Fusing a sub main

Discussion in 'Electricians' Talk' started by ajohn, Jul 8, 2019.

  1. Tony Goddard

    Tony Goddard Active Member

    I've done a few jobs on installations that were, er non compliant, for example a house last re-wired in 1927 and the owners didn't want anything done except the feed to the summer house - thats where splitting the tails and fitting a little DB saves 9, new tails from meter to henly block, tails from henley to my new shiny compliant DB for the summer house, which is therefore nothing to do with the main house wiring, covering letter stating that existing wiring in the house is likely unsafe, job done.
    The house (a 16 bed georgian mansion) is still standing with the wooden fuse boxes and lead sheathed wiring still just about working. But the summer house is 17th edition (as it was a few years back) compliant throughout!
     
  2. ajohn

    ajohn Well-Known Member

    It seems that there are still some houses around here on rubber according to the previous owner. The house was built in 1911. Don't really know when it was rewired. It was fitted with economy 7 and 2 phases, 2 meters and 2 wylex rewireable cu's. It looks like the final surviving owner was thinking about converting to flats but it never happened. 2 ladies. One lived on the ground floor and the other on the first. Then students at times on the top floor. They also go rid of all of the steel gas pipes and had larger bore copper fitted, 2 separate runs and 2 meters.

    My fathers rewire was with a rewirable. He flatly refused to have a plastic consumer unit fitted, One reason was the one that caused their demise. It happens, That was metric wiring. Don't think he liked the wilex ones. The gas and electricity boards etc used to come and talk to him now and again about how factories went about "doing it". They wouldn't let the boards do any work. Not good enough.

    I wonder if false trips are really down to too many things on a circuit. :) Also nervous about fridge freezers with plastic backs. o_OIt's even got an inverter in it. :( Probably finish up asking for more circuits than I need.

    LOL Trouble is I am a design engineer. :) I decide what's to be done.

    John
    -
     
  3. Bob Rathbone

    Bob Rathbone Well-Known Member

    Not every time Baz. Provided that the cable is protected from overload at the load end, such as in the overload units in motor starters or CB's or fuses, then the supply end fuse or device is for fault protection ONLY and may be rated accordingly to prevent overheating in the event of short circuit or earth fault.
     
  4. ajohn

    ajohn Well-Known Member

    It can relate to the cable size after a fashion. The 4mm swa has a rating of 47A worst case using the BS fiddle factor for 30C. A 40A fuse will protect that as well. Looks like a 63A faster blowing one could be used but wouldn't do the same thing. Service fuses are much the same - they are not expected to be loaded above their ratings.

    John
    -
     
  5. ajohn

    ajohn Well-Known Member

    Whoops etc. messed up the fuse selection. Eaton data sheet with a scale that is a bit odd. Might be why they updated it.

    John
    -
     
  6. Tony Goddard

    Tony Goddard Active Member

    Based on that delightful comment, I'm guessing you have George? hope you were wearing the correct PPE!
     
  7. ajohn

    ajohn Well-Known Member

    Some might find this mmmm interesting or whatever

    http://lawsonfuses.eu/lowvoltage.pdf

    :eek: It's sort of interesting. To give a chance of just the mcd tripping needs an 80A fuse to cope with L & N shorts, Worst case short to earth is 200A that would take 20 to 90sec to blow different fuses. A 40A motor fuse is shortest. That needs a screw through the SWA to get that. Frankly looked at that way I don't think any SWA arrangement will be any different. Go to 3 core and the earth conductor may not be hit by the screw so doesn't achieve anything. The cu at the end looks after itself as far as cable coming out of it is concerned even using armour for the cpc. A 100A supply fuse would just about take 200A for 20sec,

    So ??

    John
    -
     
  8. Woloumbo

    Woloumbo Member

    the current curves are in bs7671.
    you always design your circuit to ensure the earth loop is low enough to ensure disconnection in time.you make sure your earth is low enough to allow enough current to flow to blow fuse in specified time.its been this way for donkeys.
     
    Hfs likes this.
  9. Tony Goddard

    Tony Goddard Active Member

    Not changed since the concept of protective earthing and fault current was "invented" by Benno Mengele at Vienna university very early last century, he worked out the loop impedances/currents to rupture various fuses and most of that data still holds. He was working for Siemens at the time and they popularised the idea of using earth for safety purposes.
    He also invented the method of using 4 spaced conductors on high voltage transmission lines to prevent corona discharge and subsequent losses in power, you see that technique on virtually every high voltage national grid line.
    Sadly he is a largely forgotton pioneer, possibly because he shares a surname with a well known Nazi war criminal, to whom he was not related.
     
  10. ajohn

    ajohn Well-Known Member

    If you calculate shorts between armour and live that is the highest resistance the fuse needs to protect. Having worked one out with a Ze of 0R8 it indicates that any arrangement is likely to have problems allowing mcd's or rcbo's to work independently without blowing the fuse and meeting armour to live short needs. End result is likely to be the main supply fuse blowing. Because the 5 sec rule can't be met. That may have problems if it is as well.

    John
    -
     
  11. ajohn

    ajohn Well-Known Member

    Can't say I'm impressed with an invention of protective earth. Corona yes but wish my hair didn't stand on end when I walk under one in particular but fortunately they are pretty rare in B'ham and none around where i live now.

    John
    -
     
  12. ajohn

    ajohn Well-Known Member

    Have to laugh sometimes. A 45A BS3036 would do it perfectly. Odd how a piece of semi enclosed wire behaves somewhat differently to variation on the same that are enclosed. Those in an odd sort of way are 90A fuses though as it would take that for a fair old time.

    So were the old guys that did this stupid ? It's not unusual to find that they were pretty bright really. Fuses to protect cable. Appliances done via the plug.

    John
    -
     
    Last edited: Jul 11, 2019
  13. ajohn

    ajohn Well-Known Member

    Perfect item for cable fault protection ;)

    IdealCableProtection.JPG

    Discontinue and replaced with one much larger that takes the usual carridge fuses. That one can take all including an mcb.

    ;) I always wonder on things like this that if there had been some one around nearing retirement when it was replaced they might has said hang on a minute.

    John
    -
     
  14. Woloumbo

    Woloumbo Member

    lets see your calcs.
     
  15. ajohn

    ajohn Well-Known Member

    It's just ohms law using Prysmian numbers. 7R9 / km for the armour. Copper same as the guide etc. Temperature coefficient 1.2 on both as that looks to be true of steel as well.

    If I went to 6mm fault currents go up by 50% on live and armour resistance decreases by just 11%. 3 core armour resistance is a bit lower but doesn't help with live to armour shorts. 7R5 / km.

    Fuse can be seen much the same as mcb. What they do is a ratio of the rating except on a fuse it's a curve so the same thing is very likely to happen what ever size of cable is used.

    Fuse curves looked at. Several manufactures and the blue book.

    A Ze of 0R35 may well just about fix the problem and make a fuse work out but little margin over am mcb tripping out. I take it TN C S supply effectively bonds the sheath to neutral ?

    I don't think it's possible to do a calc without knowing what Ze actually is. Normally as a diy'r I'd go buy something to find out. Ze here may be very low as their is a substation around the corner but we might not be on it.

    There does appear to be a solution for running T&E without an rcd - flexible metal conduit. :( Not this helps with fuses etc but in my case may make installation easier.

    Anyway I always check things twice and yet to do the sums again but I'll be surprised if they come out any different.

    John
    -
     
  16. ajohn

    ajohn Well-Known Member

    Spot the mistake - Wylex switch fuse unit, current one
     

    Attached Files:

  17. Woloumbo

    Woloumbo Member

    overthinking
     
  18. Woloumbo

    Woloumbo Member

    put some maths down,lets see youre calculation for fault current
     
  19. Tony Goddard

    Tony Goddard Active Member

    Flexible metal conduit rusts badly enough in a slightly damp location, underground I would think it will be rusted through in a year
     
  20. ajohn

    ajohn Well-Known Member

    [
    The suppliers reckon it's used for under floor distribution. In my case that could help getting cable to the swa using t&e. Some needs to be clipped direct which the guide states is ok. That probably means that the jumble in the actual standard doesn't make that very clear. The reason the 17th allowed t&e installed correctly was due to problems that can crop up routing swa. Seems it's still ok run up walls and in view.

    Really the best way to protect a submain would be an rcd with a significant delay. Looks like that would need electronics which means they can't be used. Odd really as something like that could be produced with fixed setting that would probably last longer than the owner. ;) The mechanical bit that broke the circuit might not though.

    John
    -
     

Share This Page