The IET is carrying out some important updates between 17-30 April and all of our websites will be view only. For more information, read this Announcement

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 16 replies
  • Subscribers 31 subscribers
  • Views 14081 views
  • Users 0 members are here
Options
You may also be interested in
This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

Using RCBO to overcome high Zs but what about short circuits?

Sparkymania
This is something that occurred to me a while ago.

If the Zs is too high for the breaker to operate we ether install a downrated breaker is possible or an RCBO.

My supervisor said something about if an RCBO was installed for this reason then having to test for PSCC at the end of circuit but couldn't say why. He had been told this by NICEIC but no explanation as to why. I argued that this didn't make sense as PFC is only needed at the point where the protective device is and if it's lower at the DB than that of the PD then it will be even lower the further you go down the circuit. I've seen a video that has explained the reason. It's the current flow to trip the breaker on a short circuit rather than an earth fault. Obvious really when you think about it.

However, why would you use a max Zs for line to earth to make sure the resistance allows current flow to trip a breaker for a line to earth fault but not use the same for a line to neutral fault? Surely the impedances would be the same. You already have the max Zs to hand in many tables. With the PSCC you have to calculate what current is needed to trip the breaker or check the tables in the regs. Why do it one way for one and the other way for the other?

Therefore, if you had a C32 MCB with max Zs of 0.54 that would be the same for the line to neutral impedance. Why not just do a Ze then a Z line-neutral.

I see a problem with this as well. RCBOs are often used for higher measured Zs than max Zs if the breaker can't be downrated. But a lot of the time the impedance between line and earth and line and neutral can be almost the same and in a circuit wired with the same size CPC both impedances can be the same. So using and RCD to overcome Zs that's over the max still leaves a problem with a short circuit. This is never taken into account.

This is more an issue with thermal contraints than with shock protection.


I searched the post but couldn't find anything with this subject, although it might be there and I've used the wrong search words.
  • 0
    In most installations, the PSCC must be great enough to promptly open OCPDs but not be so great as to overstress such devices.


    If reliance is placed upon an RCD for shock protection, then the prompt operation of the OCPD is far less important. Earth faults will cleared by the RCD. Phase to neutral faults may take a lot longer but that seldom matters since a phase to neutral fault wont result in dangerous touch voltages, all that matters is that phase to neutral faults are cleared before the cable becomes dangerously heated.


    Consider a phase to earth fault that persists. During the fault, any class one appliances will be live at about half line voltage, not as bad as full mains voltage but still dangerous. An RCD should disconnect the supply almost instantly. If most of the resistance is in the earth connection rather than in the circuit wiring, then a class one appliance would be live at almost full line voltage, even worse.


    Now consider a phase to neutral fault that persists for some time. Far less important as no shock risk should result.


    As an extreme but illustrative example consider a circuit wired in 2.5mm SWA without any de-rating for grouping etc, and protected by a 20 amp fuse. If the cable was long enough, the fuse might never operate on a phase to neutral short circuit. But would the cable be damaged? I suspect not.

    In most real world cases voltage drop would be a problem first.


  • 0
    According to the book the time-current table for a 60898 B, C or D 32 would allow a continuous current of 46A and not trip.


    Using a 4mm T&E ref method C that allows 37A and ref method 100/102 allows 27A with no further correction factors.

    I'm wondering how the cable would cope with that.

  • 0

    broadgage:

    As an extreme but illustrative example consider a circuit wired in 2.5mm SWA without any de-rating for grouping etc, and protected by a 20 amp fuse. If the cable was long enough, the fuse might never operate on a phase to neutral short circuit. But would the cable be damaged? I suspect not.

    In most real world cases voltage drop would be a problem first.




    It wouldn't do much for your leccy bill would it? ?


    But yes, it's a fair point.


    I have to admit that it took me a while to realise that an RCD won't clear a short circuit, but an RCBO might. IMHO the OP's concern is justified. Good design would require that a short circuit would trip promptly.

  • 0
    Two different things are working here.


    Obviously, you're already aware of the disconnection times in Chapter 41, which covers a fault to earth.


    However, we need to look at whether the cables are suitable for fault current protection - there are two things here: fault to earth and fault to another current carrying conductor, L-L or L-N.


    The criterion for the protective conductor according to Regulation 543.1.3 is k2S2 ≥ I2t - this is only valid for disconnection times less than 5 s. This is typically fine for RCD protection for fault to earth.


    For L-L and L-N short circuits, per Regulation 434.5.2, then you also need to meet this criterion - but this time if you are using mcb or RCBO, the disconnection time has to be less than 0.1 s, and I2t is the value quoted by the manufacturer. If the disconnection time for an L-N fault exceeds 5 s for a fuse or mcb, you will need to calculate relevant values of k using BS 7454.


    So, I'm guessing the check you're doing (prospective fault current) is to ensure enough current is flowing in an L-N or L-L fault, to operate the protective device in a time to achieve compliance with Regulation 434.5.2 ?



  • 0
    Hi gkenyon


    We haven't actually started doing these tests. I had never heard of taking PSCC at anywhere apart from the DB before until my supervisor mentioned it, and even then he couldn't explain why, only that the NIC bloke said you should do it if the Zs was to high and you used an RCBO to comply to disconnection times on earth fault. Even the other sparks were confused. It makes sense now that you are checking if the current flowing on that type of fault is enough to operate the PD because if the Ze is too high than the other loops may be too high. But if that's the case using an RCBO will only overcome one problem not the other.

    In all the time I've been doing this I, and any other spark I've worked with, has never gone that deep into that part of the regs with those calculations.

    The most we do is check Zs measured to Zs max and take whatever appropriate action on that. Until recently, never thought about short circuit measurement issues although they are in the reg you pointed out.


    I don't know if this is more a design stage issue.


    However, I would have still thought you could use the same principle of line to neutral loop impedance and apply it in the same way as Zs. After all, it would be the same current required to trip the breaker (MCB part) whether the fault was to earth neutral of another line and therefore if within the max Zs for line-neutral it would be ok.

    From what I read it seems that the fault can be more that the time you mention. It says "for a fault of less than 1 second) then a certain criteria must be met, not that the fault must be less than one second.


    This is now getting more complicated than I thought it would be. 



  • 0
    Far end PSSC is a proxy for voltage drop, as well as the lowest case L-N firing current  for a dead short LN fault. (PSSC at the origin being the highest case for a dead short fault.)

    The regs pretend that not dead short faults do not exist, so there is no need to consider them.

    The high case  (origin end ) should be less that the safe breaking capacity of the breaker, and the far end case at least able to get the breaker to trip in 5 seconds.

  • 0

    According to the book the time-current table for a 60898 B, C or D 32 would allow a continuous current of 46A and not trip.



    For most devices they're required to trip within 'conventional time' (usually one hour for domestic size devices) at 1.45x In - smaller overcurrents will still trip eventually but will of course take longer. After all that's the whole purpose of the thermal element of the MCB.


     

    I'm wondering how the cable would cope with that.



    Surprisingly well actually. Cables don't suddenly explode or catch fire the moment their rating is exceeded. Generally they just run a little bit warmer with the result that the insulation degrades a little faster than usual. Given that a PVC cable might have a 70 years life span even if a fault or two knocks a few months or even years off a cable's life, is not all that significant. We accept 70 degree PVC cables getting to 160 degrees during high current faults - it will take a cable a long time to get to that sort of temperature with what's the equivalent of only about a 50% overload.


    From the point of view of needing to check that the MCB/RCBO will clear faults instantaneously (i.e. verifying that the fault current exceeds say 5x In for a B-type MCB) - that's really only needed where the cable is rated lower than the protective device (e.g. industrial motor circuits or ring final circuits) - in the majority of cases where the cables are rated the same or higher than the MCB/RCBO then they're automatically protected from low current fault currents in just the same way as they would be from overloads.


       - Andy.
  • 0


    This graph may help to illustrate the rise of cable core temp, versus the load, assuming conduction, not convection - i.e. a square law - convection cooling improves at higher temps, as air starts to be circulated faster- this model ignores that and is correct for cables in solid media, and over estimates the high end temperature  for air or liquid cooling.

    6da70367e11a83440f4802c33d69a330-huge-cable_heat.png

      Of course if you start at 20 or 25 degrees instead of 30, you have a bit more leeway, and the cable ratings assume a very long warm up, hours rather than seconds, so the if a breaker is going to trip it will have, long before the cable temp reaches the equilibrium value.
  • 0
    This and other tests are discussed here: https://youtu.be/C0Hj5K5cbdk

    it's a bit long winded but could be worth a listen generally.


    F
  • 0
    Thanks Farmboy


    I have listened to that one. 

    Talks about taking PSCC at the end of circuit but doesn't go into why you can't just use the line-neutral loop impedance and compare it to the max Zs as I would have thought that would be the same. Same fault current to trip MCB whether to earth or neutral.

Join us to get the best from IET EngX.

Joining EngX lets you personalise your experience so you stay up to date on the topics that interest you, plus you’ll be able to make connections who are looking to collaborate, exchange ideas and more.

Join us Not now

Community Rules & Guidelines