IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Weird ELI Readings...Please Help!!!!
Topic Summary:
Created On: 19 December 2010 11:08 AM
Status: Post and Reply
Linear : Threading : Single : Branch
1 2 Next Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 19 December 2010 11:08 AM
User is offline View Users Profile Print this message



sirsnortalot

Posts: 38
Joined: 25 July 2008

I have just fininshed wiring a small house and was testing it yesterday. Things started off ok, external loop Ze was 0.18, PSCC was around 1.25Ka on a TNS system so nothing unusual there. Doing the circuits, R1 + R2 values were also as you would expect. It was only when it came to the earth loop impedance values on the circuits that things started to get odd! Most of the socket rings were coming out around 1.04 Ohms give or take with R1 + R2 of 0.34!!!! I cannot understand why these values are close to the max of the 80% when the Ze is fine! I know the figures are within tolerance but they are high. Any idea's guys?
 19 December 2010 11:38 AM
User is offline View Users Profile Print this message



Phoenix151

Posts: 77
Joined: 14 February 2007

Were you using the non-trip setting on your loop tester?

What make/model of loop tester is it?

There is greater inaccuracy with the non-tripping settings, but some loop testers cope better than others... The effect is more prevelent where there is some oxidisation on the earth path... have you tried inserting the loop tester plug in and out a few times and rocking the switch back and forth, that can clean things up a bit

have
you could try doing a full loop test, which will trip the RCD and then repeating the test to see if its cleared some of the HR bits.

Can't advise you to bridge out the rcd for tesing as that is frowned upon these days! lol


As an illustration of the effect, I came across a B/E conduit system used as the CPC for lighting in a 1960's build that wasn't in very good shape... if you tested an exposed conductive part with a low ohms meter to known good earth you'd get something between 50 and 100 ohms same with a non trip loop test, but a loop tester on high current would return a sensible value on most parts of the installation (but not all, had a few that were outside of spec for the C10 mcbs. Looking at the conduit terminations into the DB where the board had been recently changed, they didn't look too great, and given that one of them I could get hold of and move around in the hole that had been drilled a bit baggy in the top of the DB, and this was acting as the CPC for the circuit, did not inspire confidence in the reliability of the earth path ... I can't help but feel that as there were no RCDs fitted anyway, that the situation would have been better before the boards were changed when the old bill rewireable boards were still in service (outline in the paintwork on the wall)... at least the conduits were probably bushed into the board properly lol

Though this is not likely to be the case in your situation, as you have got good low ohms values... and you haven't got corroding conduit joints in the ceiling (I hope )
 19 December 2010 01:43 PM
User is offline View Users Profile Print this message



PatHammond

Posts: 215
Joined: 04 December 2008

try zeroing your leads then take the Ze, R1 + R2 and the Zs in succession, then see how the results stack up

-------------------------
4.5 day wonder...had to start somewhere
 19 December 2010 02:13 PM
User is offline View Users Profile Print this message


Avatar for JonSteward.
JonSteward

Posts: 576
Joined: 04 December 2007

Guessing you've got RCD's, try measuring either side as the coils on some makes add a significant impedance adding to the overall loop measurement. Also, try your loop between L an N should be nice and low.
 19 December 2010 08:01 PM
User is offline View Users Profile Print this message



sirsnortalot

Posts: 38
Joined: 25 July 2008

Cheers for the thoughts guys! My tester is a Megger MFT1553 and yes, it was set to non-tripping loop. I didn't try it on full loop test but will give it a shot when I'm back. I've never seen it produce these figures from a new install before. Wiring is in twin and earth so oxidised earth paths are not present. Will try zeroing the leads next time I'm there and see how that goes. Interesting idea Jon....this is the first time I've fitted this particular consumer unit and I did wonder at the time if the RCD could be affecting the readings. Shall try measuring both sides and see what that shows
 19 December 2010 08:11 PM
User is offline View Users Profile Print this message


Avatar for dickllewellyn.
dickllewellyn

Posts: 1150
Joined: 19 March 2010

You'll have difficulty zeroing the leads for Loop tests on the megger, but you will find that the non trip reading varies wildly. You can do three icrntical tests consecutively and get three different results. If the calculated ZS is much lower, then I would suggest if the highest tested Zs is below limits you'll be fine!

-------------------------
Regards
Richard (Dick)

"Insert words of wisdom and/or witty pun here"
 19 December 2010 08:16 PM
User is offline View Users Profile Print this message



sirsnortalot

Posts: 38
Joined: 25 July 2008

Fair enough! Cheers for your thoughts. The highest Zs reading was still within the 80% so should be fine......was just desperate to find out why so high as it's never happened before! Cheers
 19 December 2010 08:25 PM
User is offline View Users Profile Print this message



alancapon

Posts: 5772
Joined: 27 December 2005

Loop tests are done by placing a resistor across the test connections and seeing what the voltage drop is. From this, the loop resistance can be calculated. On the normal setting, the current draw is usually about 25A. In the "no trip" mode, a much lower current is used, leading to much smaller variations in voltage being looked at. It is possible at these low levels, voltage changes caused by something happening in the network are more significant, leading to higher errors in the reading.

Regards,

Alan.
 19 December 2010 10:52 PM
User is offline View Users Profile Print this message



chrishaworth

Posts: 247
Joined: 08 March 2003

One thing that springs to mind is the circuit breaker not being correctly mounted on the busbar. Came across one a while back where only the end ones had been tightened up - the sparks must have been distracted half way through the job. The house had run for 3 years before a problem with the lighting showed up. Another one is some RCDs have a strange effect on loop readings. Try a reading at the RCD output.

Cheers
Chris
 20 December 2010 08:54 AM
User is offline View Users Profile Print this message



sirsnortalot

Posts: 38
Joined: 25 July 2008

I'm pretty sure the breakers are properly seated (Will be kicking my own ***** if they aren't!!) but will have a check today. I'm staring to think that it could well be the main RCD's in that board as it's the first time I've used it. I'm gonna try re-testing the rcds and see what comes of that. Thanks again for all your thoughts peeps.....much appreciated
 20 December 2010 10:44 AM
User is offline View Users Profile Print this message



Jaymack

Posts: 4647
Joined: 07 April 2004

What is the make of the RCD? Those to BS4293 and 63A 30mA say, can give unexpected readings using a Megger 1502/2 Meter on the No-Trip Setting.

Try a Zs test on the load and incoming terminals on the RCD, compare the difference in ohms.

The no-trip test uses a very short pulse and the RCD has significant reactance, so the test figure appears to be too high!

Regards
 20 December 2010 01:41 PM
User is offline View Users Profile Print this message


Avatar for Legh.
Legh

Posts: 3489
Joined: 17 December 2004

Loop readings can be rather problematic at times.
I usually allow between 0.3 and 0.5 Ohms difference between High loop and no-trip loop test readings

But I have found on one installation where the final circuits were fed from Betapack lighting supply the results for 'no-trip' coming out as expected and on 'high loop' the settings gave much higher values.
I have never really got to the bottom of it and assumed the best case

Legh

-------------------------
Why do we need Vernier Calipers when we have container ships?

http://www.leghrichardson.co.uk

"Science has overcome time and space. Well, Harvey has overcome not only time and space - but any objections."
 20 December 2010 02:44 PM
User is offline View Users Profile Print this message



sirsnortalot

Posts: 38
Joined: 25 July 2008

Cheers guys....The RCD is made by British General...usually I would almost always fit Hagar but this board was on offer going pretty cheap.....don't think I'll be fitting one again!! I re-visited site today and ran a non-trip loop test on the load and incoming side. Load side came out at 0.95 while the incomer was at 0.17........I think we have the answer!!
 20 December 2010 05:46 PM
User is offline View Users Profile Print this message


Avatar for JonSteward.
JonSteward

Posts: 576
Joined: 04 December 2007

Don't forget that max Zs for a 30mA in a TN system is 7666 ohms so the BG rcd is ok.
 20 December 2010 08:05 PM
User is offline View Users Profile Print this message



londonlec

Posts: 61
Joined: 29 November 2010

Originally posted by: JonSteward

Don't forget that max Zs for a 30mA in a TN system is 7666 ohms so the BG rcd is ok.


Are you sure?
 20 December 2010 10:08 PM
User is offline View Users Profile Print this message


Avatar for JonSteward.
JonSteward

Posts: 576
Joined: 04 December 2007

Well if max Zs=230/0.03, then yes I'm sure.
 21 December 2010 06:40 AM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3530
Joined: 18 January 2003

Take a look at 411.4.9.

Regards

Geoff Blackwell
 21 December 2010 10:15 AM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 11496
Joined: 13 August 2003

I suspect JonSteward might have a point - there seems to be no 50V rule for TN as there is for TT, 1x Idn should achieve disconnection within 300ms, so satisfying TN's 0.4s requirement. Yes table 41.5 may be used - but "may" isn't "must".

That said, if a TN system has Zs even 1/100th of that kind if level, there's probably something very very wrong, so needing that kind of margin is almost unimaginable.

- Andy,
 21 December 2010 10:25 AM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3530
Joined: 18 January 2003

Can you give an example of a circuit in a TN system that would have a Zs of 7666 ohms - even if we take off the maximum DNO value of 0.8 ohm for Ze (TN), I think you have a problem. Now that is unlikely to be R1 if the circuit works at all - so I would suggest R2 may require some attention .

Regards

Geoff Blackwell
 21 December 2010 10:32 AM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19688
Joined: 23 March 2004

I guess this also needs an appraisal if you are making a demand on the RCD for additional protection though.

Regards

OMS

-------------------------
Failure is always an option
Statistics

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.