Earth loop impedance problem

[cynic]

I have been asked a question by a guy who is presently preparing to undertake a C&G 2391 practical assessment. I can see solutions but not ideal so I wonder if other minds can suggest something diferent and maybe simpler. The situation is this

100A three phase supply TN-S with Ze of 0.8 ohms by enquiry (actual measured value 0.76 ohms) The first DB in the chain (DB1) includes a 63A single phase breaker protecting a single phase sub-main supplying a consumer unit in a flat above the workshop. The sub-main consists of 16mm conductors for both live and neutral and a 16mm earth conductor so Zdb at the consumer unit is measured as 0.77 ohms. The circuits protected by the breakers within the consumer unit all comply with the regs relating to max earth loop impedance taking account of rule of thumb. The problem is that the max impedance for a 63A type B MCB is tabulated at 0.71 ohms and rule of thumb takes this down to less than 0.6 ohms required. Considering the requirements as part of the test results of DB1 the circuit does not comply with requirements and the schedule of test results and notes to the intermediate inspection report reflect the situation. If this occurred in real life, rectification would be required. His question to me was how best to modify the circuit to comply. As I see it there is no possibility of gettinig the Zs down to 0.6 ohms as Ze is higher and I cannot think of any way to get a negative value of R1 and R2 for the circuit conductors. An easy solution is to provide RCD protection for this circuit but the RCD will then provide whole house (flat in this case) cover and leave the occupants vulnerable to losing lights and everything else in the event of an earth leak on one of the other circuits in the consumer unit. This is generally considered unsatisfactory by present day standards. Using a 100mA time delayed RCD and 30mA quick acting RCD or RCBO to serve sockets in the consumer unit would probably be satisfactory but can anyone suggest an easier way which woukd comply with the Regs?

[Colin Wilson]

See if the local electric board can PME the supply ?

Sink earth rods ?

[Andy Wade]

My copy of BS 7671 says 0.76 ohms for a 63 A Type B MCB to EN 60898. That would seem to be based on Uoc = 240 V and 315 A required (5 * In). Nevertheless this won't help since you're still looking for a value less than the measured Zdb. This is clearly a contrived example of an on-the-edge design.

Another point is that the rule-of-thumb method would be unduly pessimistic here since almost all of the Zdb is due to Ze. You are allowed to assume that Ze is constant and apply the temperature correction factor (F) to the R1+R2 value of the circuit conductors - this is the basis of the formula

Zs(test) An easy solution is to provide RCD protection for this circuit but

A 100 mA or preferably a 300 or even 500 mA time-delayed RCD would be quite acceptable, IMHO, and surely would be the most cost-effective solution. Or, as Colin said, get the service upgraded to PME. Driving earth rods in an attempt to reduce the Ze on a TN system will be utterly fruitless, unless there is something seriously wrong with the distributor's network (which we know is not the case since Ze is within the declared 0.8 ohm value).

Other options might be to reduce the MCB rating from 63 to 50 A, but that would depend on the assessed MD of the flat, about which we have no information, or to select a different device to protect the sub-main. A 60 A BS 3036 fuse is tabulated at 1.17 ohm for 5 s disconnection - but re-wireable fuses are rather out of fashion these days! An adjustable MCCB might provide another option - expensive though. Note reg. 413-02-13 if the disconnection time at the sub-DB (i.e. the flat's CU) exceeds 0.4 s.