rcd trips when lamp blows

[elrobbo82]

Could some one please explain why a rcd can trip when a tungsten filament bulb blows. As there is no connection to cpc i don't understand why this happens. I assume that a voltage is induced onto the cpc some where but don't know how. Just asking as I've came across this a few times and just don't know how to explain it. not long since finishing college and never had this explained to me so any input will be appreciated.

thanks

 

[RMS]

Have a read of this thread,

http://www.theiet.org/forums/forum/...5172&parentid=54992&FTVAR_FORUMVIEWTMP=Single

 

[elrobbo82]

thanks RMS that thread answers the question perfectly, just wondering if any one has a different take on the one given in the link provided?

 

[ericmark]

Much depends on how good the RCD is. The better types are able to work with spikes which will trip the more basic types. But it is the spike which causes it to trip. I find unless I switch off all MCB's when one RCD trips re-setting will often trip the other RCD.

These RCD's are if one believes what it says less likely to trip due to spikes. To save you following the link it say:-

Thunderstorms or other temporary disturbances won't unsettle the digital RCD thanks to an optimised tripping threshold and a type "G" short-time delay which is independent from the supply voltage. In addition, load situations - however close to their designed threshold - will not require additional thermal protection, thanks to the integrated overload protection mechanisms. And regardless of the supply voltage, it will always perform its duty in compliance with the IEC/EN 61008 standard. By taking advantage of the LEDs that indicate the tripping threshold is being approached - but not exceeded - you can undertake corrective maintenance to the system before the situation develops further and thereby avoid an inconvenient interruption to your power supply.

There is as far as I am aware a problem where the loop impedance is high in that a near short circuit can drop the voltage too low for the electronics to work so they will not trip. Hence why overload is included so if there is a near short circuit the overload part will insure it will trip.

This is also why RCD units built into sockets are the active rather than passive type. When installed by an electrician he will measure the loop impedance but the meters are so expensive that in the main the DIY guy will not have access to the meter and just crosses his figures and practices saying "It was like that when I moved in".

There are even auto resetting RCD's

Untitled

• ericmark
• 6 Dec 2008
• 34

but with a price tag of around £350 I think I can manage to manually re-set mine.

 

[RF Lighting]

I think it's a perfectly valid question.

How does a load connected to only line and neutral cause an RCD to trip which is only looking for leakage to earth?

There's nothing stupid about that question and I bet there's thousands of electricians up and down the land who do not know the answer. It doesn't automatically mean they are incompetent.

 

[westie101]

So I've been in the industry 38 years a colleague 45. If either of us are unsure of something we ask and discuss it, this could just be in our own office among the engineers there or by contacting other staff in other parts of the company. (these discussions take place quite frequently)

One important point we always stress to new staff is that if you are unsure ASK, this rule applies until the day we leave the industry.

So

It shows a pretty basic incompetence and I don't think he should be doing what he CLAIMS he is doing when he comes out with rubbish!

This sort of attitude from someone who claims to be a professional electrician stinks. If it discourages another less experienced electrician from asking and learning it is far, far more dangerous than someone who has the courage to ask, knows his limitations and will discuss problems.

Do you act like this with apprentices and other trainees?

Do you know everything?

Hmmm perhaps the next time I'm asked for advice by an experience electrician perhaps I should treat him as a fool.

 

[elrobbo82]

So I've been in the industry 38 years a colleague 45. If either of us are unsure of something we ask and discuss it, this could just be in our own office among the engineers there or by contacting other staff in other parts of the company. (these discussions take place quite frequently)

One important point we always stress to new staff is that if you are unsure ASK, this rule applies until the day we leave the industry.

So

It shows a pretty basic incompetence and I don't think he should be doing what he CLAIMS he is doing when he comes out with rubbish!

This sort of attitude from someone who claims to be a professional electrician stinks. If it discourages another less experienced electrician from asking and learning it is far, far more dangerous than someone who has the courage to ask, knows his limitations and will discuss problems.

Do you act like this with apprentices and other trainees?

Do you know everything?

Hmmm perhaps the next time I'm asked for advice by an experience electrician perhaps I should treat him as a fool.

thank you westie. i have made it clear that i don't want his/hers input on my thread as he/she is basically a bully. this person has not given a single bit of constructive input in any of my posts and is bringing this forum down to a childish level. i do however appreciate other peoples input as i have just finished college and am trying to become a better spark. if this question seems stupid i apologize. i understand why a mcb would trip in this situation due to overload at the point of the lamp blowing (high current). just was not clear to me why a rcd would trip as there would technically be no leakage to earth.

 

[Mrdjc]

I'm not an electrician but....

Doesn't a RCD detect an imbalance between the flow of current between the outgoing (L) and incoming (N) connection, using the CPC (PE) as reference "0"?

If a filament breaks, the outgoing (L) will be sending a flow of current out, which gets consumed by the filament but never returns to the RCD via the return (N)and so there is an imbalance between supply and return, it exceeds 30mA and goes pop?

Or am I completely wrong...

 

[elrobbo82]

i would not of thought this would cause an imbalance between line and neutral and trip an rcd. surely just a leak to earth would do this but that is why i asked the question. not as simple as some think then!

 

[Mrdjc]

Well, when a filament goes pop, what kind of behaviour would you see?

Sudden increase in temperature -> Increase in resistance -> decrease in current. ? But surely that would be the same as switching on a device..

Maybe it acts as a capacitor for a split second between the ends of the parting filament and the inert gas inside the bulb, and current increases significantly for a fraction of a second, exceeding the rating of the RCD?

 

[elrobbo82]

read the link provided by RMS as this seams like a logical answer to me.

 

[EFLImpudence]

Doesn't a RCD detect an imbalance between the flow of current between the outgoing (L) and incoming (N) connection, using the CPC (PE) as reference "0"?

Not exactly.

The RCD monitors the current in the L and N conductors within the RCD.

Normally these are 'equal' and 'balanced' so the result is no (or rather less than the nominal 30mA) residual (left over) current to operate the mechanism.

When some of the current in the installation finds a different path back to earth (for whatever reason from either conductor) then there is residual current (>30mA) detected in the RCD.
It does not measure the cpc or know or care where the current has gone - just that it's missing.

exceeding the rating of the RCD?

No, the rating of the RCD is not an overload rating, just a capacity.

 

[elrobbo82]

edit

 

[bernardgreen]

My theory as to why a lamp blowing can result in the RCD tripping is as follows

The failed lamp becomes a tungsten vapour lamp but without any control gear to limit the current so an extremely high current flows until the lamp explodes or the fuse blows / MCB operates on its magnetic trip.

The MCB is breaking a very high current that is flowing through the coil of its magnetic trip mechanism. Breaking that current will result in a very high voltage spike being generated from the energy of the collapsing magnetic flux. As happens in car ignition coils and other induction coils. It will arc across the opening contacts in the MCB and onto the live wires of the protected circuit.

This high voltage spike in itself will not trip the RCD which is current operated.

One end of the coil with the high voltage spike goes via the CU bus bar to the live conductor through the RCD sensor. The other end goes via the arc across the MCB contacts to the to the live conductor of the lighting circuit. That live conductor is capacitively coupled to the neutral conductor and the CPC ( earth conductor ) and therefore there are two capacitively coupled circuits from that end of the coil. Both these circuits return to the incoming neutral and hence via the network back the live bus bar and other end of the coil. Of these two circuit only one, the neutral, goes via the RCD sensor. So the all current generated by the high voltage spike passes through the RCD sensor's live conductor from the live bus bar but only a part of it passes through the RCD sensor's neutral conductor from the neutral bus bar hence there is an un-balance.

This un-balance is very short duration but due the the very high voltage generated at the coil and the high currents it creates the energy of the un-balance in the RCD sensor will be high and apparently high enough to operate the trip in some situations.

There may be some value in replacing MCBs with wire fuses on lighting circuits with lamps prone to becoming tungsten plasma discharge lamps to avoid producing the high voltage when the MCB operates on the high current from a lamp in plasma mode.

 

[sparkwright]

Anyone know why my recent thread titled 'Tossers' was deleted?