Rules and Guidelines
Want to participate in the discussions? Please read our Community Rules and Guidelines. Need some help? Visit Support and Answers.

  • State Not Answered
  • Locked Locked
  • Replies 14 replies
  • Subscribers 32 subscribers
  • Views 2396 views
  • Users 0 members are here
Options
You may also be interested in
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Interpolating derating factors for multicore cables

jbrameld
I'm looking at table 4B5 and am musing on interpolation of the de-rating values for differing numbers of cores from those tabulated.  I have an application I have been asked to look at with 16 loaded cores, and if I use the tabulated value for 14 loaded cores I can use the 30% rule, but if I use the value for 19 loaded cores I (just) can't.  At the factor of 19 loaded cores I am .005A the wrong side of less than 30%!


The challenge this gives me is that the OCPD is 10A rated, the design current is 2.826A, the 30% rule based on the multicore derating alone is 2.821A, and the fully de-rated capacity of the cores (allowing for a further group of the multicores in the installation) in the multicore cable cores is 6.11A.  If I can apply the 30% rule so no derating for the multicore cable then we are in the clear.  The cable is already installed and in all other respects calculates out fine.


The graph of derating factor vs number of cores isn't quite linear: 1711d4446663643ccb8ed6c700e2cdc2-huge-image-20210412170544-1.png


It nearly is in my region of interest though!  Does anyone have a formula for this calculation (if there is one?).


I know this is so marginal as to not normally be worth bothering about, but we may be dealing with a particularly picky client organisation here.


Practically, the chances of someone going up a tower, replacing some or all of the 650W lanterns with 2kW lanterns and leaving them all turned on to the point where the cable is damaged is so close to b***** all as to really but be worth considering, but right is right and wrong is wrong!


It's also not so easy to replace the OCPDs with 6A ones, as these are in a theatrical dimmer unit so modifying that would likely void the warranties, never mind the approvals!


Any other creative solutions welcomed.


As a complete aside I wonder how they chose the numbers of cores to put in the table - multiple single phase circuits would always be an even number, as would multiple TPN circuits - only 3-phase delta circuits would get you a multiple of 3 loaded cores so how you get to 7,19 or 37 cores baffles me unless you put part of the circuit in one multicore and part in another which is a bad idea, or it's some kind of control application.

  • 0
    There are important step points in the number of cores depending on how they are arranged. For example 6 cores might be arranged as a hexagonal ring, while 7 cores has 1 inner and a hexagon of 6 outers, so the inner core will have a harder time cooling. Similarly, 19 cores has an inner core, a middle ring of 6 cores  and an outer ring of 12 cores.
  • 0
    It would be unusual for all cores in a multicore to carry significant current.

    One application is stage lighting. A 19 core cable is often used for six circuits, each with its own earth, and one core unused. Therefore only 12 cores can be loaded. The total loading may be restricted by the dimmer capacity or power supply into the dimmer rack. For example 6 dimmer circuits but with only 32 amps in total due to 32 amp OCPD into the dimmer. Also the use is generally short term.


    Other applications are remote control of plant and equipment or remote pilot lamps, loading unlikely to exceed one amp per core.


    Or even telephone circuits, wired in mains voltage cable for robustness, load minute.
  • 0
    Given that heating is cross-section area to length, and the cooling is circumference to length related, I'd expect a near  1/R dependence with all other factors unchanged. It will not be quite right, as larger cables will presumably attract less  convection, and the jacket thickness is probably not scaling with number of cores, so A/R2 and B/R sort of thing is more likely.

    If all the cores are the same diameter and there are no strength cores, then the hexagon packing thing is the nearest we get to close packed circles, hence the core numbering.  You will also see 18 cores and a kevlar or steel core for suspension, and by the time you are commissioning cables with a mix of coax cables and control cores you get some very funny packing factors.

    Mike.
  • 0
    Thanks for the replies - although we are rather side tracked by my aside observation (they are always dangerous....).


    To simplify the real question - if the factor for 14 loaded cores is 0.51 and the factor for 19 loaded cores is 0.45 then what should the factor for 16 loaded cores be?  An interpolation would give me 0.474.  Is that good enough?


    Jason.




  • 0
    multiple single phase circuits would always be an even number, as would multiple TPN circuits

    Not quite - TPN circuits are usually considered to have 3 loaded cores (not 4) - as N only carries the 'imbalance' from the phases - so whatever current N is carrying, the same must be missing from at least one of the phases - so the heating effect should be no worse than all three phases fully loaded. (Harmonics notwithstanding of course).

      - Andy.
  • Former Community Member
    0 Former Community Member
    If you're interpolating the problem shouldn't arise, but I think to keep on track, you mean extrapolating.


    Perhaps not now I think more about it. but still not quite sure!


    Regards


    BOD
  • 0
    jbrameld:

    To simplify the real question - if the factor for 14 loaded cores is 0.51 and the factor for 19 loaded cores is 0.45 then what should the factor for 16 loaded cores be?  An interpolation would give me 0.474.  Is that good enough?


    Forget BED, he's tired. You need to interpolate.


    That said, do you want to do it linearly, geometrically, or exponentially? By my reckoning, linearly, the answer is 0.486 Ω - closer to 14 then 19, but it won't make any significant difference.


  • 0
    At least one cubic polynomial fit that matches the points on either side with an R of 0.996 would give you 0.4769,so things are perhaps a touch better than your simple linear interpolation.

    But calculations to this level of precision  are misleading as the raw data is not that good, and the differences will be lost in the long grass, the variation in ambient temperatures is likely to affect the answer more. Really between 0.47 and 0.48 is all that needs be said.
  • 0
    A Tower? Is it outside and cool running?


    Could a little more current carrying capacity be obtained from say applying table 4B1?


    Z.
  • 0
    Zoomup:

    A Tower? Is it outside and cool running?


    Could a little more current carrying capacity be obtained from say applying table 4B1?


    Z.


    Sadly not - I meant scaffold tower to access lighting bars at high level in a primary school.


Community Rules & Guidelines