4 Replies Latest reply on Dec 19, 2017 6:57 PM by Dale Dell'Ario

    Is Flow pro an effective tool for thermal behavior prediction of LED lamp (bulbs) design alternatives?

    Dale Dell'Ario
    Dale Dell'Ario Dec 18, 2017 7:39 PM

      We are a small startup considering the purchase of Flow Pro (without the electronics package) to help choose between and optimize a few different design approaches I have already modeled in SolidWorks 2016. I have no FEM experience. The purchase price represents what is for us a very large purchase, and I need to become proficient enough and have meaninful results in a matter of days. There is a two 1/2 day training in two weeks with our local VAR I can get in on.  I have built several real world prototypes via 3D printing. I am designing a family of lamps to fit in a variety of fixtures with a variety of orientations and over time we will likely develop other form-factor lamps. I'd like to hear from any users in simiilar applications about easy of learning, ease of use, and accuracy of results. I've watched a number of videos and spoken to one local engineer who uses Flow for fluid flow problems and thinks very highly of it. He is the only reference my VAR can provide, Thanks in advance for your opinions, experience!  -Dale

      • 381 Views
      • Tags:
        • Re: Is Flow pro an effective tool for thermal behavior prediction of LED lamp (bulbs) design alternatives?
          Mark Keown
          Mark Keown Dec 18, 2017 9:04 PM (in response to Dale Dell'Ario)

          Dale,

          Mentor makes Flow Simulation for SW - purchased from Mentor it is called Flo EFD. Flo EFD has an LED module - but you will pay a bit more for the software.  You can get the same solution with Flow Simulation with a few more steps.

          https://www.mentor.com/products/mechanical/floefd/options

           

          For your initial design I would recommend making thermal resistor networks for the solution (using MS excel).  This will let you know if you are in the ball park in seconds and is easier to see where the problem areas are than CFD results.   With a resistor network 'concept' then move to CFD.

           

          I would allow a week to run through the Flow Simulation tutorials.

          Then make some simple samples in the lab (like a single resistor on a PCB) and record results then make CFD model in Flow Simulaton.  Just because the CFD matches the lab test dose not mean that the CFD is correct.

          https://docs.wixstatic.com/ugd/ea3635_835c0ae5a7654ff28fe783889f6f6542.pdf

           

          I can provide very cost competitive simulations for you if you like.

          Mark

          mark@ecoolingsolutions.net

          • Re: Is Flow pro an effective tool for thermal behavior prediction of LED lamp (bulbs) design alternatives?
            Bill McEachern
            Bill McEachern Dec 19, 2017 11:23 AM (in response to Dale Dell'Ario)

            I have been using SWX Flow Sim code since it was introduced in 1999 at SWX World in Palm Springs. The product is well suited to system level electronic cooling applications, including LED's. It does have some issues in high specification environments as the selection manager gets a bit bogged done once you hit say about 70 items which I encountered in a LED panel with many lED's. However you can deal with it by keep your groups to something below that number and other set up techniques. You can buy the add ons as a convenience but the base product can do almost all of what is included in the add on's other than the advanced radiation in the HVAC module and the joule heating and thermo electic coolers in the electronic cooling products. Their might be some other small things I am missing but you can always add then on later if you need them. The less you know about continuum mechanics the better this product is for you as it requires only a good understanding of how the physics works. I have used quite a few flow products from panel codes to various FVM RANS codes. I think the product is brilliant but it has some limitations if you are mainly focused on smooth aerodynamics as the turbulence models are not available for getting it very close in those situations but it will provide a decent, and in my experience, a conservative estimate. It seems like it cost a lot when you know nothing about competitive products. Once you look around it will be a bargain so much so it might blow your confidence but it is a very good product - best one in the SWX line up in my humble opinion. It is hands down the best CD code out there in terms of effort hours to solution for what it is suited for which I would label "industrial aerodynamics".

            I too can offer contract services or training and support. It is quite effective with today's conferencing software.