Online assistance for electrical trade people Login  |  Register  |   Forgot Password
Assistance for electrical trade people
 

 

 

 


Click here to send Ron a pdf document for publication on this Topic

Documents must be less than 200k in pdf format

Posted By Topic: RCD for sports flood lighting circuits?

Atpfellow
Feb 26 2018 16:48

We are doing a sports field flood lighting project. The lights will be mounted on the top of 16.5m galvanized poles. The ballasts will be mounted inside the poles. The Engineer has specified one RCD in each of the final circuits. I just wonder if RCD is required for this type of work? - I am sure this is not a residential installation.
   

mf51to1
Feb 26 2018 19:56

Not mandatory based on 2.6.3. Where is the sports field located, out of curiosity?
Not a bad thing that the engineer has included additional RCD protection to go beyond the minimum requirements.
   

Atpfellow
Feb 27 2018 12:42

It is in central Auckland. The site is open and the lighting poles are accessible to the public.
   

mowgli
Feb 28 2018 20:27

How many ballasts per RCD?
   

alkyleda
Mar 05 2018 20:39

No surprises there. More people running scared of electrocuting other people. Just earth it like we used to do in the old days ;-)
   

mf51to1
Mar 05 2018 20:47

[quote]No surprises there. More people running scared of electrocuting other people. Just earth it like we used to do in the old days ;-)[/quote]

What like conductive kitchen sinks that no longer need to be? :)
   

alkyleda
Mar 05 2018 20:57

Unless you want to be on call every time the sports ground is used, so you can run around resetting the RCDS, I wouldn't bother, (hence the previous comment). Earthing stuff is much better way of doing it than RCD's.
We all have to live in the real world here and I see this as impractical. Maybe I'm wrong, but my 25 years of being a sparky has taught me a whole lot more than a bunch of people in an office making idealistic rules that we sheep should follow.
   

AlecK
Mar 06 2018 08:34

Nothing impracticable about using RCDs on a job like this. And if the work is done properly, they won't be tripping very often if at all - should be only when tested. No need for a "pissing contest" about years in the game - it's not relevant. What's relevant is Ohm's law, and anything that passes an IR test won't trip an RCD.
Sure, the IR test uses d.c., and when you put a.c on the circuit there will be some capacitive leakage. The longer the circuit, the more leakage. But properly installed should still never be an issue.

In my "real world"; if you get "nuisance" tripping of RCDs, the most likely explanation is that you've done something wrong.
   

mowgli
Mar 06 2018 08:43

Atpfellow, an RCD in this circuit is not mandatory in existing or proposed rules.

Ask the engineer what is the purpose of the RCD. If it's to protect infrastructure then use a 300mA type S and you're done.

If the engineer is expecting 30mA type A RCDs to protect against electric shock then ask for the predicted earth leakage current from the ballasts. Multiple ballasts on a 30mA RCD may result in nuisance tripping. 2.6.2.1 requires that we avoid this.

If the predicted leakage exceeds 15mA then the engineer's options would be to increase the number of RCDs (fewer ballasts per RCD), accept higher radio interference (simpler ballasts), or as alkyleda suggests, rely on another compliant method for automatic disconnection in the event of a fault.
   

gregmcc
Mar 06 2018 14:36

this is starting to sound like a part 1 design, time to ask the designer for his certified design so they can bare the responsibility for future problems.
   

mowgli
Mar 06 2018 16:06

If non-mandatory RCDs were added in a way that broke part 2 compliance then I expect those same RCDs would also fail part 1.
   

mf51to1
Mar 06 2018 20:25

[quote]ask for the predicted earth leakage current from the ballasts. Multiple ballasts on a 30mA RCD may result in nuisance tripping.[/quote]

Cant say Ive come across this issue before. Testing a new ballast has very high IR.
Cant see ballasts getting down to very low IR (RCD tripping) levels unless they had faults.
   

mowgli
Mar 07 2018 18:55

Newer high frequency ballasts may be capacitively coupled to ground to reduce interference. It may be in the order of 1-2mA so depends on how many per circuit. Noting 30mA RCD will usually trip around 20mA.
   

pluto
Mar 07 2018 21:47

Another factor that will be news to many of you is the d.c. current flowing in the a.c. input to an electronic ballast, (typically a switch mode power supply) if it is greater than 6 mA constant or pulsating d.c. current prevents the use of a Type A RCD and then you have to use a Type B or F RCD which is only capable of 10 mA max.

The reason for this limitation is the d.c. current interfers with the RCD sensing coil magnetic core, and if great enought, will prevents the RCD from tripping under earth fault conditions.

Some guidance is going to be published by Worksafe NZ shortly until this information gets into the Amendment No. 1 of as/nzs 3000:2018.
   

pluto
Mar 11 2018 09:22

mowgli Mar 07 2018 18:55

Your comment (part only)
Noting 30mA RCD will usually trip around 20mA.

My comment
The IEC 61008 or IEC 61009 standards (the official standards for RCDs in NZ for approval to sell in NZ) require that the RCD does NOT trip until the residual current reaches 22 mA but must trip at or before 30 mA.
   

mowgli
Mar 11 2018 11:14

Thanks Pluto I wasn't aware of that. ESR24 and 3190 certainly support the assertion that the limits are not less than 50% and not more than 100%.

Quick field test at home reveals my 30mA RCCB trips at 20mA with all but that subcircuit isolated.

So it's not unsafe by ESR but couldn't be sold. Do you think this is due to deterioration in service or is this likely to be a manufacturing defect.
   

pluto
Mar 11 2018 12:46

Your RCD tester is the most likely source of the error.
   

mowgli
Mar 11 2018 18:10

Fairly new Megger MFT1835 gives uncertainty +-1% +-1d for RCD trip test.

Theoretically the result 20.0ms is in the range 19.7ms to 20.3ms.

I don't have a set up to verify the calibration of trip timing, however, the other functions that I can verify remain within their calibrated tolerances. That's no guarantee that my tester is reading true, but.... if it is where else would the error come from?

Do RCD trip times change over time? What is the failure mode of an RCD?
   

pluto
Mar 11 2018 20:46

Timing is unlikely to be problem with the RCD tester, digital timer are usaully stable over a wide operating voltage range.

It is the test current that will be the most variable, it is very short term rated resistor in the RCD tester, and as it gets hotter ther test current will drop due to the test resistor heating up.

It normally only conducts for 1.5 to 2 cycles max for each test, but you need to let the test resistor cool down between each use to get the correct test resistance value and the right residual test current.