IET logo
 
IET
Decrease font size
Increase font size
Topic Title: RCD causing false Zs readings
Topic Summary: Advised by MK to by-pass RCD for test
Created On: 20 July 2013 09:25 PM
Status: Post and Reply
Linear : Threading : Single : Branch
1 2 3 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.
 20 July 2013 09:25 PM
User is offline View Users Profile Print this message



rslane

Posts: 119
Joined: 18 January 2003

Low Ze and low r1+r2 but Ze like three times the sum of them.
The only extra items in the circuit are the items in the consumer unit.
So I measure loop impedance before and after the 63A RCD of the Sentry CU. Supply side 0.34ohm; load side 1.10ohm. (Measured with Kewtech meter in anti trip mode.
So I ring up MK and say serrious problem with their RCD.
They say no, "its the electronics in the modern RCDs" and that I should turn the RCD off and by pass it for the Ze measurement.
So I try this and now get a figure of 0.41 ohm when I connect meter at the end of this botched circuit.
I can't say I am at all happy with this way of testing.
Apart from the extra work and the risks from stuffing temporary links in the CU, it is not a test of the complete cicuit. What if the RCD did actually have high reistance contacts? Of course that wouldn't ever happen, would it? I mean they are not 100A main switches! (I assume that was the reason for the recall.)
 20 July 2013 10:20 PM
User is offline View Users Profile Print this message


Avatar for leckie.
leckie

Posts: 2021
Joined: 21 November 2008

Are you getting mixed up between Zs readings and Ze readings or am I misunderstanding your post? Take both readings from the primary side of the main switch, one with earthing conductor connected, one with it disconnected.
 20 July 2013 10:34 PM
User is offline View Users Profile Print this message


Avatar for leckie.
leckie

Posts: 2021
Joined: 21 November 2008

Sorry rs lane, I think I was confused reading your 1st sentence. I think you said Ze the second time when you meant Zs? Well again, measure prior to the rcd.

However I have noticed that some rcd's give weird IR readings. I know your supposed to isolate them, and they may be vulnerable to testing but, some are not. Are you using no trip and standard earth loop impedance tests? The no trip test often gives higher readings on my meter.
 21 July 2013 03:43 AM
User is offline View Users Profile Print this message



rslane

Posts: 119
Joined: 18 January 2003

Yes, I had meant to say in first sentance that Zs was three times higher.
I had measured Ze at input to main switch with only earth to supply cable connected. Then measured r1+r2 for cooker circuit. Both readings good and low. But on connecting everything up, the overall loop impedance, measured at cooker control panel terminals was about 3x the sum of the two parts and too high and failing.
To investigate I measured loop impedance at supply terminals to RCD and then at RCD output terminals. If the test meter is giving true readings they should be almost the same as the only extra imedance in the circuit will be the contacts in the RCD. If they are not very low resistance it will overheat on full load. 63A x 63A x anything other than low contact resistance = very hot RCD.
Now what MK tell me is that even with anti trip technology I will get false Zs readings when measuring through the RCD.
 21 July 2013 07:56 AM
User is offline View Users Profile Print this message



ssohnoherewego

Posts: 63
Joined: 07 March 2008

I encounter this all of the time. Just put your calculated values onto your test sheet and state why it was calculated.
 21 July 2013 08:46 AM
User is offline View Users Profile Print this message


Avatar for leckie.
leckie

Posts: 2021
Joined: 21 November 2008

If you look at a standard rcd circuit diagram, usually printed on the rcd, it never shows "fancy" electronics. Different manufacturers much have different circuit electronic arrangements I suspect. So some, in this case MK, may give different test results. I haven't noticed differences as large as rslane has described on Crabtree, control gear, or Hager that I have used. However I have noticed that when doing IR tests with L/N connected together to earth, that sometimes I get a low reading across the rcd, with the rcd off. If I do the same test with the rcd out put disconnected the reading on the circuits is OK. But it doesn't always happen. I have put this down to electronics with the rcd. By the way, who records the rcd as being an item vulnerable to testing on EIC's?
 21 July 2013 12:26 PM
User is offline View Users Profile Print this message



perspicacious

Posts: 7267
Joined: 18 April 2006

"By the way, who records the rcd as being an item vulnerable to testing on EIC's?"

Everyone who has read 514.9.1 (iv) and looked at what must amount to 5% of the page area of p402..........

Regards

BOD
 21 July 2013 12:47 PM
User is offline View Users Profile Print this message


Avatar for Legh.
Legh

Posts: 3538
Joined: 17 December 2004

For a slightly different reason.
When Loop testing using an LT5 through a BS 4293 RCD I always connected jump leads so I could get a reasonable Zs. Now I test either side and add the values together.
Coupled with a RCD time test and a ramp test to ensure, to some degree, a level of correct operation.

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."
 21 July 2013 01:53 PM
User is offline View Users Profile Print this message



rslane

Posts: 119
Joined: 18 January 2003

"use calculated values"
Thanks. It may be what we have to do for the moment.
But I am still not happy that the manufacturers have pushed the electrician into a corner by developing a device that makes one the basic safety tests impossible with industry standard instruments.
So whats the proper long term solution?

a) Redesign of RCDs to make them testable in circuit with existing instruments.

b) Redesign of RCDs to have a foolproof test override facility thats lets us do an impedance test of the complete loop.

c) New test equipment that will give the correct answer for the as installed complete circuit.

d) Get the regs amended to say we don't need to do the test.

Now I can't see a) or b) happening any time soon, the manufacture's would need considerable preasure on them to spend that sort of money. And in any case what of all the "untestable" units in sercice.
And I can't see d) happening, nor would I advocate it. It would be the thin edge of the wedge. Next step, use calculated values for everything and throw away the test equipment. In theory if we have designed the instalation to the book it will be perfect. But then the unexpected happens and what we think we have is not what we actually have. And so we test to make sure. A calculated value for Zs fails to find a high resistane RCD, a high resistance MCB or a poor connection in the consumer unit. But then manufacurers never supply faulty equipment, none of us ever makes a poor connection and nothing ever detioriates!!
So in my opinion there still needs to be a way to test.

To me it is starting to look like c), new test equipment and yet another cost for the electrican to pass on to the customer who won't pay and will use a cowboy.
WHAT'S THE ANSWER?
 21 July 2013 03:12 PM
User is offline View Users Profile Print this message



weirdbeard

Posts: 1652
Joined: 26 September 2011

Originally posted by: rslane



WHAT'S THE ANSWER?


411.4.9?
 21 July 2013 04:05 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 6326
Joined: 18 January 2003

Are you absolutely certain that the MK RCD is NOT ADDING TO THE IMPEDANCE OF THE CIRCUIT?

Are you absolutely certain that the high figure is not the TRUE FIGURE?

If so can you explain to me why?

Andy
 21 July 2013 04:31 PM
User is offline View Users Profile Print this message



rslane

Posts: 119
Joined: 18 January 2003

You are quite right - it might be adding to the impedance and the reading might be right. That is why I am still worried.
 21 July 2013 04:33 PM
User is offline View Users Profile Print this message



rslane

Posts: 119
Joined: 18 January 2003

PS sorry to be so impersonal on this site. rslane = Bob.
 21 July 2013 04:40 PM
User is offline View Users Profile Print this message



alancapon

Posts: 5831
Joined: 27 December 2005

Originally posted by: leckie
. . . I have noticed that some rcd's give weird IR readings. I know your supposed to isolate them, and they may be vulnerable to testing but, some are not. . .

All RCDs are vulnerable to IR testing. Usually an RCD will be connected such that the supply to its internal circuitry is disconnected when it is off. The test resistor and switch is usually wired such that the "load side" phase is connected to the "supply side" neutral. If you connect it the other way round, operating the test button may energise the neutral of the installation and leave its internal electronic circuitry energised.

Regards,

Alan.
 21 July 2013 05:03 PM
User is offline View Users Profile Print this message



rslane

Posts: 119
Joined: 18 January 2003

By the way Andy I see you post far more than I do on the forum but it looks as if we both joined on the same dayy, 18.1.03!
Regards, Bob
 21 July 2013 06:02 PM
User is offline View Users Profile Print this message


Avatar for leckie.
leckie

Posts: 2021
Joined: 21 November 2008

Sorry, I know that RCD's are vulnerable to testing if they are not switched off, what I am saying is that some are vulnerable even when they are switched off I think, as some read even when off. I'm talking about doing rcd output lives linked and tested to earth. I just wondered why some are and some are not?


Why I mentioned the recording of rcd's being included as vulnerable on EIC's is because I do, and yet I very, very, rarely see it on the many certificates I see issued by others.
 21 July 2013 07:45 PM
User is offline View Users Profile Print this message



weirdbeard

Posts: 1652
Joined: 26 September 2011

Originally posted by: perspicacious

"By the way, who records the rcd as being an item vulnerable to testing on EIC's?"


Everyone who has read 514.9.1 (iv) and looked at what must amount to 5% of the page area of p402..........



Hi BOD, I can't see how an RCD is vulnerable to damage from typical testing, how else are you supposed to fill in the schedule of test results RCD columns?
 21 July 2013 08:37 PM
User is offline View Users Profile Print this message


Avatar for leckie.
leckie

Posts: 2021
Joined: 21 November 2008

Cheeky monkey weirdbeard
 21 July 2013 10:22 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 6326
Joined: 18 January 2003

"By the way Andy I see you post far more than I do on the forum but it looks as if we both joined on the same dayy, 18.1.03!
Regards, Bob"

First off I spend hours talking to myself, secondly I suffer cabin fever on account of working long hours on my own
 22 July 2013 07:13 AM
User is offline View Users Profile Print this message


Avatar for davezawadi.
davezawadi

Posts: 2816
Joined: 26 June 2002

RCDs should be designed to deal with mains spikes (and therefore insulation testing) without damage. If I ever found a make which failed permanently on a 500V test I would be at the manufacturer very hard indeed (and I haven't). It is imperative that safety devices (such as RCDs) can survive in the real world because any failure is not visible yet leads to significant danger to the user.

It may be that some RCDs give low readings of IR, but they should never be subject to failure. Whoever wrote that bit of the regs BOD certainly did not intend to include RCDs, and in my view any accessory or appliance which cannot withstand a 500V test without damage is not fit for purpose. Some poor quality items may not withstand 500V between live and neutral, but are likely to fail prematurely as a result.

There is history in this, as over voltage protection, surge suppressors etc are fairly new components (only about 25 years) so have not fully arrived in the vocabulary of product design in some areas, and can be left out to reduce costs with immediate benefit to the manufacturer, until the complaints roll in!!

-------------------------
David
CEng etc, don't ask, its a result not a question!
IET » Wiring and the regulations » RCD causing false Zs readings

1 2 3 Next Last unread
Topic Tools Topic Tools
Statistics

See Also:



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