Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Temperature related to Current flow on BBr or cables

  1. Jul 7, 2017 #1
    Temperature with hot spot by Ir thermal cameras of modern development is being used
    for measuring on the object of terminal surfaces and cable working temperatures in
    particular working loads. However, so far no measurements has been designed to
    obtain the result on the flow of current corresponding to this spotted temperature
    of on objects viz; cables, Breakers , or Bus buars of Panels and distribution boards
    in electrical system. If this is achieved by detecting of the IR cameras,(without any C.Ts )
    most of the accidents of technicians in the field and I feel the best solutions for human safety
    in electrical system.
    If any one can indicate me the short cut formulae in relations with temperature and corresponding
    ampere flow of the Electrical equipment, it would be much appreciated.

    D.Nagh
     
  2. jcsd
  3. Jul 7, 2017 #2

    berkeman

    User Avatar

    Staff: Mentor

    Welcome to the PF. :smile:
    What do you mean to protect people from current? Non-contact AC voltage detectors are already used to help technicians know when equipment is energized. And if the current paths are well designed, they should not be getting hot when current is flowing...
     
  4. Jul 7, 2017 #3
    How would using IR thermography rather than current transformers improve safety?
    Power = Volts * Amps, and Power = Amps2 * R.

    A 10 micro-ohm wire lug-to-busbar connection with 1000 amps flowing through it dissipates 10 watts.

    That 10 watts increases lug temperature, but how hot it becomes depends on a number of factors - heat sink (in this case, provided by the bus bar and cable), air velocity (convective cooling rate), the possibly larger dissipation created by the wire-to-lug connection, and so on.

    I'm unaware of any shortcuts.
     
  5. Jul 7, 2017 #4
    Dear Mr. Berkman,
    The NDT is the system to avoid most of the physical interactions on the live system inspections in the electrical field
    during operations. despite the norms, laws governing cautions to shut down the power prior to physical checkings ,
    some of the technicians, do operate by risking the checking components in Electrical panels, do not switch off the
    power prior to opening and inspect.

    Such cases the instrument like iR thermal imagers are helpful to spot defects on the components which practically
    in the stage of loose connections by various reasons, which creates the spark (Arc Flashes) any point of time during
    under the load conditions. To achieve the prevention of risk during inspections, vitally requires to read the Load current
    as well as the hot spot temperatures both for prediction of activities on the components, for the safety aspects for the operator.
    This can avoid the hooking up the Clamp on meter to check the unbalance loads of each phases in the Electrical L.V
    system distributions. Hence the question was put.
     
  6. Jul 7, 2017 #5

    berkeman

    User Avatar

    Staff: Mentor

  7. Jul 7, 2017 #6
    Dear Mr. Berkman,
    The writer is a professional IR thermographer LVl II and well under stood
    the system and operations using many models of the camera from Flir.
     
  8. Jul 7, 2017 #7
    But it can't. Current isn't the only thing involved in temperature rise.
    Power = Amps2 * R.
    The amount of heating increases in proportion to contact resistance.

    Line current measurements (more effective when combined with a series of voltage drop measurements) are useful in narrowing the field of what is causing the temperature increase, but aren't strictly necessary to know that a problem exists.
     
  9. Jul 7, 2017 #8
    Understood well the current is not only source of heat in electrical system. However, it is appreciated ,if explained to know how the current to measure through NDT method,corresponding to the temperature measurement, would be much appreciated. Any loose connections on electrical installation will definitely,
    causes the Power loss, due to voltage drop in the system.
     
  10. Jul 7, 2017 #9
    Is the definition of NDT as used above "Non Destructive Testing"?
    I don't understand what you are asking. Could describe a typical troubleshooting situation?
     
  11. Jul 7, 2017 #10
    Yes NDT means Non Destructive Testing Method. In Electrical Inspection, the latest application of
    IR thermal Imager to locate hot spots to predict due Overload or Loose connections i or faulty installations
    However, all we get the hot spot by assessing the differential temperatures with reference temperatures of ambients.
    We may not be able assess the exact load currents w.r.t temperature readings in the cameras. Always there is
    a relation in temperature , Current and thee resistance in the system. If we could achieve the result f his combined
    readings , will certainly solve many practical issues.
     
  12. Jul 7, 2017 #11

    berkeman

    User Avatar

    Staff: Mentor

    Fair enough. Please just be careful not to do any advertising of services here. Thanks. :smile:
     
  13. Jul 7, 2017 #12
    There is no adv. services. it is a question asked whether any related Formulae one can make
    in relation to the Current , temperature and the resistivity of the Conductor. Thks. for
    your kind suggestion and information.
     
  14. Jul 7, 2017 #13

    Baluncore

    User Avatar
    Science Advisor

    Power loss to the end user, due to poor connections, results in explosive heating of the faulty connections. It is not necessary to use IR to detect that very rare situation.

    Distribution of three phase power provides groups of three connectors that should be at similar temperatures. IR has been used for several years to remotely identify poor insulators and bad connections on transmission towers. Since power systems are fitted with CTs when constructed, there is no need to measure the currents again.

    I see no problem that needs a solution.
     
  15. Jul 7, 2017 #14
    If certainly would solve these issues, except I know of no practical way to accomplish it by using exclusively non-contact test methods.

    In low voltage enclosures, my practice was to perform an IR scan to find abnormally high temperatures. If any were found, the suspicious component was visually scrutinized (looking for overheating tell-tales such as metal bluing, and insulator discoloration/burning), and IR and visible light photos were taken. Depending on the nature of the component, line currents were measured and recorded, and (particularly if IR visualization was ambiguous, and so long as visual inspection didn't reveal an obviously dangerous situation such as a fuse holder in danger of breaking apart if touched), I'd carefully perform voltage drop tests to focus in on the problem.

    A classic example of voltage drop testing is for an across-the-line, three phase motor starter. It might not be obvious where the heat is coming from by studying the IR photo (for instance, it may be migrating from the thermal overload relay element), but if line currents are roughly equal to one another, a correspondingly larger voltage drop across the contact set of the 'hot' pole in relation to the cooler ones strongly suggests that particular contact set is failing.

    Properly rated test equipment must be used (class IV covers most industrial power circuitry under 1 kV), and suitable arc flash and shock hazard PPE (Personal Protective Equipment) must be worn, but a thermographer ought to be wearing PPE regardless whether they are taking electrical measurements or not.
     
  16. Jul 8, 2017 #15
    Dear Mr. Baluncore,
    Variety of Experiences teach us in many ways than what we actually study of theory ,in any respective field, especially on science.,medical professions
    Sometime our Over confidence e stops us in learning further in our life. Appreciate your very valuable & useful suggestions and answer to my chats.
    Thank you very much.
     
  17. Jul 8, 2017 #16

    Nidum

    User Avatar
    Science Advisor
    Gold Member

    (1) If you can use image data to measure the geometric extents of the hot spots and estimate the mean temperature then you are some way to being able to estimate the heat emission . For a component of known construction this could then lead to a calculation of I2R .

    (2) Tests could be conducted on samples of all types of connector and other components in use to get accurate data for converting the image data into values of I2R without need for calculations .

    It may be possible to produce a chart where technician could compare newly captured images with sample images on the chart to either get the I2R values or just get answers in form of ok / possible problem / fail

    Images would ideally have to be captured at a standard distance but I think there would be enough latitude for technician to eyeball the distance .

    (3) It may be possible to feed the images captured into a laptop and use software to perform the calculations and/or comparisons .
     
    Last edited: Jul 8, 2017
  18. Jul 8, 2017 #17
    Dear Mr.Nidum,

    Thks. the suggested answer and appreciate your initiated thinking on my question. I do consider the same.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Temperature related to Current flow on BBr or cables
  1. Heat flow (Replies: 2)

Loading...