Reverse polarity in shower on a PIR?

[BigJon]

Hi all, just a quick query.
I was discussing the consequences of reverse polarity in a shower with a work mate today. To set the scene, the mains "IN" at the 45A shower switch is connected correctly but the "Load" side is connected in reverse polarity.
When debating if/when/how the fault would be found during a PIR there were 2 sides.
In a nutshell I said it should, as you would not get an R1+R2 reading at the shower and would therefore need to be investigated and hopefully the fault would be found.
He said he takes his R1+R2 from the pull cord and he would have got the reading from the "MAINS" and that the reverse connection would be found visually.
What would be the consequences of a reverse polarity in (ONLY) the shower, would it still function as it should?
If it was protected by RCBO with functional earth would this trip?
I presume if it was just a standard 30ma DP RCD then there would be no imbalance to detect?

Thanks

 

[sparkticus]

In a nutshell I said it should, as you would not get an R1+R2 reading at the shower and would therefore need to be investigated and hopefully the fault would be found.

Well, when you take the shower cover off you will see the reverse polarity because the red/brown will be where the black/blue should be. If you ignored the colours than of course you will get an open circuit R1+R2.
But you would see the error, correct it then perform an R1+R2.

He said he takes his R1+R2 from the pull cord

That is not an R1+R2. R1+R2 is taken at the most distant point in a circuit. In this case at the shower.

What would be the consequences of a reverse polarity in (ONLY) the shower, would it still function as it should?

It would function as normal. The main switch and thermal protective device would be in the Neutral line which would leave the heating element live.

If it was protected by RCBO with functional earth would this trip?
I presume if it was just a standard 30ma DP RCD then there would be no imbalance to detect?

An RCBO and an RCD would disconnect as normal in a fault condition including an earth fault condition.

 

[JohnW2]

When debating if/when/how the fault would be found during a PIR there were 2 sides.
In a nutshell I said it should, as you would not get an R1+R2 reading at the shower and would therefore need to be investigated and hopefully the fault would be found.

True - but would you really open up a shower when doing a PIR?

He said he takes his R1+R2 from the pull cord and he would have got the reading from the "MAINS" and that the reverse connection would be found visually.

Again, probably true - although that would strictly not be R1+R2 for the circuit. A third option is that one might measure R1+R2 (again, not true R1+R2) on the shower side of the pull switch, which (assuming you went by insulation colours) essentially takes us back to your scenario above.

What would be the consequences of a reverse polarity in (ONLY) the shower, would it still function as it should?
If it was protected by RCBO with functional earth would this trip?
I presume if it was just a standard 30ma DP RCD then there would be no imbalance to detect?

I can't really think of anything that would prevent the shower and protective devices (even an RCBO with functional earth) functioning totally normally.

Kind Regards, John.

 

[sparkticus]

True - but would you really open up a shower when doing a PIR?

That's a good question and the answer is not always if you take a statistical approach to a PIR which would be typical. Having said that, I would personally always take the cover off a shower and would be more likely to perform an R2 with a wonder lead.

The reason I always pick on a shower is that they do seem to harbour reverse polarity connections - I promise I am not blaming the plumbers

 

[riveralt]

That's a good question and the answer is not always if you take a statistical approach to a PIR which would be typical. Having said that, I would personally always take the cover off a shower and would be more likely to perform an R2 with a wonder lead.

I love w a n d e r leads too.

 

[sparkticus]

That's a good question and the answer is not always if you take a statistical approach to a PIR which would be typical. Having said that, I would personally always take the cover off a shower and would be more likely to perform an R2 with a wonder lead.

I love w a n d e r leads too.

Ah yes, I must have been thinking of my wonderful 50 meter KewTech wander-lead

 

[riveralt]

True - but would you really open up a shower when doing a PIR?

Kind Regards, John.

Absolutely and for two very good reasons.
Firstly, how else are you going to get the correct test results for the schedule of test results?
Secondly, the electric shower must be one of the most potentially dangerous areas of any household.
Always seems strange to me that the casings are not more securely fitted - sometimes just one screw between you and death

 

[riveralt]

Ah yes, I must have been thinking of my wonderful 50 meter KewTech wander-lead

Blooming heck you must be rich!!!!
I make do with a 20m extension cable reel and long arms.

 

[JohnW2]

True - but would you really open up a shower when doing a PIR?

That's a good question and the answer is not always if you take a statistical approach to a PIR which would be typical. Having said that, I would personally always take the cover off a shower and would be more likely to perform an R2 with a wonder lead.

I agree that it's probably particularly desirable to look inside a shower but, as you say, I imagine it may not get done in a 'sampling' ('statistical') PIR. I don't think it's really a problem with modern/recent showers, but with some of the early electric showers I would have been very nervous about any 'avoidable meddling' with the cover - since they often had seals which could easily be damaged.

Indeed, although periodic inspection and testing is desirable, and certainly well-intentioned, I often wonder to what extent 'tampering' with the installation (particularly cable terminations) actually introduces problems as well as detects them. It reminds me in a sense of the 'no win' situation which existed with imperfect typists in the pre-word-processor days. Unless one had an incredibly good typist, it was effectively impossible to get a perfectly typed long document without the use of some Tippex etc .. a draft would be typed, one would mark all the required changes/corrections ... the typist would then retype the document, correcting those errors, but at the same time introducing a set of new errors ... and that process could go on 'ad infinitum'!

Kind Regards, John.

 

[sparkticus]

Ah yes, I must have been thinking of my wonderful 50 meter KewTech wander-lead

Blooming heck you must be rich!!!!
I make do with a 20m extension cable reel and long arms.

Yea I used to do that but those KewTech leads are bl--dy good. I borrowed one about a year ago and was immediately hooked on it. The wire is just like
a larger CSA version of a test lead and lays flat etc. I paid under 100 for it through an NICEIC promotional I think.

 

[westie101]

often wonder to what extent 'tampering' with the installation (particularly cable terminations) actually introduces problems as well as detects them

Indeed a question often raised at work. The testing requirements after work on our HV system require L - L & L - E testing on all 3 phases. This can mean us having to remove the lid and disconnect the HV windings on, sometimes a large number, connected transformers.
This always introduces risks of further faults (or even tools dropped into the tanks)

 

[sparkticus]

Indeed, although periodic inspection and testing is desirable, and certainly well-intentioned, I often wonder to what extent 'tampering' with the installation (particularly cable terminations) actually introduces problems as well as detects them.

I have almost developed a fear of testing RCDs for that very reason. I test a lot of them in a year but I think I have "broken" 5 of them while testing in the past 12 months. When I say testing I mean the standard 0.5x - 1x & 5x tests. I thought it was my tester at one point but I have two test instruments (A Seaward and a Megger) the megger was calibrated shortly before "breaking" an RCD.

 

[JohnW2]

True - but would you really open up a shower when doing a PIR?.

Absolutely and for two very good reasons.
Firstly, how else are you going to get the correct test results for the schedule of test results?

Yes, if the scope of the PIR is 'full', then one obviously cannot avoid doing it.

Secondly, the electric shower must be one of the most potentially dangerous areas of any household.

Agreed. I did say that it was something which it was 'particularly desirable' to look at.

Always seems strange to me that the casings are not more securely fitted - sometimes just one screw between you and death

Quite so - and every time I see a shower without its cover, I always re-live the same amazement I experienced the first time I saw inside one - prior to which I had assumed there would be much more extensive and obvious features to keep the water and electricity apart!

Kind Regards,John.

 

[JohnW2]

Indeed a question often raised at work. The testing requirements after work on our HV system require L - L & L - E testing on all 3 phases. This can mean us having to remove the lid and disconnect the HV windings on, sometimes a large number, connected transformers. This always introduces risks of further faults (or even tools dropped into the tanks)

Indeed. In would be very interesting to discover some statistics (for 'both sides of the cutout') on the number of faults/problems that are actually created by testing.

Kind Regards, John.

 

[JohnW2]

I have almost developed a fear of testing RCDs for that very reason. I test a lot of them in a year but I think I have "broken" 5 of them while testing in the past 12 months. When I say testing I mean the standard 0.5x - 1x & 5x tests. I thought it was my tester at one point but I have two test instruments (A Seaward and a Megger) the megger was calibrated shortly before "breaking" an RCD.

As I'm sure you realised, I was thinking more about undoing and re-doing terminations, flexing cables, removing and replacing accessories from back boxes etc. etc., as well, of course, as problems created by carelessness during testing (killing things with IR tests, forgetting to reconnect cables/bonding/whatever etc.etc.) - but what you say is interesting.

I'm actually surprised that RCDs can be killed by 'standard testing'. In my (very limited) experience, I haven't seen that happen. Any idea of the mechanism?

Just consider yourself lucky that you are not expected to test MCBs!

Kind Regards, John.