High Zs

[Spark123]

To me with the use of RCD's on all circuits the neutral v line impedance is far more important than earth v line impedance but old habits seem to die hard.

I'm the opposite, I look at things from a safety POV before convenience.
If the L-E was reasonable and RCD protected I wouldn't have a problem with it.
As for L-N impedances or volt drops, as I have said before they are a nightmare to work out on a RFC as the load is distributed around the circuit.

 

[ericmark]

The Line / neutral impedance is no harder to measure or to work out if results pass than Line / earth impedance and we have always been required to measure both but only had to record the latter.

As to volt drop on a ring circuit the organisations who control us have decided we can consider 20 amp to be drawn at the centre of the ring and the remaining 12 to be evenly distributed so for calculation we consider a 32 amp protected ring main as being 26 amp for volt drop calculations.

They also give us other standard parameters like ambulant temperature of 30 degrees C and we will normally consider the cable to be rated as 20 amp and yes these may not be true. But using the standard figures the 18mV/A/m is corrected to 16.614mV/A/m using 0.923 as correction factor Ct.

If for example we used 13.5 amp as rated current carrying capacity then the Ct value drops to 17.82551 and total cable allowed is reduced to 99.25 meters.

But we have to record the installation method on many of the schedule of test results forms although I note not on the IET version. But with programs like excel anyone checking the schedule of test results only has to enter results to highlight any possible errors.

I do realise that some people enter the given figures rather than measured figures in the parameters for the supply but then they would enter R1 + R2 rather that ELI so either way it is so easy to check for any guesstimation! that has been used. And the days of putting in guessed figures is no longer possible as one will be soon caught out. Or at least one should be if the clearing houses and LCBC are doing their jobs.

 

[Spark123]

As to volt drop on a ring circuit the organisations who control us have decided we can consider 20 amp to be drawn at the centre of the ring and the remaining 12 to be evenly distributed so for calculation we consider a 32 amp protected ring main as being 26 amp for volt drop calculations.

How are you calculating the volt drop? How long is the RFC end to end?

They also give us other standard parameters like ambulant temperature of 30 degrees C and we will normally consider the cable to be rated as 20 amp and yes these may not be true. But using the standard figures the 18mV/A/m is corrected to 16.614mV/A/m using 0.923 as correction factor Ct.

Is that for 30ºC to 20ºC? Seems reasonable that the Ct will be lower causing the length of the circuit to increase I suppose.

If for example we used 13.5 amp as rated current carrying capacity then the Ct value drops to 17.82551 and total cable allowed is reduced to 99.25 meters.

Ct=17.82551 seems a little bit on the big side!
Rated current carrying capacity of a RFC should not be less than 20A per leg.

 

[ericmark]

I used formula (B1+B3-((POWER(B5,2)*POWER(B4,2))-POWER(B6,2)/POWER(B7,2))*(B3-30))/(B1+B3) in excel where
B1 = Ambient temperature and
B2 = Correction factor Ct and (formula is in B2 box)
B3 = Max permitted operating temp tp and
B4 = Rating factor for grouping Cg and
B5 = Rating factor for ambient temperature Ca and
B6 = Design current for circuit Ib and
B7 = Tabulated current-carrying capacity of a cable It
and 30 is ambulant temperature.
The formula is taken from page 258 in red book section 6.1 and it was done after attending a lecture from IET where the speaker said we could now use 106 meters of cable in a ring main. Which was at that time a lot more than I thought we could use so I asked how he had arrived at that figure and this was the way he had worked it out.

As I say once the formula has been entered into excel to alter a few parameters is easy. But not really the sort of sum I would work out on the back of a fag packet.

 

[Spark123]

Which formula are you using to do the voltage drop calc for a RFC?
The one above gives the Ct value iirc.

 

[ericmark]

Which formula are you using to do the voltage drop calc for a RFC?
The one above gives the Ct value iirc.

106.49=11.5/(26*16.614/1000)*4

11.5 = permitted volt drop
26 = design current i.e. 20 + half of 12
16.614 = Ct x mV/A/m
1000 to convert to volts from mV
4 because it is a ring main

 

[Spark123]

I can follow the logic in that, the x4 being the mid point has 2 cables in parallel and the length of the cables is halved.
So if your ambient is 20 then for a RFC you can have up to 106m, what did you calculate the OP RFC length to be?

 

[ericmark]

It was worked out at 30 deg and sorry what is OP.

Assuming the Zs = 1.43 and Ze = 0.58 then R1 + R2 = 0.85 divide by (29+18) and multiply by 18 gives the value for R1 = 0.325532 and R2 multiply by 29 = 0.524468 so resistance line – neutral = 0.651064 we are not sure on temperature at which it was measured so is somewhere between 16.614 and 18 mV/A/m so somewhere between 144 and 156 meters of cable used. With original maximum being 85 meters this means somewhere around 60 meters of cable has been used to wire extension.
Oh dear too much volt drop I wonder why? This of course assumes it was correct in first place.

But working out what the guy has done wrong does not really help.
What I have often wondered is if replacing a long run with thicker cable is so wrong. The regulations say fuse where you change cross sectional area which is of course out with a ring main.

If from the same fuse a length of 6mm was run to centre of ring main it would reduce the reading to an acceptable level.
But as far as I can see this is not allowed. Maybe I am being thick but I can’t see why this is not allowed? But again you know when your on site normally you can find a way to get around a problem. I am sure there is a way to split ring into two.

This of course should not be able to happen under Part P as plans have to be submitted before work starts to the LCBC and they are responsible for site safety and should have picked up on this and insisted on a second ring. The same applies if done by a registered electrician. I have not used any complex maths in working out how much cable must be in the house and any tradesman should be able to do the same.

But I have not seen any reply from Rozz since first post so this is really all academic.
Why are we at all bothered as it seems Rozz isn’t?

 

[Spark123]

That amount of cable does seem excessive to me too, is it some sort of mansion?
OP = Original Post
R1+R2 in a RFC will have parallel paths which out of interest is why I asked how you worked out the length.
Rozz may be well gone by now but still it is nice to pick someone elses brain for a change!

 

[ericmark]

You are welcome. Did you want copy of excel sheet I use to work it out. I can always stick things on web site as temp place to transfer.

I will admit the temperature is a problem as normally on any PIR or installation cert there is no reference to if circuits were in use or not prior to tests.

On 16th Edition I had excel work sheet made to both check all readings and print a schedule of results that way I was sure no one will in the future check my results and highlight an error.

What has been a surprise is ELI is very rarely the reason why any installation fails. The volt drop and line - neutral impedance are far more likely to be limiting factors.

In theory when measuring ELI it should auto select line and earth and for prospective short circuit it should select line and neutral and with top of range robin it did. But it seems many testers use line and earth to measure prospective short circuit current. This again became quite obvious when all readings taken with a meter where both reading are taken at same socket 240/ELI = PSC which is all very well where earth is same size as neutral but where earth is 1.5mm and neutral is 2.5mm then if measured correct this should not be case.

Which means one has to build ones own leads to measure the PSC. Both with PAT tester and 16th Edition tester we had a second case with all the adaptors we had made up.

May be OK industrial to have a old shopping trolley with all the bits one needs but not so easy when on a construction site when playing snakes and ladders with cables and their routes?

 

[Spark123]

I generally just us 2 test leads with my Megger MFT 1552, therefore I know exactly which test it is carrying out wether I want it to measure phase to neutral, phase to phase or phase to earth pscc. As for efli, I only measure each phase to earth recording the highest value.
I also have one of those 13A plug adaptors with banana sockets which I can plug the test leads into instead of messing around with a pre built lead.
The only test which bugs me a bit is an EFLI no trip RCD which needs all 3 leads.
I have had circuits fail on efli before, I am not looking forward to a PIR on a building I maintain which I need to carry out shortly as under the 16th it passed but under the reduced figures of the 17th it will probably fail
I may need to run an extra CPC to get the loop impedance down a touch, or bin the C63 and install a C50.

 

[ericmark]

With old MCB’s there is a problem pre number or letter type as there are no longer any details available for the units and being thermal trip only unlikely they will comply with modern regulations. With the magnet part of the trip we have:-

Type 1 = 4 times rated current,
Type B = 5 times rated current,
Type 2 = 6.997085 times rated current,
Type C and 3 = 10 times rated current,
Type D = 20 times rated current.

So if we take a typical example B16 MCB will trip on the magnetic part of the trip at 16 x 5 = 80 amps or less. For 80 amps to flow the resistance at a nominal voltage of 230 volt will need to be 230 / 80 = 2.875Ω. In the old days 240 / 80 = 3Ω. But it is immaterial as to if you match 80 amps to the PFC or 2.875Ω to impedance. With a circuit protected with an RCD only the Line – Neutral need be measured with no RCD both Line – neutral and Line – Earth are measured and you take the worst of the two readings.

So what do you do when taking readings as far as voltage is concerned. The question is does the meter measure amps and convert the reading to ohms or measure ohms and convert into amps? Most meters I am sure measure amps. And are calibrated to read in ohms using the nominal voltage. Not all meters do this some do give a true reading and if you measure the loop impedance at 55, 63, 110, 230, or 240 they will give true reading.

I found if with the so called 110 volt earth loop impedance meter measured the ELI at transformer and then at end of the cable and subtracted the two readings to give impedance of cable I got a different result using a 3 phase transformer to a single phase transformer.

I also tried same with a 230 volt extension lead close to the CU and at a distance from CU and again the extension cable had changed its’ resistance.

Now the regulations say 132.2 you can get the results by calculation, measurement, enquiry or inspection. Also 313.1 and 434.1 say the same and we are ensuring PROTECTION AGAINST FAULT CURRENT so I see no reason why we should not use the direct amps reading rather than adding a load of calculations?

However should we correct for over voltage? This has been debated again and again. If you arrive at the ELI by the R1 + R2 method and use inquiry to find the PFC with cold cables then so long as the supplier does his job and gives to a supply to spec there are no problems.

But if you use actual supplied voltage and actual ELI then the results will not match those taken using inquiry and calculation. And if the ELI on a TN-C-S supply was 0.25Ω and on a 4mm² radial on a 32A MCB you get a reading of 1.35Ω should it be a pass or fail? You are allowed 1.4375Ω but at any time the supplier could change where you are getting a supply from and the Ze may rise to 0.35Ω.

Also remember the same rules go for a factory which may have a 300A three phase supply with a ELI of 0.1Ω and the transformer could well be on their premises. By time it has gone through 5 distribution boards to an office block it may has at the final distribution board a ELI of 0.45Ω. To add onto this 0.25Ω to correct for possible supply changes would be silly.

So where does that leave us. Well at the end of the day we have to remember we are skilled men and we need to use that skill to make a judgement as to what is reasonable. There is nothing wrong with writing in a report the ELI using a nominal voltage of 230 volt was not low enough but it was noted that actual voltage was 250 volt and as a result at this time no problem exists. Although it may caused problems when tested at a future date consideration should be made to improve the situation, either by de-ratting supply, or increasing cable sizes. Or even just pass it?

If we just had to follow a rule book then a semi-skilled person could do inspection and testing like they do for PAT testing. We are skilled and have to make the executive decision.

Of course when I was the person who was on salary and would need to spend more time at work to correct any faults I found there was a tendency to try and not find problems. But when self employed and work is drying up there is also a tendency to play safe on fault identification. Although not to the stage where a second opinion may be sort.

 

[Spark123]

You'll be glad to know I am not self employed, I work full time on maintenance/breakdowns for an industrial firm so I get a monthly wage wether I do anything or not.

I help out at my amateur local theatre with lighting etc hence I end up doing things like periodics and remedial work (free of labour charge) so I don't gain anything by doing it. What work I do carry out I make sure it is done right. It can get some peoples backs up but that's life! They can blame my apprenticeship for that, I served my time in a place where money was no object and an emphasis was put on doing it safely and correctly.

A few years back there was an issue with the distribution in the theatre hence I ended up installing a 3 phase C63 breaker from the main dis board feeding a smaller 3 phase board 20 or so metres away from the main board using 16mm 4C XPLE SWA. This circuit is not RCD protected.
The Zs for this is not far from the 240v max allowable measured efli, 230v max allowable might just push it over the edge. I suspect doubling up the armour with a 16mm earth will bring it down enough, I'll have to get the measurements.

The figures for comparison for Zs are available in the on site guide, not a lot I can do about Ze changing as this is out of my control - it could go open circuit tomorrow for all I know. IIRC the Ze for the supply mentioned above it is relatively low and is a TN-S.

 

[ericmark]

In the main I had a job like yours. But I have also done some odd jobs as well including a tunnel boring machine running an IT system. Much of it abroad. Often the buck stopped with me and I had to decide what to do. I am sure I made many mistakes. Worst was machine I designed cut the end off poor guys finger. Mind you he also did a daft trick as well.

Pneumatics and how the PLC controlled them was a big part. Written 1000 step PLC programs to run concrete presses. But most of the cable sizing was more guesstimate just look at run and say think 10mm will do that and normally I was well within the limits.

Bigger cables when running in 150mm then more careful. Then my son started and went into house bashing. I could not believe what they got away with. No plans! Horror! I never installed any system without plans.

God if anyone had tried to follow me and work out what my PLC programs where doing without plans they may as well just give up and start again.

He had to drill he's own holes in beams without permission. I would normally have a gang who installed all the tray work for me and drilling into beams to poke wires through them. No way. Maybe drill and tap to hold the wires but beams were never drill through.

Whole different world. But then I realised they do less calculations than me and many only seemed to know pass level and had no idea why it passed or failed.

Even going to college the lab assistance/lecture said we put EEBADS there and asked why "well that's what we always put there". And the class accepted it. The branches of the trade are so different. And when my son started on his own and I went to help I felt I was starting all over again. There were many skills I lacked and I was well down on speed never worked so hard.

So maintenance is what I like. Also did some installation again re-learn the trade. Can't mark a gland with the spanners. Every bend must match. Not a clue what you are installing. Read the plan and follow to letter. Put cable markers on upside down and not only foreman tells you off but also all guys around you. Never used so many tie wraps?

I have even done some auto-electrics.

And now on the scrap heap.

Out of interest the same subject has come up on the IET web site. at least we didn't bring Kirchoff's Laws into the discussion!

 

[ban-all-sheds]

But then I realised they do less calculations than me and many only seemed to know pass level and had no idea why it passed or failed.