were to work on a keg you would need to use some sort of foam insulation, I would also put the peltier on a copper block and solder some copper heatpipes around the keg.

Also with that little fan and heatsink I doubt that pelt is strong enough to even have the capcity to cool the keg, he would need a larger pelt or use more than one. I think to get best results from a pelt it would need to be utilized in a pool of water, somehow soldering some copper heatpipes on to it that went into the tub where something circulated water through them(a pond pump or something)

Yuris idea was good but I think it had flaws from the start. If I were to do it I would get a CPU waterblock like this: http://www.sidewindercomputers.com/thxwcpuwablf.html and have it mounted to the hotside of the pelt for cooling. This waterblock would have a returning loop that passes through the block, into a small radiator, into a small reservoir, through the pump and back through the block. This is your cooling unit for the pelt. it can be contained in a very small space(smaller than your computer case) So box this setup up, then take one of these: http://www.sidewindercomputers.com/thhrplforso7.html and use thermal epoxy to attach it to your cold plate on the pelt. Mount this piece in your tub and use a pond/fountain pump to move water over it.

Now keep in mind a pelt uses lots and lots of electricity and it will be close to water so make sure this pelt is waterproofed in some way(it shouldn't be touching water but just in case) and make sure it has short protection. I also havent studied this setup to find out if the heatsink would make the whole heatsink cold or not or if it would peter out at the base. Im sure if you use a strong enough pelt it would be great(def. good enough to get to lager temps). So in essence you have a small condenced version of a wort chiller running in your tub of water that your beer is in. I would use as small of a tub as possible such as a trash can so you have as little volume to flow over your heatsing as possible but still enough water to flow around your entire beer. There would be a trick in finding a good balance for water volume to pelt size. To much water your pelt runs for a long long time before it cools down to the proper temp, to little your pelt runs all day maintaining the temp(and you loose temp stability). Im thinking 10 gallons but this is variable of course.

So your looking at a decent price tag, probably closer to 200 bucks for the whole rig but you now have a Portable fermentation cooler(wich means lagering in the dead of summer in nearly any house), and even a wort chiller.

I say go for it Paul, it can work. We use these things on computers all the time I don't see why we can't use it on a beer. Keep us posted!

But at that price, you could get a fridge or two. I think, or at least I would want this to be cheaper than existing solutions or at least do it better. If not then really what is the point?
 
Remember: 40A @ 12V is only 4A @ 120V. (480 watts)

I tried to build a Peltier swamp bucket with some peltiers given to me by another guy in the club (from an old office water cooler), but they just weren't powerful enough to get the temp down more than a couple °C (uninsulated, but well cooled on the hot side (air cooled)).

Beer%20-%2020.preview.jpg


S
 
that's assuming you have good efficiencies with your power supply

but even at 4 amps, you are getting above the power load of a refridgerant based system.
 
This guy has a cool thing going on. I wonder how well it works and if he'd chime in here.

962-FermentationCooler.JPG


962-FermentationCooler2.JPG

"Two carboys inside one side of the fermentation cooler. The top shows the inner side of the peltier cooler outfitted with a fan. The temperature sensor wire is seen running down to an insulated sleeve which detects the wort temperature more accurately than if it were just hanging inside the cooler."

962-FermentationCooler3.JPG

Fermentation cooler dual controller and one of the two peltier units.

I wonder if he can get that thing down to lagering temps.
 
Remember: 40A @ 12V is only 4A @ 120V. (480 watts)

I tried to build a Peltier swamp bucket with some peltiers given to me by another guy in the club (from an old office water cooler), but they just weren't powerful enough to get the temp down more than a couple °C (uninsulated, but well cooled on the hot side (air cooled)).

Beer%20-%2020.preview.jpg


S

The problem here is simply the lack of insulation.

Lets do some basic calculations:

If we approximate a 5G carboy by a cylinder 46 cm high and 27 cm in diameter then the surface area is about .5 m^2 . A carboy weighs 5.5 Kg. The density of glass is 2.5 gm/cm^3. From this we find the average thickness of the glass is 0.44 cm (assume uniform thickness over entire surface).

Now lets assume the glass in the insulator between the beer and the outside world. The thermal resistance of glass is 1.1 W/(m-K). So the thermal resistance of the carboy is about 123 W/K or W/degree C.

If the beer in the carboy is 3 degrees C colder than the outside environment then there is a net flow of 3 * 123 = 369 Watts out of the carboy. This seems very consistent with the results obtained using the Swamp Bucket.

You can actually verify the thermal resistance by putting cold water into the carboy and monitoring the liquid and ambient temperatures over a period of a few hours, as the water warms up. I did and the actual resistance seems to be just a little higher than my simple calculation (remember the water is also an insulator of sorts).

I am now more sure than ever that Peltier Lagering will work. That it will cost me about $75 by the time I'm finished and will be more energy efficient than a refrigerator. I guess the proof will be in the putting. I'm building the very well insulated box on the weekend and the Peltier device should arrive early next week.

I'll keep you posted.

Paul
 
I do think this may work well with an insulated box (Son of fermentation chiller type) because insulation from the outside world will be so much more important. I also think that if insulated well enough this could be used out in a garage (maybe with more than one peltier device) then in the winter you can use it in reverse to keep it warm enough for fermentation.
 
FWIW, I was measuring the temperature of the chill water in my tests, not the temperature inside the cooler. Yeah, I do think insulation would have helped, but I'm sure that the main problem was underpowered peltiers/supply.

S
 
I tried using one (had R-40 insulation in the box). Keeping the hot-side cooled is very difficult. Between the power supply and the heat-sink's output, it was like having an electric heater in the room. A 2L bottle of ice changed out every day works much better.
 
I tried using one (had R-40 insulation in the box). Keeping the hot-side cooled is very difficult. Between the power supply and the heat-sink's output, it was like having an electric heater in the room. A 2L bottle of ice changed out every day works much better.
but can you get lager temps with that?
 
The problem here is simply the lack of insulation.

Lets do some basic calculations:

If we approximate a 5G carboy by a cylinder 46 cm high and 27 cm in diameter then the surface area is about .5 m^2 . A carboy weighs 5.5 Kg. The density of glass is 2.5 gm/cm^3. From this we find the average thickness of the glass is 0.44 cm (assume uniform thickness over entire surface).

Now lets assume the glass in the insulator between the beer and the outside world. The thermal resistance of glass is 1.1 W/(m-K). So the thermal resistance of the carboy is about 123 W/K or W/degree C.

If the beer in the carboy is 3 degrees C colder than the outside environment then there is a net flow of 3 * 123 = 369 Watts out of the carboy. This seems very consistent with the results obtained using the Swamp Bucket.

You can actually verify the thermal resistance by putting cold water into the carboy and monitoring the liquid and ambient temperatures over a period of a few hours, as the water warms up. I did and the actual resistance seems to be just a little higher than my simple calculation (remember the water is also an insulator of sorts).

I am now more sure than ever that Peltier Lagering will work. That it will cost me about $75 by the time I'm finished and will be more energy efficient than a refrigerator. I guess the proof will be in the putting. I'm building the very well insulated box on the weekend and the Peltier device should arrive early next week.

I'll keep you posted.

Paul

I'm thinking about welding up a aluminum box just big enough for two carboys side by side. I'd fix the TECs to the outside of the box. Put the carboys in the box, fill the box with water. For the hot side of the TECs, heat sinks sitting in plastic ductwork with a blower fan forcing the air. Insulation on the outside of the metal box. What do you think?
 
Any updates?

I had the same idea and was already planning circuit design. Basically, I was planning a manual PWM control over the cell, with an automatic "too-cold" ON/OFF circuit (like powering off the peltier cell below 8°C and then back on at 12°C). Your thread caught my attention. I'm impressed with your calculations. I was planning to put the peltier cell on top of the insulated box and the carboy slightly higher than the bottom to put ice packs. Keep us updated!


The problem here is simply the lack of insulation.

Lets do some basic calculations:

If we approximate a 5G carboy by a cylinder 46 cm high and 27 cm in diameter then the surface area is about .5 m^2 . A carboy weighs 5.5 Kg. The density of glass is 2.5 gm/cm^3. From this we find the average thickness of the glass is 0.44 cm (assume uniform thickness over entire surface).

Now lets assume the glass in the insulator between the beer and the outside world. The thermal resistance of glass is 1.1 W/(m-K). So the thermal resistance of the carboy is about 123 W/K or W/degree C.

If the beer in the carboy is 3 degrees C colder than the outside environment then there is a net flow of 3 * 123 = 369 Watts out of the carboy. This seems very consistent with the results obtained using the Swamp Bucket.

You can actually verify the thermal resistance by putting cold water into the carboy and monitoring the liquid and ambient temperatures over a period of a few hours, as the water warms up. I did and the actual resistance seems to be just a little higher than my simple calculation (remember the water is also an insulator of sorts).

I am now more sure than ever that Peltier Lagering will work. That it will cost me about $75 by the time I'm finished and will be more energy efficient than a refrigerator. I guess the proof will be in the putting. I'm building the very well insulated box on the weekend and the Peltier device should arrive early next week.

I'll keep you posted.

Paul
 
Just a quick reply, I only read most of the posts loosely,...

Peltiers are rocket science !
They sound very attractive, but you need to read a lot about them. I have read quite a bit, and gave up my project.
There are different types of peltiers with very different specs, they do not cope very well with moisture, you need _lots_ of cooling on the hot side, you need a high quality powersupply,... these are just some of the things I remember,...

But do by all means share any experience, I would love to use one !


cheers,
walther
 
the only problem I see if cost. A johnson or ranco temp controller is ~50 on the high side. Add $0-$40 for an old freezer off craigslist and you have a fermentation cabinet for under 100 that holds 4-6 carboys.
 
Well I've finally gotten back to my peltier lagering project. I manage to find a small fridge with two large peltier units already attached. I've cut out the top so it can accommodate 5 gallon and 6 gallon carboys as well as Corney kegs. I am in the process of making an insulated cover. Now I need a controller. I want to be able to heat and cool - I love Belgian ales and Lagers. This means a dual control. Problem is the I got the fridge for $50 new and a dual controller is at least $60 to $70.

I was thinking in the long run I might want 3 or 4 of these chambers. Is there some other control module I could get that would handle this? It would need to handle 4 thermocouples and control 8 outputs. It would be nice if I could hook it up to an old laptop to control and monitor. Alternately, something with a web interface would be nice. Any ideas or suggestions?

Paul

IMG_0087.jpg
 
If you want to send it to me, I can have it installed in my chamber in a matter of days and report back with how it is working! :)
 
I can only get my chamber down to about 60F with the peltier that I robbed from a wine cooler. Not lagering temps but what I wanted was a way to hold ferm temps steady in my upstairs where the temp swings during the hot summer. I guess I could better that number with a second peltier of the same output.
 
Back
Top