Eh up
I don't do EICRs - not enough experience - but the things I see on some reports done by others really dismay me. Perhaps some of my understanding needs improving, I dunno. Or perhaps the CPSes have their own set of guidelines different from the sources I use? In the examples below, I'm taking my knowledge from:
- BS7671
- GN3
- Electrical Safety First, specificallly Best Practice Guide 4 and their FAQs of periodic inspection, testing, certification and reporting.
Gripe number 1: made up numbers, using maths
I'm not talking about the "van test" here, whereby you look at the previous EICR and enter similar numbers, without even getting your tester out. I'm talking about:
- measure Ze
- measure Zs at a random point (lowest ceiling? As per the legend widdler's rant in this thread)
- in the R1+R2 box, put the number that is mathematically Zs-Ze
- if a RFC, determine r1 and rn using 4 x (R1+R2) x (1.5/4), and r2 using 4 x (R1+R2) x (2.5/4) (which would meet the theoretical criteria "r1+r2 = (R1+R2)/4" for a RFC wired in 2.5mm T&E with 1.5mm cpc).
So you end up with a schedule of test results that looks like you've done loads of tests, but in fact done one (per circuit).
I would argue that for an existing installation that is energised, that you have confirmed has a means of earthing, and bonding is in place, you would not need to measure R1+R2. To back this up, I quote "note 4" under table 3.4 of GN3 (p84) - "The earth fault loop impedance test may be used to confirm the continuity of protective conductors at socket-outlets and at accessible exposed-conductive-parts of current-using equipment and accessories." And also the answer to FAQ Q2.12 here.
GN3 also gives guidance to whether ring circuit continuity is necessary (table 3.4, p 83): "Where there are records of previous tests, this test may not be necessary unless there may have been changes made to the RFC."
So in many cases, I would argue they don't need to put these numbers! But perhaps there's a culture of, "there's a box for it, it needs a value." But a value in the box implies that you've measured it (e.g. that you've tested continuity of ring circuits) when you haven't. How would the person reading the report know?!?
It's even more obvious when there's more than one DB, and they've used Ze rather than Zs (@DB) to calculate R1+R2! That's just sloppy!
And to cap it off, an operational limitation: "No continuity tests on circuits that cannot be isolated without affecting your trading." Huh.
Gripe number 2: Fake IR results.
So, you've successfully identified which circuits are three phase, and which are single phase - well done you! You've put ">500" in the former, and "N/A" in the latter for "Line/Line (MΩ)." All other boxes: L/N, L/E and N/E, you've put ">500." Uh huh. Despite the operational limitation, "No IR tests applied to circuits containing equipment subject to damage." Nah, sorry, don't buy it. But the customer thinks, "Ooh, that all sounds ok."
Again: if you've not tested it - and GN3 gives guidance here too - why put numbers in the boxes?!?
Gripe number 3: C2 for things that don't need it
- Lack of diagrams, charts or schedules at or near equipment (all circuits marked up on boards)... table 3.5, GN3, p88 (example of C3 code)
- Lack of "415V" warning notice... 514.10.1 ?!?
- White flex used for exterior equipment exposed to sunlight (where no visible sign of deterioration is present... and after all, its IR is >500MΩ!)
It just dismays me. Am I right to be dismayed?
I don't do EICRs - not enough experience - but the things I see on some reports done by others really dismay me. Perhaps some of my understanding needs improving, I dunno. Or perhaps the CPSes have their own set of guidelines different from the sources I use? In the examples below, I'm taking my knowledge from:
- BS7671
- GN3
- Electrical Safety First, specificallly Best Practice Guide 4 and their FAQs of periodic inspection, testing, certification and reporting.
Gripe number 1: made up numbers, using maths
I'm not talking about the "van test" here, whereby you look at the previous EICR and enter similar numbers, without even getting your tester out. I'm talking about:
- measure Ze
- measure Zs at a random point (lowest ceiling? As per the legend widdler's rant in this thread)
- in the R1+R2 box, put the number that is mathematically Zs-Ze
- if a RFC, determine r1 and rn using 4 x (R1+R2) x (1.5/4), and r2 using 4 x (R1+R2) x (2.5/4) (which would meet the theoretical criteria "r1+r2 = (R1+R2)/4" for a RFC wired in 2.5mm T&E with 1.5mm cpc).
So you end up with a schedule of test results that looks like you've done loads of tests, but in fact done one (per circuit).
I would argue that for an existing installation that is energised, that you have confirmed has a means of earthing, and bonding is in place, you would not need to measure R1+R2. To back this up, I quote "note 4" under table 3.4 of GN3 (p84) - "The earth fault loop impedance test may be used to confirm the continuity of protective conductors at socket-outlets and at accessible exposed-conductive-parts of current-using equipment and accessories." And also the answer to FAQ Q2.12 here.
GN3 also gives guidance to whether ring circuit continuity is necessary (table 3.4, p 83): "Where there are records of previous tests, this test may not be necessary unless there may have been changes made to the RFC."
So in many cases, I would argue they don't need to put these numbers! But perhaps there's a culture of, "there's a box for it, it needs a value." But a value in the box implies that you've measured it (e.g. that you've tested continuity of ring circuits) when you haven't. How would the person reading the report know?!?
It's even more obvious when there's more than one DB, and they've used Ze rather than Zs (@DB) to calculate R1+R2! That's just sloppy!
And to cap it off, an operational limitation: "No continuity tests on circuits that cannot be isolated without affecting your trading." Huh.
Gripe number 2: Fake IR results.
So, you've successfully identified which circuits are three phase, and which are single phase - well done you! You've put ">500" in the former, and "N/A" in the latter for "Line/Line (MΩ)." All other boxes: L/N, L/E and N/E, you've put ">500." Uh huh. Despite the operational limitation, "No IR tests applied to circuits containing equipment subject to damage." Nah, sorry, don't buy it. But the customer thinks, "Ooh, that all sounds ok."
Again: if you've not tested it - and GN3 gives guidance here too - why put numbers in the boxes?!?
Gripe number 3: C2 for things that don't need it
- Lack of diagrams, charts or schedules at or near equipment (all circuits marked up on boards)... table 3.5, GN3, p88 (example of C3 code)
- Lack of "415V" warning notice... 514.10.1 ?!?
- White flex used for exterior equipment exposed to sunlight (where no visible sign of deterioration is present... and after all, its IR is >500MΩ!)
It just dismays me. Am I right to be dismayed?
Last edited: