UK I am getting very high zs readings

[banny07]

Hi Guys. When measuring zs I get very high reading. Ze is 0.3 (tns), zdb is 0.21. r1 and rn are is 0.42. r2 is 0.72 , r1+ r2 is 0.29 . when I test zs it 8.26 ohm new to testing. I even buypassed rcbo but it stays same. Do you think my mft if faulty? Any thoughts guys? I am new to testing.

 

[Julie.]

None of those figures make sense to me.

I assume zdb is at the db after measuring ze so should be higher than ze not lower.

How are you doing these tests?

What is the state of the circuit for each test?

 

[westward10]

If Ze was measured with parallel paths removed then Zs at the DB with all paths reinstated can be lower.
If you believe your test equipment to be faulty then recheck it at your known source of Zs at the DB.

 

[banny07]

None of those figures make sense to me.

I assume zdb is at the db after measuring ze so should be higher than ze not lower.

How are you doing these tests?

What is the state of the circuit for each test?

I am teasting ze with main earth disconnected, zdb with earth connected.

 

[banny07]

None of those figures make sense to me.

I assume zdb is at the db after measuring ze so should be higher than ze not lower.

How are you doing these tests?

What is the state of the circuit for each test?

On lighing circuit r1+r2 is 0.86 but zs comes up to 7.66 ohm.on another radial circuit r1+r2 is 0.09 but zs jumps up to 0.52.

 

[timhoward]

Formatted slightly differently....
Please note r1, rn and r2, and R1+R2 are different things.

Ze is 0.3 (tns), (OK)
Zs@db is 0.21. (OK - parallel paths I assume)
Ring Final Circuit:
End to End r1 and rn are is 0.42. (OK)
End to End r2 is 0.72 (OK, assuming twin and earth - expected cpc= * 1.67 and is close enough)

R1+ R2 is 0.29 (please use capitals for this - OK - 0.42+0.72 / 4 = 0.28 )

. when I test zs it 8.26 ohm new to testing. NOT OK!

What make and model of MFT do you have?
Are you using different test lead / probe / clip for Zs?
If recent Megger are you selecting 2 wire or 3 wire non-trip Zs?

 

[westward10]

Have you tried a line to neutral loop. I often do this with obscure loops, one it verifies your meter is working and two it verifies the line and if this is okay you could have an earthing issue. With higher than expected loops return to the board and verify your loop at the board again it isn't uncommon for a TN-S system to intermittently break down.

 

[banny07]

Have you tried a line to neutral loop. I often do this with obscure loops, one it verifies your meter is working and two it verifies the line and if this is okay you could have an earthing issue. With higher than expected loops return to the board and verify your loop at the board again it isn't uncommon for a TN-S system to intermittently break down.

R1 and Rn are same on lighting circuit

Formatted slightly differently....
Please note r1, rn and r2, and R1+R2 are different things.


What make and model of MFT do you have?
Are you using different test lead / probe / clip for Zs?
If recent Megger are you selecting 2 wire or 3 wire non-trip Zs?

Megger 1741+ . using plug leads for sockets and using original test leads to test lighting circuit. using 2 low or 3 low leads with same result.

 

[timhoward]

Well something is afoot!
I think you need to retrace steps and do several Zs@DB in a row to check results are repeatable, and also double check main earth is properly reconnected.
If the high current loop is working consistently and reliably for the board, there is no reason it shouldn't also work reliably when testing the final circuit with RCBO bypassed.
Try (carefully!) using the exact same probes at the back of a socket to rule out a faulty plug adaptor test lead.

It's plausible that the non-trip Zs is being affected by interference/noise, but the high current should be fine.
(There is a way to use high-current with RCBO's but I won't confuse the issue by talking about that yet.)

 

[banny07]

Please note r1, rn and r2, and R1+R2 are different things.

Thanks. point noted.

 

[banny07]

It's plausible that the non-trip Zs is being affected by interference/noise, but the high current should be fine

I used 2 lead on high setting when I bypassed RCBO.

 

[Julie.]

I am teasting ze with main earth disconnected, zdb with earth connected.

Fair enough, although you should be testing for zs using main earth only, removing parallel paths.

Zs can vary a bit even from test to test, but shouldn't be that far out.

You can easily check your mft for correct zs, at home or wherever, measure the zs (say 1. 3 ohm), then measure it again but place a 1 or 2 ohm resistance between your meter leads and the same point, the new zs readings should increase by the value of the resistance you added. (To 3.3 ohm in this example) if it's all OK, them you are measuring something wrong, or you have a poor connection when you reconnected either the circuits, or the main earthing.

There is a possibility of cross voltage from another property, is the bonding correct?

 

[timhoward]

I was about to suggest similar to @Julie -
Another thought - if you have a long wander lead with you, measure it's resistance on the continuity range.
Then do Zs@DB again normally. Then Zs@DB with the wander lead between the tester and the CPC bar. It should increase by the resistance first measured.

 

[banny07]

you should be testing for zs using main earth only, removing parallel paths

I didn't know that. I thought zs db/circuit should be tested with all parallel path connected.

 

[banny07]

I was about to suggest similar to @Julie -
Another thought - if you have a long wander lead with you, measure it's resistance on the continuity range.
Then do Zs@DB again normally. Then Zs@DB with the wander lead between the tester and the CPC bar. It should increase by the resistance first measured.

I will do this when I go back there. I have another mft (1553) which I will be also using . My concern is how to proceed if It also gives same results.

 

[westward10]

If you do you must go back to the beginning. As @timhoward suggests do repeated Zs at the DB to ensure it is constant then repeat your dead tests.

 

[Julie.]

I didn't know that. I thought zs db/circuit should be tested with all parallel path connected.

There are two reasons to measure ze/zdb , the first is with everything connected I.e. all parallel paths - this will give the maximum fault level, you can then check the protective devices are suitable for that fault level; the second which applies to ze/zdb and zs is with the minimum fault level likely to occur - so at the furthest point with all the parallel earth paths removed, in this case it is to ensure that the protection will operate in the required time in the worst case

 

[banny07]

There are two reasons to measure ze/zdb , the first is with everything connected I.e. all parallel paths - this will give the maximum fault level, you can then check the protective devices are suitable for that fault level; the second which applies to ze/zdb and zs is with the minimum fault level likely to occur - so at the furthest point with all the parallel earth paths removed, in this case it is to ensure that the protection will operate in the required time in the worst case

so at the furthest point with all the parallel earth paths removed

by disconnecing main bonding conductors to water and Gas?

 

[timhoward]

I didn't know that. I thought zs db/circuit should be tested with all parallel path connected.

That's actually how I was taught and what GN3 appears to say; Julie's logic and reasoning is spot on, and in this case I might have done it anyway to help build a picture of what is going on.

There are two reasons to measure ze/zdb , the first is with everything connected I.e. all parallel paths - this will give the maximum fault level, you can then check the protective devices are suitable for that fault level; the second which applies to ze/zdb and zs is with the minimum fault level likely to occur - so at the furthest point with all the parallel earth paths removed, in this case it is to ensure that the protection will operate in the required time in the worst case

No argument about the logic/reasoning. I'm also mindful that GN3 implies that in all circumstances where you have live final circuits (required for these tests) you should have the bonding connected. But it also doesn't specifically describe the Zdb test as far as I can tell.

It only seems to discuss two cases - testing external earth loop impedance with main earth disconnected, and all other earth loop impedance tests where it says that "for safety and practical reasons neither the connection with Earth nor bonding conductors are disconnected". (Page 66)
So while I think it might have been a handy test to do I wouldn't say the OP is doing it wrong either.

 

[westward10]

Zdb is nothing more than a Zs.

 

[littlespark]

Did anyone else only read the first five words of the thread title?

 

[Pretty Mouth]

R1 and Rn are same on lighting circuit

I think what @westward10 meant was to carry out a L to N live loop test, as opposed to a dead continuity test of R1 and Rn

 

[Julie.]

That's actually how I was taught and what GN3 appears to say; Julie's logic and reasoning is spot on, and in this case I might have done it anyway to help build a picture of what is going on.

No argument about the logic/reasoning. I'm also mindful that GN3 implies that in all circumstances where you have live final circuits (required for these tests) you should have the bonding connected. But it also doesn't specifically describe the Zdb test as far as I can tell.

It only seems to discuss two cases - testing external earth loop impedance with main earth disconnected, and all other earth loop impedance tests where it says that "for safety and practical reasons neither the connection with Earth nor bonding conductors are disconnected". (Page 66)
So while I think it might have been a handy test to do I wouldn't say the OP is doing it wrong either.

That's really why measuring R1+R2 dead, and calculating zs from the ze/zdb + the measured circuit loop resistance is the usual practice, it allows you to obtain the minimum zs whilst not exposing users to an un-bonded system.

 

[banny07]

Did anyone else only read the first five words of the thread title?

good shout!

 

[pc1966]

by disconnecting main bonding conductors to water and Gas?

Exactly.

Just as @Julie. says when you have the bonding in place you might have parallel paths lowering the supply's R2, hence you might see a much higher PFC and that is something to take in to account when verifying that OCPD is able to safely interrupt it.

However, such parallel paths are not guaranteed to remain. So if someone replaces some of a metal water pipe with plastic you might lose a parallel path and so the only safe assumption is the supply earth is there for faults. So checking Ze allows you to check that under a change of bonded part(s) you are still able to disconnect fast enough under fault conditions.

 

[mart spark]

Technically speaking if performing zs measurements then main protective bonding must in place as the installation is live and if there are fault conditions, removing it to take away any parallel paths will result in extraneous conductive parts becoming live in said fault conditions. I could be wrong though.

 

[pc1966]

Yes, typically the only way to measure Ze is to isolate the installation supply (usually via the incoming switch) so it is completely dead, and then disconnect the supply earth and measure to it, and put it back immediately. Removing the bond conductors might do it, but as you can have links from the installation CPC to extraneous metalwork via boiler CPC, immersion heater CPC, or supplementary bonding, etc, you can never be sure.

Measuring Zs is simpler as everything is still in place. Of course if the supply earth is faulty then you would risk raising the CPC and any extraneous parts with a high impedance to potentially dangerous voltages, but most MFT will usually throttle the test current if they see a voltage change that could exceed 50V to keep things reasonably safe.

 

[Inteificio]

Megger 1741+ .

I'm considering getting one of these.

It is supposed to tell you if the Zs reading is reliable. It's one of the things that made this meter stand out.

I normally would ask how you are finding the unit...I might hold off until you have checked it against something else =)