I want build a cooler box using two Peltiers (TEC-12706) and have certain set point

Thread Starter

jack rau

Joined Apr 1, 2018
5
As mentioned above , I want build a cooler box where both TECS rated (12V/6A) are connected in series where it is powered up using Lipo battery (11.1V where fully charged will be around 12.5V).And im using a separate battery (another lipo) to power up arduino , motor driver and fan .So in this project , there are few set points available in the range of (2-15C) and I want the cooler reach the set point within 15 mins where the capacity is about 15Litres. In order to control the temperature , i want to use the PID where it controls the current pass thru the peltiers accoding to the set point but ,my point here is ,how to implement this ? I have read somewhere by using mosfet but im not clear abt it. Can anyone help me on this ? Apart from that should I use heatsinks and fans at the cold side of tec or the attach it to the interior of the box which is made up of aluminium sheet , which will be better ?
 

danadak

Joined Mar 10, 2018
4,057
I want the cooler reach the set point within 15 mins where the capacity is about 15Litres
You will have to do thermal calculations taking into account thermal mass, starting temp, thermal capacity
of Peltiers.....Or do just a basic measurement where you simply turn on Peltiers and time how long it takes
to meet target temp.

i want to use the PID
Here is an example using PSOC, principles the same for any micro -

http://www.cypress.com/file/139661/download

http://www.cypress.com/documentation/application-notes/an89346-psoc-4-intelligent-fan-controller

Apart from that should I use heatsinks and fans at the cold side of tec or the attach it to the interior of the box which is made up of aluminium sheet , which will be better ?\
You have to evaluate the thermal R of each approach to determine what approach
will meet the overall criteria for costs, size, power....

Regards, Dana.
 

danadak

Joined Mar 10, 2018
4,057
If you are interested in PSOC both ap notes have an example project
you can download into the IDE tool, PSOC Creator. Creator is also
free -

http://www.cypress.com/products/psoc-creator-integrated-design-environment-ide

Each project has a recommended part. Once you work with your requirements
you can always change the part in the project to meet requirements and cost.

Google "PSOC 101 videos", there is a series 1 - 16 that will get you started.

Regards, Dana.
 

MrChips

Joined Oct 2, 2009
30,621
Connect the TECs in parallel, not series. You will need heat sinks on the hot side of the TEC. The heat sinks and cooling fan has to go on the outside of the box.

You need some way of monitoring the temperature. A thermistor is one way to do this. You need an analog to digital converter and a microcontroller in order to implement the control algorithm.

You do not necessarily need a full PID algorithm. You can use PI or even a simple ON/OFF control. Or you can buy an off-the-shelf temperature controller.
 

OBW0549

Joined Mar 2, 2015
3,566
So in this project , there are few set points available in the range of (2-15C) and I want the cooler reach the set point within 15 mins where the capacity is about 15Litres.
I don't think two TEC modules are going to be able to bring a box with an internal volume of 15 liters down to 2°C in 15 minutes, even if the box is empty; an hour (or even longer) would be more realistic. And if the box contains anything with significant thermal mass, the time to reach that temperature will be a lot longer.
 

Thread Starter

jack rau

Joined Apr 1, 2018
5
Connect the TECs in parallel, not series. You will need heat sinks on the hot side of the TEC. The heat sinks and cooling fan has to go on the outside of the box.

You need some way of monitoring the temperature. A thermistor is one way to do this. You need an analog to digital converter and a microcontroller in order to implement the control algorithm.

You do not necessarily need a full PID algorithm. You can use PI or even a simple ON/OFF control. Or you can buy an off-the-shelf temperature controller.
Thanks for the reply .May I know , how does it make the difference by connecting series or parallel?
 

Thread Starter

jack rau

Joined Apr 1, 2018
5
I don't think two TEC modules are going to be able to bring a box with an internal volume of 15 liters down to 2°C in 15 minutes, even if the box is empty; an hour (or even longer) would be more realistic. And if the box contains anything with significant thermal mass, the time to reach that temperature will be a lot longer.
Thanks for the reply.But is there any other ways for me to achieve the desired temp faster if not 2°C but still I hve other set points .Or what do you think of the size of the box ?
 

wayneh

Joined Sep 9, 2010
17,495
Thanks for the reply.But is there any other ways for me to achieve the desired temp faster if not 2°C but still I hve other set points .Or what do you think of the size of the box ?
In theory, every aspect of this problem can be modeled and simulated before anything is built. It's 'just' thermodynamics, and everything you need to know about the cooler modules can be found in the technical literature, and the properties of the box and material inside that you need to cool are all knowable things. You'd need the mass and heat capacity of the stuff in the box. You then calculate the amount of heat you need to move from one side to the other and the time allowed to accomplish it, and determine how much cooling hardware is required to do all that.

In reality, it's a fairly tough calculation even for someone that knows what they're doing. No offense, but I don't think that includes you. For one thing, the performance of the TEC will depend on the ∆T across it, the current supplied to it, and critically on the heat transfer (sinks, fans) attached to both sides. The hot side will need to dissipate as much as 10X the heat entering the cold side, and so the hot-side sink and fan are usually more elaborate than the cold side. A TEC is more efficient at low current, but cannot reach as large a ∆T at less than full power.

From what you've described so far, I don't think you can do what you want. A single-stage TEC cannot reach to freezing on the cold side while exposed to ambient on the hot side, even with infinite heat dissipation. (In reality the hot side will be hotter than ambient, causing the cold side to be that much warmer.) And when operating near the maximum ∆T, very little heat will actually be moved from the cold side to the hot. Cooling to near freezing may require a two-stage TEC arrangement and a lot of capacity.

To control the TEC, you'll need a temperature sensor and a thermostat. A simple on/off thermostat is probably fine, since that can hold to within a °C or so. Thermostats are pretty easy to build (yes, with a MOSFET as you noted) but you can buy them also. Consider the thermostat as the least of your concerns. Getting enough cooling capacity will be the bigger challenge.
 

MrChips

Joined Oct 2, 2009
30,621
Thanks for the reply .May I know , how does it make the difference by connecting series or parallel?
The TEC-12706 is rated at 6A @ 12V, i.e. 72W of electrical power.
If you connect two TEC-12706 in series, you will need a 24V, 6A power supply to run at max power.
If you connect two in parallel, you will need 12V, 12A power supply to run at max power.
 

OBW0549

Joined Mar 2, 2015
3,566
But is there any other ways for me to achieve the desired temp faster if not 2°C but still I hve other set points .
Sure, there are all kinds of ways; just Google "cooling methods for electronics". You can circulate chilled water through a heat exchanger (e.g., a radiator & fan) inside the box. Or (if you can stand the noise) you can blast super-cold air into the box with a Ranque-Hilsch vortex tube fed by a compressed air source. Or just use more Peltier modules-- a LOT more.

Or what do you think of the size of the box ?
What I think (or what anybody else thinks) of the size of the box is of no importance. That is for you to decide, based on your needs. All I can suggest is to not make the box any bigger than it needs to be, since that would make cooling take longer.
 

Thread Starter

jack rau

Joined Apr 1, 2018
5
Sure, there are all kinds of ways; just Google "cooling methods for electronics". You can circulate chilled water through a heat exchanger (e.g., a radiator & fan) inside the box. Or (if you can stand the noise) you can blast super-cold air into the box with a Ranque-Hilsch vortex tube fed by a compressed air source. Or just use more Peltier modules-- a LOT more.


What I think (or what anybody else thinks) of the size of the box is of no importance. That is for you to decide, based on your needs. All I can suggest is to not make the box any bigger than it needs to be, since that would make cooling take longer.
Thnx for the reply .
 
Top