Simulation Forums

Find Simulation answers, share expertise, and connect with your peers

Simulation CFD


TEC (Peltier) Transient Study

121 Views, 9 Replies
12/19/2017 9:45 PM

Hello,

 

I am looking to simulate a transient study with a TEC device. I have the device modeled and working well in a steady state situation. I would like to model a transient solution where I can turn the TEC on and off at specified moments in time and see how the device performs. Can this be done? Can the TEC be turned "on" and "Off" transiently? Or is my best bet to perform a piecewise simulation and solve it in steps with the device in on and off states?


Thanks,

 

Grant 


Report Inappropriate Content
Message 1 of 9 ( Views: 121 )

Re: TEC (Peltier) Transient Study

12/20/2017 8:33 AM

Good morning Grant,

 

Unfortunately we cannot simulate a TEC being turned on and off during a transient run.

 

A couple of steady state solutions should give you what your looking for. In my opinion for this case running transiently wouldn't give you anything other than an increased run time and larger result files!

 

Hope this helps,

David


David Short
Technical Support Specialist, Simulation

Report Inappropriate Content
Message 2 of 9 ( Views: 107 )

Re: TEC (Peltier) Transient Study

12/20/2017 2:33 PM

David, 

 


Thanks for the quick reply. I figured that. However, I DO NOT want to run a steady state simulation for this as in real use the Peltier will be cycled to achieve temperature on the cool side. I am trying to investigate how long the Peltier needs to be "on" to achieve that from a base temperature to a target temperature. The Peltier Module we are using as negligible thermal mass so I am not worried about the TEC model not having thermal mass. What I would like to do now is start at a transient simulation with the Peltier "on" and investigate the temperature change over time in my system to see if the Pelier we are considering has enough power to drive the change in temperature we need in the amount of time we need it. Is this possible or is the TEC material completely unable to run anything accurately in transient? If the Peltier is unable to work correctly in transient do you have a suggestion for how to model this system? Which boundary conditions could I drive to simulate this Peltier set up? I feel like purely driving the temperature differential on the TEC would over simplify the study and not inform me as to if the wattage needed in the Peltier is over its design. Any ideas are welcome!


Thanks,

 

 

Grant 


Report Inappropriate Content
Message 3 of 9 ( Views: 99 )

Re: TEC (Peltier) Transient Study

12/20/2017 3:01 PM

Hi Grant,

 

The TEC will work absolutely fine in a transient run. You just can't vary its settings during the run. 

 

Your idea to simulate transiently with TEC 'on' to see if the TEC settings are sufficient to drive a temperature change in a given time sounds good to me.

 

Good luck with the simulation,

David


David Short
Technical Support Specialist, Simulation

Report Inappropriate Content
Message 4 of 9 ( Views: 94 )

Re: TEC (Peltier) Transient Study

12/20/2017 3:35 PM

David, 



Thanks! This should allow me to find the information I'm looking for. 

Best, 


Grant 


Report Inappropriate Content
Message 5 of 9 ( Views: 87 )

Re: TEC (Peltier) Transient Study

04/27/2018 5:41 AM

Hi,

Is it possible to simulate TEC "on" for example 10 s ? and after that "off" and log information about voltage or current as running as TEC Electric Power Generator ?  Is it possible to simulate and log Voltage and Power working TEC as Electricity Generator ( after "off" as Heat pump ? )

 

Thank you

 

M.G


Report Inappropriate Content
Message 6 of 9 ( Views: 49 )

Re: TEC (Peltier) Transient Study

04/30/2018 6:31 PM

M.G.,

 

This is sort of possible. What you need to do is treat it like a piecewise step function as you cannot change the TEC parameters mid solution. Each different TEC step has to be it's own solution but you can set the environment to previously solved for values at each step. However you can't always use the previously solved solution directly as the simulation will crash depending on what TEC control values you are using. 

 

For example:

If you had a simulation where a TEC was heating a module on SIDE A and cooling a heatsink on side B for 10 seconds and wanted to reverse the heat/cool side for 10 seconds you would first solve a transient solution for heating SIDE A. When that is complete you can then change the TEC control parameter and hit solve from time step (What ever time step you ended on). However, this may crash depending on the TEC control variable. If this crashes then you must set up a separate simulation and set the temperatures of the module and heatsink as initial conditions manually using averages from the previous simulation. I hope this makes sense. Let me know if you need help setting this up. 

 

Best, 

 

Grant Chapman

VP of Operations

Glassboard Product Development 

 


Report Inappropriate Content
Message 7 of 9 ( Views: 38 )

Re: TEC (Peltier) Transient Study

04/30/2018 10:12 PM
Hi.
Thanks your help.
I need simulate TEC as Heat pump - normal mode as TEC and After 10 s i
want stop supply voltage, and after that i need to log voltage from TEC
working as
Electric Power Generator. So: 10 s works like normal TEC, and after 10s
works as Electric Power Genertor - i need plot voltage changes in time
for Genereting Electricity Mode. Is it possible to log voltage changes in
Power - electricity Generating Mode ?
I need solve this unstationary condition,

Thank you

M.G

Report Inappropriate Content
Message 8 of 9 ( Views: 36 )

Re: TEC (Peltier) Transient Study

04/30/2018 10:26 PM

M.G., 

 

No matter how you solve this you will need 2 studies as you can't turn off the TEC. 

 

So for your first study it should be very straight forward. Set up the TEC using one of the control techniques, lets assume you want to control the power to the TEC, (I am assuming during the 10 second first step you will either control Voltage, Current, or Power in Watts). So set up the TEC material and use the Power control method and enter in a number that you'd like to use. Also make sure to correctly define the TEC Material properties with information from the manufacturer.... the built in ones in CFD aren't always going to behave like the one you want to buy. Set up the rest of the simulation as you would, ensuring to not but a boundary condition directly on any of the TEC surfaces at that tends to crash things. Then hit the solve button. 

 

When that is done solving you are going to use the results to set up your next study. Whatever temperature you're different devices connected to your TEC got to will be the initial conditions for your next simulation..... Now this is where it gets tricky as I have never done this. There is no way to just set the TEC to act as a generator. But you can set it to control either of its side for temperature. I would then set the control method to control for temperature on the side of the TEC you know has a larger thermal mass and good conduction. Then set the temperature to whatever the end temperature was on the last simulation you ran. If this is done correctly step one of the 2nd simulation should not require power from the TEC as its surface it at least at temperature or already exceeded what it needs to meet. You can then see the Voltage and Current in the results section under the Parts task. 

 

I'm not sure this will work. But its the closest thing I could think of. If you are still needing help modeling your solution let me know. 

 

Best, 

 

Grant Chapman 


Report Inappropriate Content
Message 9 of 9 ( Views: 34 )