Plumber asked me to connect an Ariston 3kW water heater for him but when I checked the phase to earth resistance and found it was less than 15k ohm across the immersion heater terminals I refused to connect it and called the service engineer out. He wanted to know what happens when I switched it on! I said its not safe to switch on and it would be likely to trip the RCD (not that there was one).
So I started looking for a spec figure as to what the insulation should actually be and can't find one - I guess in the order of mega ohms as I would expect to be able to pass a circuit insulation test with the heater switch on.

So what should the insulation resistance be?

The manufacturer's help technical help line told me I was using the wrong terminology - "Insulation is what you put round the tank to keep it warm mate!"

Also not impressed that their instructions suggested use of 13A plug and socket as alternative to FCU.

Probably just a bit of damp got in to the element ends in storage. If it was me I'd run it up through a variac so it gets gently warmed.
Alternatively, put it in SERIES with a 100W bulb, this will run the heater at much reduced power, but should be sufficient to drive out any moisture; then do your IR test.

You might be onto something - plumber got everything soaked.

If there's no rcd in circuit just stick the mains on it for half an hour and then do an I/R test,

Dave.

Thanks for the suggestions - perhaps I am worrying too much!
There isn't an RCD so I could fire it up, and then recheck it and I guess it may well have dried out.
But do you know what the spec is for an immersion heater's insulation resistance?

Not sure what the 'spec' is but I would expect >500M ohm.

RCD should hold at 15k ohms.

When I used to play around with 3 phase boilers, steam ovens etc; the minimum IR for an element of that nature was 1 Mohm and as far as I know that has not changed, you will find when they are new they are often very low but after a couple of uses the IR increases remarkedly..

rehards

-------------------------
"Take nothing but a picture,
leave nothing but footprints!"
-------------------------
"Oh! The drama of it all."
-------------------------
"You can throw all the philosophy you like at the problem, but at the end of the day it's just basic electrical theory!"
-------------------------

interesting the" In-service inspection and testing of electrical equipment 3rd edition" COP page 75 note 1 suggests doing as above.

Indeed primo - I had applied ohms law for 15K and even at 250V an RCD should indeed hold - if it is the only leaky device.

Rocknroll - thanks for your input - I will see if I can get it to dry out and get the insulation into the mega ohm range.

Thanks Zeeper - will fire it up and recheck after a while. Having to be a bit careful with this heater - its in a public hall and right above the ladies loo! Not the best of locations but its a replacement job and not many other options now.

Mineral insulated elements are often a bit 'leaky'.
Off the top of my head I think the max permitted leakage is about 3.5mA - which in simple terms would equate to around 66kΩ (in reality somewhat lower might be acceptable as the voltage drops along the element, so lowering the leakage for the same resistance, but even so it's shouldn't be less than say half that - 33kΩ)

- Andy.

Ran the immersion heater for a few hours this afternoon and actually managed to get a reading on the insulation tester, 0.01M, the lowest reading it will give. So that's 10K at 250V, which I think is progress compared with 15k on the multimeter. I should have brought my old analogue insulation tester out of retirement then I would have got an initial reading on it when it's digital successor was refusing to display. I have now left the water heater on, although unless any hot water is drawn off I doubt if the element will actually be working much. I will see what the insulation resistance looks like tomorrow.

But still not sure at what point (what insulation resistance value) I can say it is good enough to sign off.

Good question ,
I would have expected 1M min .

I suppose that as 8K relates to 30mA at 240V then owt above that could be tried under carefully controlled conditions to see if would improve after warming (or better still 24K being 10mA ).

Ilike the answer you got off the "Tech Line"

-------------------------
Regards,
Ebee (M I S P N)

Knotted cables cause Lumpy Lektrik

One web site for Hubble Electric Heaters www.hubbellheaters.com/pdf/HeatElemOM.pdf
spoke of getting the insulation resistance up to at least 20Meg by baking at high temperature in an oven at before installing and switching on. Now they make industrial heaters but the principles are much the same for small scale water heaters so I guess I may have a substandard one.
As I said earlier I am still looking for an actual specification value that I should be achieving to sign it off. However its in an earthed titanium tube (Zs = 0.54 ohm) and protected with a local 13A fuse as well as the breaker in the board, so one could say that if it works, it works; but if it fails it will take the fuse out pretty fast without risk to anyone. Not an ideal situation though.

Leaving it on continuously for a few days was the answer. It now has an insulation resistance of 351M ohms at 500V.

However I still don't know what the pass level is! But I agree with ebee's suggestion of a meg - it feels about right!

I guess the element had absorbed moisture in storage. You would think they would bag the whole heater better and pack it with some desiccant!

Thanks for all your inputs.

Whatever you do, never test a sheathed oven/grill/hotplate element on an electric cooker for insulation resistance!

-------------------------
Norman

The Baby Belling because it was usually fitted with a plug and subjected to PAT testing, always caused howling and moaning as the PAT instrument sometimes turned it down as faulty.

Suppose nowadays if it don't trip a RCD it must be OK.

Where Baby Belling cookers still fail the insulation resistance test after being heated up, they should be given a more thorough visual examination. If all appears normal, given a technical visual pass as noted on the record, then this still complies with Regulation 4 of the Electricity at Work Regulations 1989 in that a safe system of maintenance has been established

I PA test in a place that has a small Belling cooker in their kitchen and that always fails the test unless I remember to warm all the elements up for a few minutes before testing,

Dave.

Unfortunately you are looking for something that is not there using the basic equipment you have, as I pointed out heating elements new or even cold can register very low IR readings but can improve as the unit gets used or even hot and that is because oxidisation forms on the conductor or heating element and improves the dielectric strength sometimes quite slowly over a period of time.

The biggest problem here is you are using an instrument (basic IR tester) that is only designed for cable where the conductors are in close proximity to each other, within 1mm as beyond that the readings become erroneous, in actual fact a simple IR tester is not really suitable for MI or even armoured cable and in the areas I have been it would not be used.

The minimum dielectric strength of a 3kW heating element is 0.0001A @ 1000/2000v versus temperature and this can only be measured with a hi-pot tester and this tester was what we used for MI and armoured cable.

With regards to MI cable and your domestic tester where these methods are now somewhat lost the correct procedure was to install the new cable into service even though the standard IR may have been often lower than 1M and around a month later recheck the IR, but in my case where we used a hi-pot tester and the dielectric strength was within limits there was no need to do any further testing.

The issue of damp can border on the 'wives tale' boundary a bit, the improvement is mainly because of oxidisation on the element or conductor, once the oxidisation has hardened and the element or cable is used within the bounds of 'moderate temperature' (the average range of say an oven) no further deterioration of the element or conductor takes place.

Men in suits rule. LOL

regards

-------------------------
"Take nothing but a picture,
leave nothing but footprints!"
-------------------------
"Oh! The drama of it all."
-------------------------
"You can throw all the philosophy you like at the problem, but at the end of the day it's just basic electrical theory!"
-------------------------