11
\$\begingroup\$

I built a box from 5cm thick Styrofoam with outer dimensions 60x55x80cm. I wanted to use it as a fermentation chamber for home brew purposes. The reason why I did not choose regular (some used cheap) refrigerator is I need to have it in a small chamber with floor dimensions of 80x130cm. Also door of that chamber is only 55cm wide. So getting inside some refrigerator would be most likely impossible.

I decided to buy 2 peltier modules - TEC-12715 and attached them to a "hot side heatsink" of dimensions 23x17x4cm and "cold side heatsink" of dimensions 10x17x5cm. I am powering each separately from the dedicated ATX power supply which each can handle 18A max at 12V, so there is a reserve. Hot side is being cooled by 5 fans removed from old cases and atx power supplies. The smaller, cool heatsink is inserted into the top part of the box and sealed. On the cold heatsink there is one 12cm fan.

Unfortunately I can only get max of approx 18 degree Celsius inside (35L barrel with approx 20L of a beer inside + 4x 1.5L bottles with water just as a accumulation of the "cold"). Hot heatsink has most of the time 50 degrees Celsius (I have a temp sensor sticked on it's middle part (near to the peltiers which are also in the middle)). Ambient temperature is now 29 degrees Celsius. So it is a difference of max 11 degrees of Celsius.

Should I consider this to be "normal" when considering inefficiency of peltier modules? Or should I be able to squeeze more cold inside? I read somewhere on the internet the peltier's efficiency is terrible once the hot side is over 40 degrees, but I am not sure if this is true. I planned to do some temp regulation as well, but as I wanted to get closer to 15 degrees Celsius, this does not have the reason now.

Any ideas, suggestions?

\$\endgroup\$
4
  • 4
    \$\begingroup\$ For fun? maybe. But a compressor based cooler would use about 1/3 as much energy for the same cooling, and that means about half the heat to get rid of. There are 50cm and smaller refrigerators for boats and motorhomes or you can build your own around a compressor : search "Danfoss BD35" as a starting point \$\endgroup\$
    – user16324
    Commented Apr 4, 2016 at 14:14
  • \$\begingroup\$ The maximum ΔT of your elements is 70℃. Stick thermal sensors on hot and cold sides (below the sinks) to see how much of this ΔT you're seeing. \$\endgroup\$
    – Dmitry Grigoryev
    Commented Apr 4, 2016 at 14:33
  • \$\begingroup\$ @Dmitry Grigoryev: I currently do not have any spare temperature sensor and I have to complete current fermentation, so I will try this later. \$\endgroup\$
    – user241281
    Commented Apr 4, 2016 at 18:25
  • 4
    \$\begingroup\$ Do not forget the yeast is dumping quite a bit of heat into the system. See homebrewtalk.com/showthread.php?t=145767 - 20l beer will output several Watts depending on the fermentation speed. Got to measure the temp difference with the next beer batch, never thought of it. When making wine, the temp difference was very noticable with bare hands - but wine has about twice the sugar to burn. \$\endgroup\$
    – Posipiet
    Commented Apr 4, 2016 at 22:03

5 Answers 5

Reset to default
6
\$\begingroup\$

It would be worth while to try and reduce the temperature of the hot side. The biggest heat leak into your fridge is going to be through the TEC. The hotter it is the more heat it leaks.. and the more it has to work to get that heat back out. The "classic" mistake with TEC coolers is to make the hot side heat sink too small the first time. (Welcome to the club :^)

\$\endgroup\$
5
  • 3
    \$\begingroup\$ I did an experiment - I switched the power supply from 12V to 5V, just to see what happens. Interesting is the hot side heatsink has now around 30 C and after 8 hours the temperature inside of the cooler is 17.4 C (a sensor not directly attached to the fermentation vessel) and the beer has 19 C (a sensor directly attached to the fermentation vessel). So it looks I am able to get similar, if not better results with using 5V instead of 12V as a power source for the peltier modules. Maybe this is because the fact I am not "overheating" the hot side and whole room. \$\endgroup\$
    – user241281
    Commented Apr 5, 2016 at 6:05
  • \$\begingroup\$ It looks it even if powered with 5V, I still need more cooling power. Currently the beer has 19.8 C, air in the box had 18.5 C and the hot heatsink has 32 C, room has 25 C. One more idea - I have 2 spare peltier modules TEC-12706 with attached heatsinks and I could add them to the box and power them as well with the 5V. Do you think this would add reasonable amount of cooling power? I would need to have ~15 C in the box. Switching back to the 12V will not solve the problem as there is huge amount of heat produced and I am getting over 30 C in the room. \$\endgroup\$
    – user241281
    Commented Apr 5, 2016 at 11:36
  • \$\begingroup\$ @user241281 Well, perhaps there is a maximum cooling point somewhere between 5V and 12V? \$\endgroup\$
    – Dmitry Grigoryev
    Commented Apr 5, 2016 at 11:51
  • \$\begingroup\$ @DmitryGrigoryev Yes, maybe. Unfortunately I currently do not have the possibility for regulating voltage (or the current). So further experimenting can be done once I have the regulation. \$\endgroup\$
    – user241281
    Commented Apr 5, 2016 at 13:17
  • 2
    \$\begingroup\$ @user241281, with too small a heat sink the whole TEC turns into a heater and melts itself. So the first thing I'd do is get more air cooling or a bigger heat sink for your one. If you add more... (you could add the second in series and run them both off 12V... 6V each) then you are also increasing the thermal path from the outside. \$\endgroup\$
    – George Herold
    Commented Apr 5, 2016 at 16:26
3
\$\begingroup\$

You might try looking at what commercial units do. Here, for instance is a commercially available TEC cooler chest. It's smaller than yours, but it only draws 3.5 amps at 12 volts. Note that it is specified for a 36 F (20 C) temperature differential.

You don't say how long you've run your cooler, but be aware that 26 liters of water takes a lot of cooling. As per the Amazon link, you should wait at least 24 hours before making your measurements.

\$\endgroup\$
4
  • \$\begingroup\$ The beer's temperature was exactly 16.2 degrees Celsius when I put it into the fermentation chamber. Water in the bottles had ~15 degrees Celsius. And it's been over 12 hours since I started the cooler. So the peltier did not have to cool down the stuff, just to keep it cold. Fermentation produces some heat as well - but I am not sure how much) \$\endgroup\$
    – user241281
    Commented Apr 4, 2016 at 18:30
  • \$\begingroup\$ Fermentation can produce quite a lot of heat. A better "test" of the TEC would be with plain water in there. To save your beer, half-fill plastic bottles with water, freeze them, and swap them with the bottles you have in there now (half-filling and freezing them in the meantime.) Small compressor-based units are easily available and far more efficient than TECs. The waste heat from the TEC will be heating up your closet, which makes it more inefficient, which makes it heat up more.... \$\endgroup\$
    – Ecnerwal
    Commented Apr 4, 2016 at 19:05
  • \$\begingroup\$ @Ecnerwal: do you have an example of such "small compressor unit"? \$\endgroup\$
    – user241281
    Commented Apr 4, 2016 at 21:08
  • \$\begingroup\$ In general, what is typically sold as a "dormitory refrigerator", a "Bar refrigerator" or a mini-refrigerator will all have examples that easily pass through a 55 cm doorway. The "dormitory" variant is often obtained inexpensively at the end of college or university terms when students move out, and can be cut open (carefully, not to disturb the refrigerant system) to get the relevant guts out - or you just take the door off and mate it to your insulated box. Other examples were given above. \$\endgroup\$
    – Ecnerwal
    Commented Apr 5, 2016 at 16:33
2
\$\begingroup\$

I notice no one has mentioned using water cooling for hot side, my best wee chilly bin, uses a peltier at each end, with water cooling via a water pump, and small radiator and fan. Will achieve a inside temp of 27 deg below ambient, while full of beer.

\$\endgroup\$
1
\$\begingroup\$

I am using a peltier in an old mini fridge. It’s working well; it keeps freezing the inner cold sink.

I’m using a small 120V fan from Walmart on the hot side, and a little computer fan inside. It gets cold until it builds up too much ice, after which I have to turn it off for about 30 min.

Remember that the more heat you dissipate the colder it will get; less moisture also means greater cooling.

\$\endgroup\$
4
  • 2
    \$\begingroup\$ Welcome to EE.SE, Mike, but this is a question and answer site, not a chat forum. It's not clear if this is supposed to be an answer to the question at the top of the page. Is it? \$\endgroup\$
    – Transistor
    Commented Jan 13, 2019 at 14:12
  • \$\begingroup\$ Yes it is an answer to his question about wanting to get colder and how I did it by adding a higher velocity fan to the heat sync. But informing him in doing so the next issue he may face in doing so. With the possibility of someone correcting me with another method that may help him farther. It’s my first time posting anything on any site let alone yours. I apologize I will attempt to answer more directly next time. \$\endgroup\$
    – Mike the mechanic
    Commented Jan 15, 2019 at 21:58
  • 1
    \$\begingroup\$ That's fine, Mike. Just a suggestion, but you, like many younger writers, leave out the grammatical subject of sentences and this can cause confusion (as well as being bad grammar). In your answer's last sentence it's not clear whether you mean "I need to find a way ..." or "You need to find a way ...". One other confusion: you have written "cold sync". Did you mean "cold sink"? i.e. somewhere to sink the removed heat? I've never used a Peltier device. \$\endgroup\$
    – Transistor
    Commented Jan 15, 2019 at 22:37
  • \$\begingroup\$ Well my answer isn't all that helpful or technical if you would like to remove it all it wouldn't hurt my feelings. I never actually answer these things usually anyway I just thought I would share what I have learned because he is having the same problem... And the peltier device has a hot side and a cold side. My setup uses a heat sink or sync I'm not sure of the correct spelling on the hot and cold side. And I've found that adding a better fan to disapate the heat caused the cold side that also uses a sink or sync and fan to get colder... And actually freeze \$\endgroup\$
    – Mike the mechanic
    Commented Jan 16, 2019 at 23:32
1
\$\begingroup\$

I have attached two Peltier (TEC12706) modules to an 11L icebox. They draw around 8 Amps of the current in total. The temperature inside the box reaches 10 degrees C within an hour (with four 1L bottles) and to a minimum of 1 degree C after waiting for two hours. The room temperature is 30C. I use the water cooling solution. You can compare these results with your setup and find answer yourself.

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.