RCD TRIPPING after hoover gets plugged in

[EFLImpudence]

I see what you are saying but the 11.2ms is the difference in the tripping times of the two RCDs when tested with, presumably, the same meter and parameters.

So, according to Harry, does that mean the second RCD should NOT have tripped when subjected to the fault in question.

 

[martygturner]

I see what you are saying but the 11.2ms is the difference in the tripping times of the two RCDs when tested with, presumably, the same meter and parameters.

So, according to Harry, does that mean the second RCD should NOT have tripped when subjected to the fault in question.

No, if the second is connected somehow... joined on same N bar then, or there is a shared neutral out on the circuits, lighting for example.

As for the RCD operating time and the difference they will always vary in their response time, no two identical units will work at the same speed, brand new ones may be close in operating speed but not identical, to get identical all the variables would need to be the same, take operating temperature how would you control that in a consumer unit , how would you measure it, if temperature increased what happens to resistance and the speed of the wave that triggers the rcd components, faster or slower ? Is the connection between the RCD and its circuit as good as its neighbour, does 2.5n/m give exactly the same connection every time, is the copper a bit oxidised. Are the components exactly the same, any variation in the iron core, different destiny ?

Then there is the condition of the rcd over time, does one rcd have a harder life, is it on the downstairs sockets, whats the effect on the rcd components being hammered by the wave of electrons, that's why they are tested to see if they are still within their design spec as over time the rcd components will degrade and affect the response time. So hence we have a maximum permissible disconnect time ... lower than 40mS AND greater than 30mA.

I think the confusion has arisen due to the difference between the fault condition and the trip time, a standard rcd will trip when over 30mA of voltage imbalance is seen, however it can take up to 40mS to actually complete the trip and achieve isolation, given the very small circuit inside the rcd a wave travelling through the rcd @ 3.2 meters per second will be well on its way to the second rcd if they are linked by the time the first has achieved isolation.

But of course that does depend on a number of other factors...eg say there is 28mA leakage on the co-joinied rcd's, rcd 1 at or close to leakage limit say 28mA, if rcd 2 is down at say 26mA then 3mA extra would trip rcd 1 but not 2. Whilst they may be linked the first toridal would absorb a few mA and shield 2 from all the leakage. That's why we have high integrity boards, so that leakage on one set of circuits does not propagate to the 2nd.

 

[EFLImpudence]

No, if the second is connected somehow... joined on same N bar then, or there is a shared neutral out on the circuits, lighting for example.

As for the RCD operating time and the difference they will always vary in their response time, no two identical units will work at the same speed, brand new ones may be close in operating speed but not identical, to get identical all the variables would need to be the same, take operating temperature how would you control that in a consumer unit , how would you measure it, if temperature increased what happens to resistance and the speed of the wave that triggers the rcd components, faster or slower ? Is the connection between the RCD and its circuit as good as its neighbour, does 2.5n/m give exactly the same connection every time, is the copper a bit oxidised. Are the components exactly the same, any variation in the iron core, different destiny ?

Then there is the condition of the rcd over time, does one rcd have a harder life, is it on the downstairs sockets, whats the effect on the rcd components being hammered by the wave of electrons, that's why they are tested to see if they are still within their design spec as over time the rcd components will degrade and affect the response time. So hence we have a maximum permissible disconnect time ... lower than 40mS AND greater than 30mA.

Just to point out that RCDs must trip somewhere between ½ and 1 times their rating; i.e. between 15 and 30mA - usually they are around 26mA.

I think the confusion has arisen due to the difference between the fault condition and the trip time, a standard rcd will trip when over 30mA of voltage imbalance is seen, however it can take up to 40mS to actually complete the trip and achieve isolation,

Given all parameters being equal: when you say 'complete the trip', does that mean the residual current must be present for the whole of that 40ms (or whatever time it takes) or is there a recognition time which begins the process of disconnection and then a time the device takes to actually open?

 

[martygturner]

Just to point out that RCDs must trip somewhere between ½ and 1 times their rating; i.e. between 15 and 30mA - usually they are around 26mA.


Given all parameters being equal: when you say 'complete the trip', does that mean the residual current must be present for the whole of that 40ms (or whatever time it takes) or is there a recognition time which begins the process of disconnection and then a time the device takes to actually open?

Yep, but its DC which is injected which will be equal to 1.4 times AC as AC has a full wave Peak,0,-Peak... so dependant on your kit at 1 times say 30Ma you will be getting 42 mA at the rcd and the limiting factor is the solenoid/relay.. how long will that take to fire... less than 40mS...once it is triggered ain't no stopping it.