Continuity testing with Fluke 1663

[Daryl H]

Hi all,
I was testing a domestic premises In Ireland during the week for Continuity, pre-connection.
usong the Fluke 1663 megger, I had a small issue during the day related to the mA setting on f4.
During continuity testing you can set to 10mA or 250mA. My question is which is the correct setting used for this pre-connection test. Some readings were way off (High resistance) on one setting for sockets, but the same setting gave 0 ohms resistance on the lighting cable then.
Megger was calibrated only 2-3 months ago.
thanks in advance

 

[dodger421]

Hi all,
I was testing a domestic premises In Ireland during the week for Continuity, pre-connection.
usong the Fluke 1663 megger, I had a small issue during the day related to the mA setting on f4.
During continuity testing you can set to 10mA or 250mA. My question is which is the correct setting used for this pre-connection test. Some readings were way off (High resistance) on one setting for sockets, but the same setting gave 0 ohms resistance on the lighting cable then.
Megger was calibrated only 2-3 months ago.
thanks in advance

According to the manual 10mA is used to avoid tripping the RCD, although it even warns the test should be carried out with the circuit de-energised so that shouldn’t occur anyway.

I always use 250mA on my 1664 and any odd readings I get are usually attributed to poor connections with the probes or not zeroing properly after changing from probes to the mains lead or vice versa. I’ve found it pretty sensitive to poor zeroing so worth a check.

 

[static zap]

I always thought too the 250mA test current was the whole reason for using an MFT.
(rather than chancing things with weedy/unsafe/cheap DMM)
Than an alternative with low test current.

 

[static zap]

For any low ohms tests ,RFC end to end or figure of eight tests ..
the best stable results should be achieved with 250mA setting.
Inconsistent 0 results imply. Either poor zeroing , or confusion of
IR and low ohms ranges.
The meter has been calibrated but are the leads any good - if readings wander.

(A cal card -or known sample of cable ..are a good confidence check)

 

[pc1966]

Really you should not trip an RCD by ohms testing - if you do you are testing a live circuit and hopefully you can see that is a bad idea in all cases.

Unless you actually have a N-E fault already, any bond/CPC test will never trip an RCD!

I can't see any reason to use 10mA for electrical supply. Yes, there might just be the odd very low power device that PAT testing in the tens of amps region might cause PCB damage, but if damaged at 250mA I would deem the device to be crap anyway.

You need a decent test current to see what the impedance is going to be like under fault conditions, even then 250mA test it is more of an approximation as real faults will be in the hundred amps and above range.

 

[dodger421]

Really you should not trip an RCD by ohms testing - if you do you are testing a live circuit and hopefully you can see that is a bad idea in all cases.

Unless you actually have a N-E fault already, any bond/CPC test will never trip an RCD!

I thought so too, then wondered about the effect on split load boards. Maybe doing a test on a live installation where you’ve locked off the circuit you’re testing 250mA would cause the RCD to trip?

Say cross connecting at the halfway point of a RFC rather than at the CU to allow the owner to have some power elsewhere while you fault find. Obviously it’s not something you’re going to be doing very often, if ever, but I guess that might be the reason for the 10mA test?

 

[pc1966]

Most MFT and IR testers like the Meggers that have a low-ohms range will not test if they detect more than the odd volt of AC anyway.

So even with split boards, etc, I can't see any testing in a manner that is safe/sane to be doing in the first place tripping the RCD.

Not to mention Type A/AC RCDs that will largely ignore DC anyway...

 

[dodger421]

The Fluke inhibits the test if it detects voltage too.

I agree, I don’t think there is a safe/sane way of doing a continuity test that would trip an RCD either but Fluke must think there is to have included the option.

For arguments sake maybe doing the test on an isolated circuit would cause any upstream RCDs to trip, but why would you be testing continuity on an isolated circuit without taking the cover off the DB in the first place? And if you’ve got the cover off the installation should be isolated so the RCDs won’t trip anyway.

This is from the manual pages dealing with continuity testing:

“2. 1663/1664 FC: Choose the test lead pair to use with F1 and use the appropriate terminals for this test.

1662: Use the L and PE (red and green) terminals for this test.

This option is for tests at the ring installation or to verify the connection between protective earth and neutral at a mains socket. To avoid tripping the RCD, use the 10 mA test current.”

And

“4. Choose the maximum test current with F4. To not trip an RCD, use the 10 mA setting for a ring installation test that includes the neutral or phase wire.”