Fermenter Cooler project

**fish4fun** · 01-08-2012, 07:01 AM

I have been playing with an idea using a TEC (ThermoeElectric Cooler) for chilling/heating a fermentation vessel. These little semiconductor devices offer ~85W of cooling/210W of heating at a very low initial cost (< $ 0.10/W) and a high degree of temperature separation from ambient (~60C). Honestly associate fans/heat sinks/insulation are more primary costs. The general idea is to build a cost effective temperature controlled environment for fermentation. While I fully intend to explore the venue for my own purposes, I am curious if such a device would be of interest to others.

Fixed costs of such a cooler would be the TEC & heat sinks, the PS, the "box" and the controller; variable costs would be the size of the box, the amount of heat removal (cooling) and the fit and finish of the "box".

Everything I am talking about can certainly be achieved with a craigs list freezer and an external thermostat, perhaps for less money. The cost of a 6 gallon carboy sized device would be close to $200, but the cost of a 1/2bbl Sankey keg sized device wouldn't be much more (perhaps $250). The advantage to individual fermenter coolers would be independent temperature control and space savings (pick a temp between 32F and 86F with +/- 0.25F or greater resolution and +/- 0.5F accuracy) with data logging for the entire fermentation process (180 days or more in user selectable time increments easily imported/monitored with a PC.)

I am curious if anyone would be interested in such a cooler @ ~$200-$300; or, if anyone would be interested in an open source project of basically the same cost basis. (Honestly in "production" the cost (w/o shipping) of a finished unit would probably be less than DIY). The finished dimensions of the cooler would essentially be the fermenter vessel + 4in in all planes except the "top" where the "internal box clearance" would need to be height + ~9in.

The expensive parts of the system would be:

1) The liquid heat Transfer System (~$70)
2) The Pourable Foam used for insulation/construction(~$60)
3) The Power Supply (~$25)
4) Cosmetics ($20)
5) The Cooling Fan & Heat Sink ($15)
7) The Misc Alum parts (~$15)
8) The uController PCB & Components (~$15)
9) TEC (~$7)

The mold for the foam (with a modest CNC router or careful hand work) shouldn't cost more than ~$50 in materials. The various "finishing" parts should be trivial in material cost, but would require either careful hand or CNC processing.

Obviously the cost estimates DO NOT include the fermentation vessel, and as the volume of the fermentation vessel increases the various costs associate with construction increase (but with some economy of scale, eg, the FOAM & "finishing" components increase as volume increases, but the power supply, cooling fans & heat sinks remain fairly constant).

While I plan to follow this project through to a few finished units for personal use regardless of other interest, I am curious if others might be interested in 1) an open source project where the design, PCB, firmware/software and construction were documented, or 2) where the finished product were offered for sale. The two are NOT mutually exclusive. I have a day job, and am not looking to support myself from this endeavor; if I made these units for sale it would be on a hobby scale, much like a wort chiller: any profit would come from purchasing materials in bulk. The primary advantage to individual cooling units for fermentation would be to those interested in simultaneously fermenting different batches at different temperatures, or for people who have space restrictions. From an energy point-of-view a TEC (peltier cooler) is ~ 80% less efficient than a conventional liquid based heat pump, but they are far less expensive to construct than a similarly sized heat pump on a small scale. TEC cooling requires CONSIDERABLY less daily effort than "ice" based cooling but is otherwise similar in cost of construction, and perhaps less expensive from an energy foot-print point-of-view.

Again, the project does not involve "magic", and there are certainly ways to achieve the end result less expensively (eg a $50 craigs list chest freezer and an external temp controller), but if there are people out there who are interested in fermenters with a small footprint, exacting temperature control AND logging with minimal effort then perhaps this project is of general interest.

Thanks in advance for any comments or responses :-)

Fish

**muthafuggle** · 01-08-2012, 02:18 PM

If I read this correctly, the peltier cooler is 80 % LESS efficient than traditional refrigeration?

You pretty much had me right up until that.

I would love two fermentation vessels, one to ferment in and one for lagering. Tracking and logging temps intrigues me, too. Did I just misunderstand your post about the efficiency?

**fish4fun** · 01-09-2012, 12:27 AM

muthafuggle,

Quote:

Did I just misunderstand your post about the efficiency?

Sadly you did not; peltier cooling is roughly 80% less efficient than refrigerant based cooling; however, this does not mean such a device would make your electric meter spin like a top. A worse case scenario would be if the cooler had to run 24/7. Since the max power input to the cooler is ~150W (including pump & fans), assuming the power supply is at least 80% efficient this would imply a maximum power consumption of 150/0.80 < ~190W. (190W/ * 24Hrs * 30days * $0.12/kWh)/1000W/kW = $16.41 per month. In a more realistic example where the cooler cycled on only 5 minutes per hour a 30 day cost of operation would be < $1.50. If you compare this to the cost of an ice based system that required 4kg of ice per day (roughly one gallon of ice frozen in your freezer), the energy required to freeze the water would cost ~ 15.3kWh * $0.12/kWh = $1.84. (This is derived from: 30 Days * 4L * ((1Kg * 4.182kJ/kg * 25C) + 334kJ/kg + (2.11kJ/kg * 10C)) = 55,080kJ/3600seconds = 15.3kWh; this assumes your freezer temperature is -10C (14F), and it assumes your freezer is 100% efficient.)

The fact that a tightly sealed 3 inch thick foam "box" gains/loses considerably less heat than a typical consumer refrigerator/freezer goes a long way to leveling the playing field. Following is a table of projected heat losses for the proposed "box":

Code:

Room    Fermenter  Power    100% Eff
Temp     Temp     Per Day  Cost/Month
 F         F         Wh	       $
 77       68        120       0.43
 77       59        257       0.85
 77       50        356       1.28
 77       41        474       1.71
 77       32        593       2.13
 77       23        711       2.56

To achieve ~80W of cooling we need to input 135W. So, to maintain a temperature over the course of the day we need to remove heat at the same rate it is coming in. So, if 120Wh per day is coming in, we need to remove 120Wh per day. In order to remove 120Wh, we need to turn on our peltier cooler for 120Wh/80W = 1.5 hours. Operating the cooler for 1.5 hours will consume 135W * 1.5 hours = 202.5Wh per day. 202.5Wh * 30 days = 6075Wh or 6kWh. 6kWh * $0.12/kWh = $0.72.

In the event this horse is not sufficiently beaten: HVAC efficiencies are frequently compared using SEER (Seasonal Energy Efficiency Ratio) which is mathematically the BTU output/Wh input. To calculate the SEER of the proposed peltier cooler we would convert 80Wh to 273BTU, then divide 273BTU/135W = 2 SEER. A typical home HVAC system has a 10 SEER to 16 SEER rating. A resistive heater that converts 100% of the electricity to heat has a 3.4 SEER.

Anyway, the projected cost of operation of the fermenter cannot exceed the afore mentioned $16.41/month, and in all but the most extreme conditions (for example storing the fermenter in a 120F attic and attempting to maintain lagering temps) should cost less than $5/month. The advantage to using a peltier cooler is the relatively low cost to mfg a specialty device like a chilled fermentation environment, and the very small footprint of the cooling mechanism (the peltier cooler itself is only 40mm x 40mm x 3mm (1.5" x 1.5" x 1/8")).

Fish

shotgunner · 01-09-2012, 04:02 AM

Subscribed. I'm in for the open source version. I have been looking at peltier units for a long time. Planning on arduino control and therefore logging??

**fish4fun** · 01-09-2012, 12:04 PM

Shotgunner,

I plan to use an ATMega32 for my prototyping. The firmware is in AVR-ASM, so it would be fairly trivial to port the firmware to any of the Arduino's, but I would leave that task to others in an open source project; I do, however, plan to develop the TEC power supply/current controller in two phases: 1) Using a standard PC ATX power supply 2) Using a custom built power supply. Assuming the prototype testing goes as expected, I plan on building several of these for myself; from a purely aesthetic point-of-view I would prefer not to have PC power supplies cluttering things up.

I honestly expected a bit more interest in this project, but I will document the progress of my build/testing here, and perhaps as things progress there will be more interest.

The time-line for initial development and testing should span the next 4 to 12 weeks (a lot depends on available free time and initial testing results). My prototype box is made from 1" and 2" "pink foam" available @ building supply stores. I will use this box for all initial testing of the other components. Assuming the peltier cooler design works as expected, I will then build the mold for Sankey Keg sized fermenter boxes made from pourable foam and begin refining the physical integration of the components with the box. I have to be out of town on business a big chunk of the next two weeks, but all of the various components should be here when I get back, and I will have some time to work on the firmware while I am gone, so hopefully it won't be too long before I have some pictures, figures and conclusions to post.

Thanks for reading!

Fish

**muthafuggle** · 01-09-2012, 01:04 PM

Just out of curiosity, why sanke keg size? Unless you are talking about commercial beer , I don't know anyone who uses sanke kegs as kegs.

Corny kegs or 5 gallon carboys or 6.5 gallon acid carboys would all be more applicable than a sanke keg for most of us here.

**fish4fun** · 01-09-2012, 01:56 PM

muthafuggle,

For fermentation I like doing 10-11 gallon batches, hence a 15.5 gallon fermenter. I keg into 1/6th bbl Sankey kegs, so each batch fills two 1/6th bbl kegs. As far as the project goes, the electronics, heat sinks/fans are the same, obviously for a 6 gallon carboy or bucket the box itself could be smaller. I do plan to build a mold for a 6 gallon carboy for a brewing buddy who bottles (if everything else works out as planned).

Fish

airving · 01-09-2012, 08:18 PM

Peltiers are cool ( no pun intended) but I suspect the subdued response might be from prior experience with them here. A number of other attempts to make use of them have met with less than stellar success, mine included. At some point I couldn't justify spending more effort trying to get it to work when a compressor based solution is so inexpensive.

Are you planning cooling liquid to pump around or through the fermentation vessel, mounting the peltier directly on the fermenter, or creating a cold air chamber?

Not to discourage you - I fully recognize that a good part of the enjoyment of this hobby for some folks (I'm in that bucket) is in the tinkering and the gadgetry. And for full disclosure, I'm neither an electrical engineer nor a refrigeration expert, so my attempts may have been ill fated from the start.

shotgunner · 01-09-2012, 08:20 PM

Quote:

Originally Posted by fish4fun

Shotgunner,

I plan to use an ATMega32 for my prototyping. The firmware is in AVR-ASM, so it would be fairly trivial to port the firmware to any of the Arduino's, but I would leave that task to others in an open source project; I do, however, plan to develop the TEC power supply/current controller in two phases: 1) Using a standard PC ATX power supply 2) Using a custom built power supply. Assuming the prototype testing goes as expected, I plan on building several of these for myself; from a purely aesthetic point-of-view I would prefer not to have PC power supplies cluttering things up.

fugly!

Quote:

I honestly expected a bit more interest in this project, but I will document the progress of my build/testing here, and perhaps as things progress there will be more interest.

The time-line for initial development and testing should span the next 4 to 12 weeks (a lot depends on available free time and initial testing results). My prototype box is made from 1" and 2" "pink foam" available @ building supply stores. I will use this box for all initial testing of the other components.

great way to prototype.

Quote:

Assuming the peltier cooler design works as expected, I will then build the mold for Sankey Keg sized fermenter boxes made from pourable foam and begin refining the physical integration of the components with the box.
Fish

What density pourable foam? 0.5#, 1#, 2#??

Thanks. I look forward to seeing your efforts. I live in SoCal and the weather here is too warm to reliably ferment without cooling and basements in earthquake country are rare as hens teeth!

**fish4fun** · 01-10-2012, 01:47 AM

airving,

My plan is to use one of these:

http://www.amazon.com/Corsair-Cooling-Hydro-High-Performance-CWCH60/dp/B004MYFOE2/ref=pd_cp_e_1
with the radiator on the outside of the cooler and the peltier cooler and heat sink on the inside of the cooler. The "Hot" side of the peltier cooler will be thermally bonded to the "heat sink" of the liquid cooler. A fan will be mounted to the "cold" side of the peltier cooler. The biggest problem with peltier coolers is that as soon as you remove power from them the two sides quickly equalize in temperature. While "off", if one side of the peltier cooler is directly coupled to ambient while the other side is directly coupled to the controlled environment then the TEC (ThermoElectric Cooler) becomes a TEG (ThermoElectric Generator) as long as there is a temperature difference between the controlled environment and the ambient environment; while the amount of power generated is trivial, it is basically like "opening a window" to the controlled environment. Anyway, the bottom line is that using a liquid media to remove the heat from inside the controlled environment is the only practical solution. The above system is what I have ordered as a "starting place". If you used a similar approach I would be interested to read about your findings/problems.

Shotgunner,

Quote:

fugly!]
.
.
.
great way to prototype.

Thanks?

I plan to start with 2# foam. ( Urethane Foam , Expanding Marine Polyurethane Foam ) I plan to use 0.010" sheet aluminum as outside and inside "sheathing" with the foam "binding them". Imagine two sheet aluminum cylinders one 6 inches in diameter smaller than the other and placed "inside" the other. This will provide room for a 3 inch foam layer between them. I am uncertain if the 2# foam will provide enough structural integrity, but I suspect it will.

Fish