Intentional earth leakage

[LastManOnline]

Just thinking out loud. IEL (intentional earth leakage) is already having a significant impact on domestic socket circuit design. It's impact is going to continue to increase as virtually every apiance incorporates electronics. RCBO populated DB, s will become increasing necessary to cope with this change. However is it conceivable that at some point a "fourth conductor" may become standard in domestic circuits, (L, N, E + E2), specifically for IEL? This would then remove the impact of IEL from circuit design

 

[timhoward]

Interesting idea. But it seems unlikely as appliances and accessories would have to be constructed that way first, and then a whole new range of plugs and sockets / FCU would be needed.
There would have to be a way of using appliances equipped for this in normal installations too.
So it would be a massive change to embrace at all levels.

 

[snowhead]

It's far easier to fix an appliance than it is to fix an installation.

A fourth conductor would be as bad as requiring metal C.U's to contain a fire caused by bad C.U connections design and poor installation.

I'd say it will go the opposite way in that intentional leakage will have a limit put on it such that an average number of appliances with leakage won't cause an issue in an average older installation.

However I suspect that the potential issue is already being looked into by those who plan and make decisions for the future.

 

[pc1966]

The current (old?) PAT testing values where:

• Heating appliances should have a current that should be less than 0.75mA per Kilowatt up to a max of 5mA.
• Portable/handheld appliances have to have earth leakage current of less than 0.75mA to be considered safe.
• Other Class 1 appliances such as IT, movable and stationary equipment have a limit of 3.5mA
• Class 2 appliances – 0.25milliamps
• For Class III appliances the limit is 0.5milliamps

It used to be a lot of computer equipment was quite leaky, but now less of a problem it seems (guess just better EMC filter design, etc) though it is still an issue for any sort of data centre or IT lab with dozens of machines off the same supply point.

I suspect the practical solutions will be more regulation of appliance leakage limits and a move to all-RCBO boards so you have less cumulative leakage problems. Which most of us have been advising for years but now cost is down to a point where it no longer makes eyes water!

Drat, forgot about AFDD...eyes watering again folks!

 

[loz2754]

The 5th Edition of the CoP for ISITEE page 74 simply states "the values should not exceed 5mA for AC-powered equipment".
It no longer differentiates between the various types of equipment.

 

[LastManOnline]

So it would be a massive change to embrace at all levels.

Well the jury, s verdict is pretty clear on this one and I am very much inclined to agree. Would like to take the opportunity though to once again thank the appliance manufacturing Industry for the wonderful manner in which they have introduced "intentional leakage" in to our lives. Little or no planning. Sparks largely left in the dark.Confusion still reigning over which rcd to use. There is very much a feel of " Create a problem and let someone else tidy up the mess" about the whole situation

 

[freddo]

(guess just better EMC filter design, etc)

A lot of manufacturers seem to totally omit filter components from the final product, chokes being replaced with wire links, capacitors not fitted and so on. No doubt the components are fitted when samples are sent for EMC approval...

 

[pc1966]

A lot of manufacturers seem to totally omit filter components from the final product, chokes being replaced with wire links, capacitors not fitted and so on. No doubt the components are fitted when samples are sent for EMC approval...

Yes, anything to save money even if they cause radio problems. Not that they ever seem to get prosecuted for failure...

 

[Lucien Nunes]

I think this is a storm in a teacup. Operational leakage has always been around; sheathed heating elements have always had a tendency to leak resistively, SMPSU line filters have always leaked capacitively, large white goods have for many decades had a delta cap across the supply etc. Any increase is as much due to uptake of more appliances by users, than a deliberate increase in leakage per item. In cases where leakage has increased, again this is as much due to the need to comply with more stringent EMC requirements as anything decided by the manufacturers.

Increases will have come from:

• A large number of small SMPSUs now in use, which might previously have been wire wound transformer-fed and hence low leakage. But the change from linear to SMPSU usage is swamped by the dramatic increase in the number of devices in use.
• More stringent EMC requirements. I am not sure how much impact this has on the total because although the leakage per kW might have increased, the power requirements of most devices have decreased. For example, work that routinely needed a desktop PC and CRT monitor in 2000 can now be done with a laptop, so the two PSUs of say 300W and 100W each have given way to one charger of 90W that probably doesn't leak over 0.5mA.
• New types of high-leakage appliance that have only recently come into widespread use, such as induction hobs.

The idea of a 4th conductor is a red herring. There is already a conductor present to take away current at or near earth potential, which is the neutral. Any leakage that could be directed into the 4th conductor could already be directed into the neutral. Earth leakage arises through coupling, (either unavoidable or deliberately introduced for RF filtering reasons) from live functional parts to parts connected to the protective earth or to the outside world. Adding the 4th conductor would not remove the need for the existing filter leaking into the protective earth, but instead would increase the leakage as it would also need filter components leaking into it.

In reality, there is no sensible argument against the universal adoption of RCBOs, which in all normal domestic and many commercial situations will completely overcome any issues with increased aggregate leakage. The overall RCD or split board was a stop-gap adopted while the unit price of RCDs decreased to a level supporting one per circuit.

 

[Rockingit]

We deal with this in the events industry almost daily. LED screens (the big walls you see) are notoriously leaky and typically need to be on 100mA B types, sometimes even higher. Often we'll use RCM devices and some human judgement for items out of reach.

 

[Lucien Nunes]

Large LED displays fall into the category of 'new types of high-leakage appliance' and are pathologically leaky and inrushy on account of having a large number of medium-sized SMPSUs all separately filtered. But people do not randomly plug large LED walls into general-purpose supplies, they are a special case for which the operators are aware of the special requirements. More insidious are the smaller LED installations e.g. in retail and corporate premises, where the leakage might require a high-integrity CPC but no-one has thought that far ahead, and instead merely provided eight double 13A sockets on a 4mm radial protected by a C32 RCBO.

 

[marcuswareham]

Are switch mode power supplies not causing RCDs to trip due to harmonic content, in particular 3rd harmonics and how they affect the sine wave causing apparent higher neutral current and therefor causing an imbalance in rcd coil, and in fact has nothing to do with leakage to earth

Thats why I have had alot of SMPU which are in double insulated items with no CPC in the equipments flex yet accumulatively tripping RCDs, this must be down to harmonic distortion because of the way the SMPU draw current to charge capacitor then suddenly stop drawing current when capacitor starts discharging, etc etc, and not via leakage as there is no return path for any leakage (no cpc and insulated enclosure)

So would an 4th "leakage" conductor do anything really?

 

[Lucien Nunes]

Zero-sequence harmonics do sum in the neutral, but they do not cause an imbalance at the RCD because the vector sum of the line currents is equal and opposite. Since the RCD sees the sum of line and neutral currents, the residual is zero. Kirchhoff's first law applies here.

Although there is no CPC for a Class II SMPSU to leak into, they will often produce a small amount of touch leakage from the SELV output. If the output is deliberately or casually earthed, e.g. where multiple Class II home cinema components are connected via HDMI cables to at least one Class I device such as a projector, all have a route to earth for their leakage and it begins to add up.

Larger SMPSUs increasingly have active power factor correction, which uses a DC-DC converter between the rectifier and the reservoir capacitors to shape the input current into a sinusoid in-phase with the supply voltage. They generate much lower distortion with pf up to 99% as standard.