This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

MCCB Max ZS & Correction factor for temperature rise under fault.

Hello All,

I have recently completed an EICR which has MCCB's fitted most are Merlin Gerin NS with TM D or STR blocks all feeding large final or submains, allowing for a 5s dis connection time although the data tables the zs is the same for 0.4 or 5s. I have calculated the max Zs values for these taking into account the tolerance permitted by Schneider and the C min value. I have used these as the Max Zs recorded on the test sheet. However when coding I have taken 80% of this value in lieu of the possible temperature rise under fault conditions. The client has since queried this as it has caused some circuits to fail. When I spoke to Schneider they said as I have there maximum Zs values from there tables these can be used (which are the same as what I calculated on site) but there calculation are done at ambient of 30 degrees. Therefore I believe a derating factor would still need to be applied to allow for the possible temperature rise in the conductors under fault condition. And that where the measured Zs does not meet this corrected max Zs a C2 would be the correct coding.  

Look forward to hearing your opinions. 

Thank you

  • I wouldn't get involved in a technical decision with the client on this one. The 0.8 factor for the impedances in Tables 41.2 to 41.4 of BS 7671 is actually provided for you in the guidance in BS 7671 itself, Appendix 3 (page 410 in the Brown 2022 book).

    I would answer the query by replying to that effect.

    The 0.8 factor is to be used with respect to look impedances in Tables 41.2 to 41.4. If you use the 'maximum measured loop impedance' tables in OSG or GN3, the 0.8 factor is already taken into account for you, and you can just use the loop impedance value from those tables.

    You could also answer the query regarding older circuits that used to be compliant, that, in earlier Editions of BS 7671 and the Wiring Regulations, higher impedances were permitted, but they have been adjusted in later 16th Edition (for change from 240 to 230 V as the nominal voltage), and in 17th Edition with the inclusion of Cmin. This means that some circuits that used to be compliant are no longer considered to meet the latest safety recommendations.

  • Thank you, I have explained to them the added derating factor of Cmin which was not likely taken into account on the previous inspection, and that we have to ensure compliance to 0.8 of the maximum Zs. As MCCB's aren't covered in the wiring regulations due to the vast differences in tripping characteristics I just wanted to verify that any max Zs; even if given by the manufacturer, is taken at ambient temperature and not at full load would still be required to be taken at 0.8 as per appendix 3. They are being reasonable but are now worried about the cost of correcting the issues if what I'm saying is correct as some of the given values are extremely low. 0.03 at 100% for example a Zs they are unlikely to achieve at the end point especially when taken at 80%. 

  • even if given by the manufacturer, is taken at ambient temperature and not at full load would still be required to be taken at 0.8 as per appendix 3.

    Yes, although the conductor temperature (70 deg C vs ambient of 30 deg C) is what's important. The manufacturer's loop impedance values for 30 deg C align with the Tables I quoted in BS 7671.

    On an EICR, if you don't have the original design data regarding circuit design, assumed fault currents and disconnection times, it's not possible to do a "re-design" on the hoof, so I guess the only option would be an "FI" (which definitely leads to "unsatisfactory") and doesn't help the client ? If you did have that data, you could take it into account.

  • 80% is a somewhat arbitrary rule of thumb that has a number of assumptions behind it. It's not so much for temperature rise under fault conditions as for the difference in temperature between when the impedance is measured (often when the system is unloaded) and normal operating conditions (i.e. when the conductors are much warmer). If you know the conditions you can likely calculate a more accurate factor - usually based on 1+0.004 per ºC (at 20ºC) - based on the actual temperature of the conductors when the test was carried out and their maximum operating temperature.

       - Andy.

  • Thank you for the guidance so far, unfortunately the previous EICR's carried out (by the company i now work for) simply put the maximum Zs as non verified and placed a limitation on the EICR, now that I have come in and started to source the actual maximum Zs figures its causing issues as circuits that should have likely failed previously but were passed are now failing. We were doing the testing on behalf of another contracting company who now need to explain the C2's to the client and feels allowing a 20% decrease in the maximum Zs is unfair, unfortunately knowing the operating current would require a thermal of all the conductors to be done and the more accurate factors allowed for at the time of test which has now passed. Although thermal imaging was carried out of the distribution boards as a whole and no conductors were noticeably sitting much above ambient at the time of test. 

    The contracting company has asked since the incoming supply is greater than 230V (around 245v) can the measured voltage be used instead of the 230v nominal line to earth voltage as stated in the regs. The transformers on site are client owned. However i can see In ESQCR that unless specified in writing the voltage shall be assumed to be 230V. Would this therefore only be possible if the generator/distributor of the supply states it will be 245 and not change.

    As much as id like to say accept the report, I am unfortunately now dealing with the incompetence of my predecessors who failed to verify these measurements on previous inspections. 

  • Do you mean the main worry is the  temperature rise under fault ? As in the extra heating of the wire that occurs  in the time between someone damaging the cable or otherwise shorting it out,  and the trip firing off ?  If so, this does not normally cause issues for faults at both ends of the cable at once, and so long as one is happy to accept that after a really unlucky tripping event the cable will need checking and maybe some of it at one end or the other may need replacement (and if it has been nailed or drilled then it may need replacement anyway) it is not really a C2.

    Is the bigger problem that of a high Zs, that the trip may not operate fast enough or in a really bad case of a far end fault, never at all ? If  you assume the cable is already running hot at the start of the fault ?   If so it may be worth checking voltage drop at full load as well, as that normally fails before the loop impedance, unless the earthing is very weedy.  Before re-wiring or specifying earth fault relays I'd be wondering how close to pass/fail it really was.

    Also note that a Zs on a running system may already have a hot live and a cool earth so keep your wits about you about the true test conditions.
    Mike.

  • Hello, the second point, I worded it wrong on the title and is relating to the 0.8 factor that is applied in accordance with appendix 3, some of the results are extremely close to the limits within 100% of the max Zs but not with in 80%, I try to be reasonable with the measured Zs due to equipment tolerances and temperature levels, some of the MCCBs are 630A or 400A and have the ability for the magnetic settings to be reduced which should help pass some circuits, its the fixed magnetic TM100D that are more of a problem.

    The contracting company when installing some of the new equipment a while back used Amtech to verify there cable choices, runs and expected Zs. I have not seen any of this data however or the installation certs but it wasl all given to the client at some stage to prove compliance when it was installed.. They are claiming that Trimble is giving higher allowable zs readings but have not received what these are to compare to my readings or what numbers they are using to get the values.

    Although i know that for at least some of these circuits there is likely to never be an issue i cant prove it, the regulations errs maybe to much to side of caution in some instances but without the initial design data i cannot confirm it would be safe. I feel if we were to use a low ohm earth impedance tester we may get slightly better results on those that are closer to the transformers which I believe is what was used when the install was commissioned. But was not asked to be used or priced for in the EICR report. 

    As a side note would you say these are often used for periodic inspections on larger installs. I have rarely seen them used on sites before but have also not had many issues relating to the MCCB's and maximum Zs's before now so have had little need for them.

  • Hello, the second point, I worded it wrong on the title and is relating to the 0.8 factor that is applied in accordance with appendix 3, some of the results are extremely close to the limits within 100% of the max Zs but not with in 80%,

    What is the in-service accuracy of your test instruments? Are they within the 0.8 limit, given the stated accuracy of the test instrument? If so, you can't simply say "FAIL". If it's an actual loop test, some instruments are not that accurate at all depending on the conditions.

    How does 'Measured ZS' stack up against ZDB + (R1+R2)?

  • Also note that a Zs on a running system may already have a hot live and a cool earth so keep your wits about you about the true test conditions.

    Good point. Similarly not so long ago the appendix said we need only apply the adjustment factor to the installation's conductors (R1+R2) and could use the external (Ze) part as measured - that option seems to have disappeared, but still appendix 3 is still only "Informative" and 'other methods are not precluded', so something along those lines could still be appropriate.

       - Andy.

  • The contracting company when installing some of the new equipment a while back used Amtech to verify there cable choices, runs and expected Zs.

    Therein, I suspect, lies some of the problem. If you just put figures into the software, you have no idea whether the results are suitable or not; nor whether they are safe by a good margin or borderline.