Online assistance for electrical trade people in New Zealand and Australia Login  |  Register  |   Forgot Password
Assistance for electrical trade people




Click here to send Forum Admin a pdf document for publication on this Topic

Documents must be less than 200k in pdf format

Posted By Topic: What to do?

Jun 25 2018 15:12

Working in a domestic swbd today noticed a C20 MCB marked gate wired in 2.5mm TPS. The gate is 80m from the board.

Curiosity got the better of me. My heat gun draws about 8 amps and on the above numbers predicted volt drop is 9.5V. Actual was 10V confirming that the cable hasn\'t been tailed down to 2.5.

The cable terminates in a double 10A weatherproof socket outlet. The home owner uses the second socket to trim the hedge via a long extension lead.

The cable is undersized. What would you advise the client? De-rate the circuit? Keep his extension leads short? Don\'t use anything bigger than 1200W? Track down the last sparky for a please explain?

Jun 25 2018 15:53

How much does the gate draw?
is it permanently connected or plugged in?
GenCalc gives quite a big VD beyond a 2.5mm at 80m @ 20a, but i doubt a gate will draw anything over 1A. Being permanently wired (if it is) the circuit at that length will never draw more than what it connected too.

You can do a volt test at the gate and if you get close to 230v than the VD is fine (within 5%)

I had someone tell me they ran a 35mm2 CU 300meters which by GenCalc is 9% VD. But they tested, still within 5% so complied legally

But again, what are you doing on the board that you are making this circuit youre problem?

Jun 25 2018 17:00

The job is a new sub circuit supplying a garage. So nothing related to the gate circuit. This is a good client so always looking to deliver best service.

The gate\'s draw is minimal and plugs into one socket. The second socket is used for garden power tools up to 10A, limited by the plug. Volt drop with a 10A load is over 5%. This could damage equipment.

The circuit should have a 5A MCB on it to limit volt drop. So far I\'ve suggested the client only connects tools under 1200W. I was wondering what others would have suggested.


Jun 25 2018 20:10

Hi Mowgli , i think your on track with your thinking , it would seem that the socket outlet is the issue , if the gate was hard wired the 2.5 would probably be ok but having the socket outlets changes things.
You have not said though what the actual voltage was. What is the normal site voltage 230 or 236 or what ? as that is the only thing that may allow some wriggle room ,Gen Cal is based on 230 so the lower acceptable level is 218.5 so did it drop below that.


Jun 25 2018 21:24

TheDon, unbranched circuit terminates in a double socket outlet. The gate plugs in.

No load voltage at the socket was 236V. Voltage with an 8A load was 226V. This was in the early afternoon.

In theory a load at peak could drop below 218V. But then how likely is it that he\'ll be gardening at dinner time on a cold winter\'s night?

Jun 25 2018 21:47

Just keep in mind guys, that the 5% allowable VD is from the point of supply, to any point within the installation.

Codaxx mention of 5% above, seemed to refer to only the mains?

And Mowgli\'s mention of the 5% @ 10A, was only for that 2.5mm2 circuit.

Before you can check how much voltage drop can be allowed on a final subcircuit, you would first need to calculate the voltage drop on the mains. So if you have 3% VD on the mains, you only have 2% allowable for each final subcircuit.

Generally, you would have each one roughly between 2-3%.
Long mains to a small house - more on the mains, and less on the house.
Short mains to a big house, less on the mains and more in the house.

As far as the gate circuit goes, it only has to be calculated by the actual load, not the 20A circuit protection. Although being connected by a power point, alters the calculation.

Another reason where that circuit may fail (I haven\'t calculated it), is if the earth loop impedance is too high to be able to clear the 20A MCB. Changing to a 10A MCB would solve that problem.


Jun 25 2018 22:33

I\'m back there on Thursday. I\'ll take some impedance measurements from the socket and the board to complete the picture.

Jun 25 2018 22:45

6A would solve your problems in terms of limiting current on circuit and get close to 3% VD if using GenCalc @ 6A and you still get 1300W.

whats a wattage of a hedge trimmer?

Table B1 for max route length


Jun 26 2018 07:54

If, and it should be, the circuit is protected by an RCD then EFLI becomes a moot point.

Of course you could remove the power point and permanently connect the gate. Make the customer run multiple extension leads to use the electric weed eater around the gate.

Of course the professional electrician who installed and certified the gate supply cable should have done the volt drop calc using the 10 amps if it was originally intended to fit a socket oiutlet at the gate.
And of course it is not 5% but what ever is left after the Mains VD has been taken into account.

Jun 26 2018 08:53

The idea that doing a test - obviously on-load - can confirm VD is OK, is flawed. As others have pointed out, the total VD includes VD in mains (and in any submains). So unless those are also \"on-load\", i.e. loaded up to their max demand, the voltage at end of tested subcircuit may give a false impression.

As for EFLI, as a general rule, if you meet VD requirements you\'ll also meet EFLI requirements. Tghis is because the two effects are based on Ohm\'s law applied to the same bit of cable in different ways. However need to recognise that in larger cable sizes (where PEC is smaller than N), the A-N loop for VD will have lower impedance than the A-E loop for EFLI. So there\'s really no excuse for getting to end of job and then finding an EFLI test fails due to design error - any failure should be due to an installation error (eg bad join). New edition has additional info on the relationship between VD & EFLI.

I don\'t believe it\'s valid to assess a socket outlet at anything less than 10 A when assessing VD.
Using Table 42 of \"3008.1.2\"; the single-phase VD for 2.5 is 16.29 mV/A/m (using cable temp 45 deg).
16.29 x 80m x 10A = 13,000 mV (13 V.
This is well over the permitted limit even if the mains VD is 0. So the circuit is non-compliant (even assuming that 80 m is the cable route distance and not just the direct distance).

As Sarmajor indicates, having RCD protection means an EFLI test is not needed, mostly because with RCD the earth fault trip time is less than the maximum pretty much as long as the PEC is not open-circuit.

However earth continuity still must comply. Table 34 gives resistance of 2.5 mm2 as 8.14 ohms / km; so for 80 m the resistance will be 0.65 ohm. Which is more than the maximum 0.49 allowed by Table 8.2 of \"3000\" for C-curve mcb. So non-compliant on that score also.


Jun 26 2018 09:16

As you have pointed out, the 0.65ohm is the maximum with a 20A C curve. But for a 10A C curve, it the maximum is 0.98ohms per conductor. (I\'m not sure that the 8.14 ohm per km is at the right cable temperature though?).

As for the \"protected by an RCD then EFLI becomes a moot point\", it still needs to comply for short circuit protection doesn\'t it? (A-N)

Jun 26 2018 15:02

I did a SC calc, using guestimated data, but got the \"Foprbidden\" message when i posted.
Bottom line is that SC protection works differently from fault protection.
With fault protection, we have a set max trip time that applies to the entire circuit; and adding too much impedance makes it non-compliant.
For SC protection, the max trip time varies as a function of PSSC for that point of the circuit; and adding impedance actually gives a longer permitted trip time. Worst case is at origin of circuit, so this may well comply.

8.41 ohms is for cable temp 40 C; which is simply the lowest temp column in the Tables.

And yes changing the mcb rating and / or trip curve makes a difference.

Jun 26 2018 21:52

De rate the circuit.

Jun 26 2018 22:24

If I was to disconnect that circuit, for whatever reason, then I would have to derate it before I could issue an ESC to reconnect. That part\'s a no-brainer.

It\'s non-compliant but is it imminently unsafe? I think not. Given that I am not working on that circuit I don\'t think I\'m obliged by regulation to do anything. Am I correct?

Jun 26 2018 22:52

Correct. You are not responsible for other people’s work.

[quote]This is a good client so always looking to deliver best service[/quote]

Then may be best to mention it as professional advice. However then you would need to act on it. Which shouldn’t be a major to “fix” anyway.

Jun 27 2018 10:04

Agree you have no legal responsibility.
not even to report, as you\'re right that it is not an \"immediate danger to life or property\" so ESR 19 is not triggered. The only risk is to their plugged-in equipment, and then only if the load is high.

You do have a moral obligation to advise the owner; after which whether they choose to act (or ask you to act) is up to them.

Jun 27 2018 12:25

Getting back to \"I don\'t believe it\'s valid to assess a socket outlet at anything less than 10 A when assessing VD\", would you consider it acceptable if the socket is a labelled restricted use socket? And would it make a difference if it\'s on a 6A MCB? I\'ve come across a few 3-phase appliances with an integrated socket for supplying ancillary equipment where the socket is protected at 6A to keep the whole machine\'s maximum current below its plug rating.

Jun 28 2018 08:53

Unless you want to put your name on it I just wouldnt touch it..

Jun 28 2018 09:52

There\'s a world of difference between restricting the current available for plug-in extras on an appliance (equipment) to keep within rating of supply plug & lead; and assessing MD for VD on installation wiring. So not really a valid comparison.

WRT installation wiring; since \"limitation\" is a valid method for assessing MD of installations / part installations, have to accept that it can reduce the MD for VD purposes. ny earlier statement was in the context of the suggestion that it was OK because only low-load appliances were intended to be used. That\'s not valid. If it were; we\'d be able to apply the same logic to overcurrent protection of conductors (\"it\'s OK, they\'ll never plug more than 4 kW of appliances into the 10 sockets on the circuit, so it doesn\'t need to be protected against some silly bugger doing it.\" yeah, right).


Jun 28 2018 12:05

The on-load VD comparison was not intended to validate the installation VD. It was only used to establish that a larger cable hadn\'t been tailed down prior to the switchboard.

Given the quality of the work in evidence above ground, I\'m not prepared to take responsibility for the work underground. This rules out de-rating as I\'d have to issue an ESC for the circuit before reconnecting.

My advice to the client at the time will stand. The circuit is adequate for the gate, and likely safe for plug in loads up to 1200W or 5A.