AnsweredAssumed Answered

Active control of thermoelectric cooler?

Question asked by Michael Bush on Jul 30, 2013
Latest reply on Aug 2, 2013 by Jared Conway

Like • Show 0 Likes0
Comment • 16

Hi-

I'm trying to model a system where the power draw of the TEC is driven by a PID loop, such that the chip temperature is held constant. Seems as though a goal dependent dependency would be just the thing-- the help files discuss using such a beast, but it's not clear whether this functionality is available for the TEC element in flow simulation.

I've tried inputting an equation based on a SurfaceGoal, but it gets spit out as being "wrong number format"-- the syntax I tried is "=0.5+1.5 * (SG_1 - 293 )^2 / 40^2" , where SG_1 is a surface goal that shows up in the project tree.

Am I just messing up the syntax, or am I attempting the impossible?

Thanks,

Michael

Jared Conway Aug 1, 2013 8:48 PM

Correct Answer

checked your model, looks like a limitation, if you go to the surface heat gen and click on the fx, you can see that you can select the goal and then click "goal" to plug it into the equation. so it seems for thermo electric coolers, it can only be a constant.

See the reply in context

No one else had this question

Outcomes

16 Replies

Chris Michalski Jul 30, 2013 3:39 PM
Mark Correct
Correct Answer
Did you type in SG_1 or click on that goal from the tree?
Like • Show 0 Likes0
Actions
- Michael Bush @ Chris Michalski on Jul 30, 2013 3:42 PM
  Mark Correct
  Correct Answer
  typed it. Just now tried clicking on it; no luck-- the color of the bar at the top of the window pulses, but nothing shows up in the window
  Like • Show 0 Likes0
  Actions
  - Chris Michalski @ Michael Bush on Jul 30, 2013 3:46 PM
    Mark Correct
    Correct Answer
    And that goal only contains a single temperature, no other values?
    Like • Show 0 Likes0
    Actions
    - Michael Bush @ Chris Michalski on Jul 30, 2013 3:50 PM
      Mark Correct
      Correct Answer
      correct-- see attached screen capture
      Like • Show 0 Likes0
      Actions
      - Michael Bush @ Michael Bush on Jul 30, 2013 5:17 PM
        Mark Correct
        Correct Answer
        sadly, there's an open enhancement request (SPR 618617) that asks for this functionality-- maybe not any time soon
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Michael Bush on Jul 30, 2013 6:06 PM
        Mark Correct
        Correct Answer
        michael, can you post the dialog where you're trying to type the equation? did you click depencncy to get into it?
        
        i find if equations aren't working for me i try something simple like multipling the temp by 4. if it works, then i know it is a syntax problem vs a functionality issue.
        
        what do you want to learn from the overall analysis? maybe there is another way to do it.
        Like • Show 0 Likes0
        Actions
        Michael Bush @ Jared Conway on Jul 31, 2013 9:04 AM
        Mark Correct
        Correct Answer
        Screen capture below.... I've tried it with and without quotes around SG_1
        I'm attempting to model overall system thermal properties-- the waste heat from the TEC is a significant contributor, and is relatively sensitive to the temperature of the laser diode... I can do the analysis iteratively (set a TEC current, run the analysis to see where the diode ends up, adjust the current slightly, run again until the diode temp converges on what I want....), but it's time consuming....
        
        Like • Show 0 Likes0
        Actions
        Chris Michalski @ Michael Bush on Jul 31, 2013 9:08 AM
        Mark Correct
        Correct Answer
        I haven't worked with the TECs so bear with me - do you input the current to control them? it shows 1.1A in the box. If so then the problem is likely that you are using a temperature (Deg C) in the equation and the units don't match.
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Michael Bush on Jul 31, 2013 11:52 AM
        Mark Correct
        Correct Answer
        if you have the ability to choose dependency, it should work.
        
        for creating your equation, just click on the goal in the feature tree to add it to your equation. that should add it with the right syntax.
        
        if you try a simple equation, does it work? what version/service pack are you using?
        
        Chris: Good call on units. However I THINK that solidworks considers goals without units. They are just numbers. I think.
        Like • Show 0 Likes0
        Actions
        Michael Bush @ Jared Conway on Jul 31, 2013 12:01 PM
        Mark Correct
        Correct Answer
        see above.... clicking on the goal does nothing other than pulse the color of the window
        A simple equation works, but of course isn't useful....
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Michael Bush on Jul 31, 2013 4:28 PM
        Mark Correct
        Correct Answer
        version/service pack? any chance you can share the model? jared@hawkridgesys.com
        Like • Show 0 Likes0
        Actions
        Michael Bush @ Jared Conway on Jul 31, 2013 4:42 PM
        Mark Correct
        Correct Answer
        2012x64, SP4 Sorry, can't post the model
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Michael Bush on Jul 31, 2013 4:44 PM
        Mark Correct
        Correct Answer
        coiuld you build a simplified model that show the same behavior? i'll check to see if 2013 sp4.0 helps at all or if there is anything in 2012 that you can do differently. based on what you're showing, it should work
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Jared Conway on Aug 1, 2013 8:48 PM
        Unmark Correct
        Correct Answer
        checked your model, looks like a limitation, if you go to the surface heat gen and click on the fx, you can see that you can select the goal and then click "goal" to plug it into the equation. so it seems for thermo electric coolers, it can only be a constant.
        Like • Show 0 Likes0
        Actions
        Michael Bush @ Jared Conway on Aug 2, 2013 10:05 AM
        Mark Correct
        Correct Answer
        Bummer-- that's a pretty big limitation; I don't know of a TEC in existence that operates open loop; makes for a simulation that can't be trusted. Time to lobby for another seat of Ansys.
        
        Thanks anyway
        Like • Show 0 Likes0
        Actions
        Jared Conway @ Michael Bush on Aug 2, 2013 11:04 AM
        Mark Correct
        Correct Answer
        does CFX have a smart part for TECs? i wonder how they implement transience.
        
        probably what i'd do is double check with your reseller about that limitation. they can connect with solidworks and the flow simulation development team to get the "final answer" on this. also, they may have a workaround for it. and better yet, they should be able to help you escalate the enhancement to get into a version of solidworks earlier.
        Like • Show 0 Likes0
        Actions

Active control of thermoelectric cooler?

Outcomes

Related Content

Recommended Content