- Joined
- 11 Jan 2004
- Messages
- 44,059
- Reaction score
- 2,912
- Country
No. Absolutely not.
Just sign an affidavit that John was, is and always will be, right and then admit defeat.
PME Trip Times etc advice
- Thread starter T0nyguitar
- Start date
No. Absolutely not.
Just sign an affidavit that John was, is and always will be, right and then admit defeat.
Sponsored Links
Except, of couse, when he argues with BAS
He could be in deep trouble if he did sign such an affidavit since it would clearly be untruthful - as witness the fact that on this occasion (as many others) John was seen, by him and witnesses, to write some things which were very wrong
Kind Regards, John
The reason I thought they were tested in isolation is to remove any capacitive components from the test. A tester basically operates by creating a leakage current to trip the RCD and then measures the time it takes for the tester to see a pre-determined voltage, I imagine this will be somewhere around around 0v.
Anything which holds charge i.e. capacitors, possibly even cables will increase the time taken for the voltage to decrease to 0v.
So even though the RCD may trip within limits the tester may see it taking longer.
Anything which holds charge i.e. capacitors, possibly even cables will increase the time taken for the voltage to decrease to 0v.
So even though the RCD may trip within limits the tester may see it taking longer.
Sponsored Links
Interesting point. I've often wondered exactly how the meters determine the point of tripping to a precision of a millisecond or two - if tripping happens to occur at/around the zero-crossing point of the waveform, nothing would actually 'happen', voltage-wise' at that point in time, and one would have to wait at least a few milliseconds for the meter to 'realise' that the waveform was not going to move from zero in the way it would if the RCD had not operated. By the same token, the magnitude of any effect of capacitance such as you describe would presumably depend upon the point of the waveform at which tripping occurred - if near the zero-crossing point, it would presumably have little/no effect.
[I'm not really used to having to think about the behaviour of capacitors upon removal of an AC applied voltage - at which point I presume they essentially behave as if it had been DC, of a voltage equal to the instantaneous voltage of the AC at the moment it disappeared].
Kind Regards, John
There are too many variables involved to fully test an RCD "in the field" with a single test instrument and provide accurate times to disconnect. The present tests are ( apparently ) a basic Go-NoGo test
An RCD ( of the type fitted in domestic installs ) can only be aware of an earth volt when there is voltage to drive current through the fault to ground. That happens 100 times a second as the supply voltage is AC,
The time before disconnection is mostly determined by two variables. The impedance of the fault circuit and the time it takes for the mechanical action of opening the contacts. A third factor is load impedance but this is probably insignificant in a Go-NoGo test.
The fault current will vary as the AC waveform changes. At the zero crossing point there will be no leakage current and at the peaks of the voltage the leakage current will be at its highest.
The time for the mechanical trip action to complete will depend on how much "kick" energy is produced by the current un-balance in the sensing coil to kick the mechanical mechanism into tripping.
Assuming a trip current ot 30 mA
a fault impedance of 1 ohm will create a fault current when the AC waveform reaches 30 mV, almost immediately after the zero crossing point. The kick will last the whole half cycle and will probably lead to non -stoppable tripping in one half cycle.
a fault impedance of 1000 ohms will create a fault current when the AC waveform reaches 30 V, The kick from this will last less than a half cycle and two or more kicks may be needed to push the trip mechanism in a non stoppable trip
a fault impedance of 10,000 ohms will create a fault current when the AC waveform reaches 300 V approximately 5 mSec after the zero point. The kick from this will last less than a mSecond several kicks may be needed to push the trip mechanism in a non stoppable trip.
The meter no doubt works on a RMS current value from a sine wave current which is the most likely wave form from a purely resistive fault path to ground.
It gets complicated when copper oxide on a live conductor touches a different metal that has a path to ground, That could create a diode effect and produce a fault current that is not sine wave.
In my opinion provided the RCD begins a non stoppable tripping action during the first half cycle after the fault current begins then the device has done the best it can. But at low fault currents there may not be enough energy in one half cycle to fully trip the mechanism in which case a couple of half cycles may be needed before the trip has moved enough to be non-stoppable.
I think using milliseconds as the test result can be misleading and the number of complete half cycles before the supply is disconnected would be a more accurate indication of the operation of the RCD.
An RCD ( of the type fitted in domestic installs ) can only be aware of an earth volt when there is voltage to drive current through the fault to ground. That happens 100 times a second as the supply voltage is AC,
The time before disconnection is mostly determined by two variables. The impedance of the fault circuit and the time it takes for the mechanical action of opening the contacts. A third factor is load impedance but this is probably insignificant in a Go-NoGo test.
The fault current will vary as the AC waveform changes. At the zero crossing point there will be no leakage current and at the peaks of the voltage the leakage current will be at its highest.
The time for the mechanical trip action to complete will depend on how much "kick" energy is produced by the current un-balance in the sensing coil to kick the mechanical mechanism into tripping.
Assuming a trip current ot 30 mA
a fault impedance of 1 ohm will create a fault current when the AC waveform reaches 30 mV, almost immediately after the zero crossing point. The kick will last the whole half cycle and will probably lead to non -stoppable tripping in one half cycle.
a fault impedance of 1000 ohms will create a fault current when the AC waveform reaches 30 V, The kick from this will last less than a half cycle and two or more kicks may be needed to push the trip mechanism in a non stoppable trip
a fault impedance of 10,000 ohms will create a fault current when the AC waveform reaches 300 V approximately 5 mSec after the zero point. The kick from this will last less than a mSecond several kicks may be needed to push the trip mechanism in a non stoppable trip.
The meter no doubt works on a RMS current value from a sine wave current which is the most likely wave form from a purely resistive fault path to ground.
It gets complicated when copper oxide on a live conductor touches a different metal that has a path to ground, That could create a diode effect and produce a fault current that is not sine wave.
In my opinion provided the RCD begins a non stoppable tripping action during the first half cycle after the fault current begins then the device has done the best it can. But at low fault currents there may not be enough energy in one half cycle to fully trip the mechanism in which case a couple of half cycles may be needed before the trip has moved enough to be non-stoppable.
I think using milliseconds as the test result can be misleading and the number of complete half cycles before the supply is disconnected would be a more accurate indication of the operation of the RCD.
Thx bernardgreen very interesting.
Very interesting - but what about the question I effectively posed last night? Once disconnection has occurred, how quickly is the meter going to 'realise' that this has happened - particularly in the worst-case situation of disconnection happening at the time of zero crossing of the waveform?
Kind Regards, John
Not an expert but -
the RCD does not become aware of an earth volt.
There are no earth volts when operating the test button.
It 'merely' detects residual current on the L and N conductors.
Unless it is known how the mechanism works it cannot be deduced how various fault impedances affect it.
For example, once a residual current of 30mA is detected, for how ever short a time, do the electronics and the mechanism then 'go on' to switch off the device regardless of the fault continuing?
I took that to be a (semi-'phonetic') typo - such that Bernard actually meant to type "earth fault", not "earth volt".
Yes, that also occurred to me. Like you, I don't know, but it seems quite possible that the electronics are simply 'triggered' (into a 'fault state') by the first appearance of a 30mA imbalance, and remain in that triggered state (until disconnection occurs) regardless of the perisistence (or otherwise) of the fault.
Kind Regards, John
Oh, could be. I really didn't think of that.
Ad John has said, a typo by me....
That word "residual" is mis-leading. It is the difference when the currents in Live and Neutral are added algerbraically.
Which means a test meter might need to be different for different makes of RCD unless all makes have exactly the same mechanism, which they don't.
In an "electronic" or "active" RCD the difference ( aka residual ) is measured and if it exceeds 30 mA for a "debounce" time then the electronics use power to trip the mechanism. These can be made very predictable as to operating current and duration required to trip the mechanism.
In a passive RCD the energy required to move the latch to and beyond the trip point comes from the output of the sensing coil which is affected by many things and thus passive RCDs are not predictable in operation other than in general terms. It will operate within X miSeconds if more than Y mA of difference occurs.
He subsequently usd the phrase "fault to ground" in the same sentence, which I took to support my belief that "earth volt" was a typo. I'm an expert in this area, since 'semi-phonetic typos' are a speciality of mine when my brain and typing fingers are in autopilot mode
Kind Regards, John
Is that true of all passive RCDs? I had imagined that at least some, just like active ones, had electronics which triggered a power-operated trip mechanism. If the electronics and trip power came from the supply side of the RCD, that would probably be OK, since in the hypothetically complicating situation of the supply failing (hence removing power for electronics and tripping), the need for RCD operation would usually be fairly moot (and, in any event, could be supplemented by a secondary 'passive' back-up mechanism if one wished).
Kind Regards, John
It's not misleading. That's what residual means - 'left over', the residue.
It is the difference between the currents in Line and Neutral i.e. left over. (adding would not be useful)
DIYnot Local
Staff member
If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.
Select the supplier or trade you require, enter your location to begin your search.
Are you a trade or supplier? You can create your listing free at DIYnot Local
Sponsored Links
