• Welcome to engineeringclicks.com
  • Testing the impossible

    Discussion in 'The main mechanical design forum' started by tcs106, Sep 24, 2011.

    1. tcs106

      tcs106 New Member

      Joined:
      Sep 2011
      Posts:
      2
      Likes Received:
      0
      Dear All,

      This post is not directly related to a design issue and is more targeted to general mechanical or thermal engineers.

      I need to test a part which is subjected to high thermal loads during service. The upper face of the part is situated in the path of a hot gas flow whereas the bottom face of the part is continuously being cooled by cold gas during service. Due to the uneven thermal loading, the part will bend outwardly from its radial direction and I would need to find a way to test the part in these conditions in a test rig to inspect this effect.

      We currently have a wide-range of furnaces available but as you know it is very difficult to be bringing cold air in a furnace running at over 1200C.

      I welcome any suggestions from any engineers who faced similar issues in the past.

      Any other suggestions are of course more than welcome!

      Thanks for your time!
       
    2.  
    3. AndrewNew

      AndrewNew Well-Known Member

      Joined:
      Jul 2010
      Posts:
      73
      Likes Received:
      0
      Could you use hot and cold gas streams in an open lab (rather than in a furnace)? E.g. by directing a blowlamp on to the upper surface and a compressed air jet on the lower? You could play around with the burner/jet configuration if you need a particular heating pattern (a blowlamp would give you a fairly localised heating effect I would imagine). You could hire a thermal imaging camera if you need/want to validate the temperature distribution in the part.

      Cheers

      Andrew
       
    4. dakeb

      dakeb Member

      Joined:
      Sep 2011
      Posts:
      13
      Likes Received:
      0
      You cannot do this properly and cheaply.

      You will need to design and build a device that simulates the installation in use, and add various types of monitoring gauges.

      Alternatively, actually install the things and run them in situ, measure performance, then remove them for examination.
       
    5. AndrewNew

      AndrewNew Well-Known Member

      Joined:
      Jul 2010
      Posts:
      73
      Likes Received:
      0
      Why can't you do it "properly and cheaply"? Providing you represent the service conditions there is benefit in simplifying to the lab condition (reproducibility of conditions, for one). This is the way nearly all R&D is done.
       
    6. srdfmc

      srdfmc Well-Known Member

      Joined:
      Jun 2011
      Posts:
      114
      Likes Received:
      0
      Well, it looks like as if the good old venturi will play its part again. ;)

      On way cold air sucked
      On the other the hot hair is blown
      The more the temp raise the colder you'd get !

      Obviously a well dimensioned system needs to be studied carefully if you don't want to cut some metal endlessly. I am available :mrgreen: !
       
    7. Michael Davis

      Michael Davis Member

      Joined:
      Aug 2011
      Posts:
      14
      Likes Received:
      0
      :?: Sorry, I have a stupid question. What are you testing for? Temp? deflection? strength? :?:
       
    8. tcs106

      tcs106 New Member

      Joined:
      Sep 2011
      Posts:
      2
      Likes Received:
      0
      Thanks for the replies. We are now thinking of using a burner rig where one face of the part will be heated by the flame while the other face is cooled by external cooling tubes. The main challenge will be to know exactly what the temperature is on each face.

      I believe we will use a variation of pyrometers as well as thermocouples to control our experiment and obtain a correct temperature profile.
       
    9. dakeb

      dakeb Member

      Joined:
      Sep 2011
      Posts:
      13
      Likes Received:
      0
      For temperatures around 1200C you could consider an infrared thermometer
       
    10. srdfmc

      srdfmc Well-Known Member

      Joined:
      Jun 2011
      Posts:
      114
      Likes Received:
      0
      I am pretty sure that cheap Thmocpls won't work at that temp (but wld be plsed to know if I'm wrong). Use laser or simple thermography.

      If you are going down to some CFD go for Comsol (expensive). It wld lead you easly trought this design.

      [Joke mode = ON] Btw if you Britts had a good laugh on my little misspelling think how I wld feel when UK wld have o bail the EUro :twisted: [Joke mode = OFF]
       
    11. AndrewNew

      AndrewNew Well-Known Member

      Joined:
      Jul 2010
      Posts:
      73
      Likes Received:
      0
      You should be able to get thermocouples that will survive 1200°C - the limiting factor is usually the capability of the lead wire insulation as I understand it. Have a look at these for example http://www.omega.co.uk/ppt/pptsc.asp?ref=XC&Nav=tema06

      If you want to go the CFD route (which sounds like it might be a very good idea to me, it could save you a lot of time and experimentation in specifying and building your test rig) there are cheaper alternatives to COMSOL. Try http://www.symscape.com/ - this is based on the OpenFOAM open-source CFD code which is pretty well respected in the industry. CFD is not the easiest thing for beginners to master (especially if you want it done properly!) - give me a shout if you want to discuss.

      Good luck.

      Cheers

      Andrew
       

    Share This Page