I have built a couple of mains filter boxes. Each contains 4 high performance 3 – 6 A commercial EMI filters connecting to 4 IEC sockets with the ground wire going through some ferrite cores. They have worked well. One is used for my stereo sound system and the other is used for my home theatre. The 2 mains filter boxes never get turned on at the same time.
One minor issue is that leakage current of the combined filters (from the Y capacitors) is relatively high so the house main circuit breaker can be tripped more often when the supply power line is noisy and other household appliances are working. This has happened about twice for the past 6 months.
My understanding is that the wall switch only switches off Hot/Fire, but not Neutral. So, when the wall switch is turned off, the Y capacitor from the Neutral line is still connected and working.
The question is, when switch off, i.e. Hot/Fire is disconnected, the leakage current from the Neutral line, which is not switched off, should have minimal amount of leakage current? Or not?
One minor issue is that leakage current of the combined filters (from the Y capacitors) is relatively high so the house main circuit breaker can be tripped more often when the supply power line is noisy and other household appliances are working. This has happened about twice for the past 6 months.
My understanding is that the wall switch only switches off Hot/Fire, but not Neutral. So, when the wall switch is turned off, the Y capacitor from the Neutral line is still connected and working.
The question is, when switch off, i.e. Hot/Fire is disconnected, the leakage current from the Neutral line, which is not switched off, should have minimal amount of leakage current? Or not?
Yeah since the neutral is normally connected to earth at the substation somewhere not too far away, the potential on 'neutral' is a couple of orders of magnitude below that on 'live'. So the leakage currents will be relatively insignificant.
There will be a capacitive leakage at all times that will be insignificant to an RCD trip set at say 30mA. If however, your electric cooker draws a lot of current, the power return to the Neutral line may be enough to trip it, if there is a poor Neutral connection in your household wiring. Any more than 4 Volts between Neutral and Earth is a fault.
I said in a post a couple of days ago that the leakage current passing through a UK RCCB should be less than 25% of it's rating.
That gives enough margin for interference to not trip the detector.
If your household circuits have a total leakage exceeding 25% of rating, then find a way to split your circuit loads and put in an RCCB for each reduced load circuit.
Two circuits each with a 30mA RCCB gives you a potential 15mA of leakage and still stay below the 25% of rating.
That gives enough margin for interference to not trip the detector.
If your household circuits have a total leakage exceeding 25% of rating, then find a way to split your circuit loads and put in an RCCB for each reduced load circuit.
Two circuits each with a 30mA RCCB gives you a potential 15mA of leakage and still stay below the 25% of rating.
For a standard residual current cb, the operate time is quite 'long' (100's of ms) when the residual current is near to the setting, so a tripping RCD is experiencing quite a level of leakage current (either high peak level, or sustained low level).
Given that the RCD trips very infrequently, there would have to be doubt as to whether it is the addition of leakage current from your audio gear that is an influence, and also doubt about whether the trip is from 'interference' rather than a more sustained leak from say an insulation breakdown somewhere.
Nowadays, residential switchboards are more likely to use individual RCD cbs for each house circuit, rather than one RCD for the majority of house circuits. That helps to initially to localise a failing device to just one circuit, and minimises disruption to other circuits.
You may not have the tools to further define what the source of the leakage is, and whether your audio emi filter is complicit, other than to try and define trip events by addition/subtraction of connected devices.
Further definition is likely to include a voltage analyser on the neutral at various locations, and a current analyser on the neutral in to your mains filter boxes. But these may not define insulation breakdown events within the house, until a breakdown occurs.
Given that the RCD trips very infrequently, there would have to be doubt as to whether it is the addition of leakage current from your audio gear that is an influence, and also doubt about whether the trip is from 'interference' rather than a more sustained leak from say an insulation breakdown somewhere.
Nowadays, residential switchboards are more likely to use individual RCD cbs for each house circuit, rather than one RCD for the majority of house circuits. That helps to initially to localise a failing device to just one circuit, and minimises disruption to other circuits.
You may not have the tools to further define what the source of the leakage is, and whether your audio emi filter is complicit, other than to try and define trip events by addition/subtraction of connected devices.
Further definition is likely to include a voltage analyser on the neutral at various locations, and a current analyser on the neutral in to your mains filter boxes. But these may not define insulation breakdown events within the house, until a breakdown occurs.
Thank you very much for your comments.
Let us assume that the electrical wiring in the house is done properly. Since the voltage on Neutral is substantially lower than Fire/Hot, the leakage current from Neutral should be much lower than that is from Fire/Hot. So I don't need to worry about what is written on the data sheets of the EMI filters that have a combined leakage current of 5mA when these EMI filters are switched off (disconnected from the Fire/Hot while connected to Neutral). When one of the two Mains Filter box is turned on, there will be a maximum leakage current of 2.5mA, which is within the acceptable range. Can I conclude it that way?
Let us assume that the electrical wiring in the house is done properly. Since the voltage on Neutral is substantially lower than Fire/Hot, the leakage current from Neutral should be much lower than that is from Fire/Hot. So I don't need to worry about what is written on the data sheets of the EMI filters that have a combined leakage current of 5mA when these EMI filters are switched off (disconnected from the Fire/Hot while connected to Neutral). When one of the two Mains Filter box is turned on, there will be a maximum leakage current of 2.5mA, which is within the acceptable range. Can I conclude it that way?
Not sure I understand this sentence?
Do you have a link to emi filter datasheet?
here it is the datasheet:
FN 2090
I built 2 mains filter boxes with each box containing 4 such EMI filters (mostly 6A rating). So one box would have leakage current of 4 x 0.61mA = (approximately) 2.5mA. Two boxes would have leakage current of 2 x 2.5mA = 5mA.
But only one box would be switch on at one time. So the leakage current would be 2.5mA. The other one would be switched off.
The question is that for the mains filter box that is switched off, since the Neutral is still connected, how much leakage current it would produce to cause problems.
FN 2090
I built 2 mains filter boxes with each box containing 4 such EMI filters (mostly 6A rating). So one box would have leakage current of 4 x 0.61mA = (approximately) 2.5mA. Two boxes would have leakage current of 2 x 2.5mA = 5mA.
But only one box would be switch on at one time. So the leakage current would be 2.5mA. The other one would be switched off.
The question is that for the mains filter box that is switched off, since the Neutral is still connected, how much leakage current it would produce to cause problems.
Ta. The datasheet rated leakage current is for 250VAC 50Hz, and is effectively for the capacitive current from L to PE (which is supported by their comment that leakage would increase if neutral was disconnected - ie floating and allowing current to flow).
Based on simple capacitor current, a filter with just neutral and PE connected would cause a neutral current proportional to the N-PE voltage, and normalised to 0.61mA if N was 250V with respect to PE. So a 4V rms level of N-PE would cause 0.61 x 4/250 = 0.01mA.
For a neutral line to have 4Vrms wrt PE at a particular GPO would be due to neutral current flow to other devices being powered from that particular circuit. It could be assumed that their is no voltage between the main neutral bar and the main earth bar in your switchboard.
Your wiring should not allow say more than 5% (I can't recall the limit) voltage drop to a device from the switchboard. That drop is split between active and neutral. Eg. 2.5% of 240V is 6Vrms drop.
Based on simple capacitor current, a filter with just neutral and PE connected would cause a neutral current proportional to the N-PE voltage, and normalised to 0.61mA if N was 250V with respect to PE. So a 4V rms level of N-PE would cause 0.61 x 4/250 = 0.01mA.
For a neutral line to have 4Vrms wrt PE at a particular GPO would be due to neutral current flow to other devices being powered from that particular circuit. It could be assumed that their is no voltage between the main neutral bar and the main earth bar in your switchboard.
Your wiring should not allow say more than 5% (I can't recall the limit) voltage drop to a device from the switchboard. That drop is split between active and neutral. Eg. 2.5% of 240V is 6Vrms drop.
Thanks again. Your comments are very much appreciated. Now I have a peace of mind connecting both mains filter boxes to the wall switches.
- Status
- Not open for further replies.