Schneider Electric Square D Rcbo Reliabiliy Problems

[Pib]

Nuisance trips are, well, a nuisance. In this case since they are on a water heater there's no real issue, it's not like it's your freezer.

Consider the choice, either have a trip that opens occasionally for no apparent reason or have no earth leakage protection when you accidentally connect yourself to the supply.

I know what I would rather have

If occassionally was ever few months, no problem; but so far with these 10ma RCBO "occassionally" is turning out to be every few days...and the one I took back had reached the point of every few minutes. That's too occassional for me.

Edited April 16, 2012 by Pib

[Pib]

electau... All good information, but as I stated in my original post, the failures are quite real. Taken out of the load center, they are VERY difficult to arm to the on position. The folks at HomePro could not get them to arm at all. These are real failures, not an application problem.

The three I've had so far would arm easily whether installed or uninstalled. You could flip the breaker switch on or off easily.

[Pib]

OK, just completed another test of my heater circuits and RCBOs by inserting an ammeter to check for any current flow/earth leakage when the heaters are turned off. As mentioned in earlier posts the RCBOs have only tripped when the heaters are off.

Got my digital multimeter, set it to the AC current function (it can measure down into the microamps), turned the RCBO off, removed the RCBO Live Load wire, hooked the ammeter between the RCBO L connector and L wire running to the heater, turned on the RCBO, measured the current flow. Repeated this setup and measurement on the N load side. Then I repeated this L and N current measurement process on the other heater circuit and RCBO. I got zero current flow on each leg of both RCBOs which indicates no current leakage to earth and also shows me there is no current flow with the heaters off…just as it should be.

I didn’t test the current flow with the heaters turned on since my inline ammeter only goes up to 10 amps and the 6000W heaters pull around 27 amps. Plus I can’t see how that would have told me anything useful especially since the amp draw would have varied slightly as the heating element got hotter and then me having to quickly disconnect/reconnect the meter on the N wire to take another measurement which would have varied due to heating element temperature change. Plus, all my RCBO trips have been when the heaters have been off…off for at least hours…and for one heater days.

OK, you may be thinking my ammeter was not working, not measuring down into the milliamps although it’s capable of measuring down in the microamps plus it’s an auto- ranging multimeter. I thought the same thing. So, I got my 7W bulb rig I mentioned in an earlier post which I used to test the RCBO (a double check of the Test button on the RCBO) and hooked my multimeter up, hooked into a wall socket, the bulb came on, and the multimeter measured 45.5ma…guess this cheapie little 7W bulb has some loose manufacturing tolerances as 45ma means it was really burning at around 10W on my 225V line.

I then went around the house again turning lights, A/Cs, heaters, TVs, etc., on and off to see if either of the RCBOs would trip…they didn’t. At this point in time I’m still thinking surges/spikes on the power feeding the house…surges/spikes in the millisecond ballpark which are too short for a human to notice and probably even electronic devices to notice (except those surges/spikes are probably slowly killing the devices like high blood pressure slowly kills a person). I know RCBOs are suppose to be built to not trip on such surges/spikes especially the really brief ones. Any ideas?

[electau]

OK, just completed another test of my heater circuits and RCBOs by inserting an ammeter to check for any current flow/earth leakage when the heaters are turned off. As mentioned in earlier posts the RCBOs have only tripped when the heaters are off.

Got my digital multimeter, set it to the AC current function (it can measure down into the microamps), turned the RCBO off, removed the RCBO Live Load wire, hooked the ammeter between the RCBO L connector and L wire running to the heater, turned on the RCBO, measured the current flow. Repeated this setup and measurement on the N load side. Then I repeated this L and N current measurement process on the other heater circuit and RCBO. I got zero current flow on each leg of both RCBOs which indicates no current leakage to earth and also shows me there is no current flow with the heaters off…just as it should be.

I didn’t test the current flow with the heaters turned on since my inline ammeter only goes up to 10 amps and the 6000W heaters pull around 27 amps. Plus I can’t see how that would have told me anything useful especially since the amp draw would have varied slightly as the heating element got hotter and then me having to quickly disconnect/reconnect the meter on the N wire to take another measurement which would have varied due to heating element temperature change. Plus, all my RCBO trips have been when the heaters have been off…off for at least hours…and for one heater days.

OK, you may be thinking my ammeter was not working, not measuring down into the milliamps although it’s capable of measuring down in the microamps plus it’s an auto- ranging multimeter. I thought the same thing. So, I got my 7W bulb rig I mentioned in an earlier post which I used to test the RCBO (a double check of the Test button on the RCBO) and hooked my multimeter up, hooked into a wall socket, the bulb came on, and the multimeter measured 45.5ma…guess this cheapie little 7W bulb has some loose manufacturing tolerances as 45ma means it was really burning at around 10W on my 225V line.

I then went around the house again turning lights, A/Cs, heaters, TVs, etc., on and off to see if either of the RCBOs would trip…they didn’t. At this point in time I’m still thinking surges/spikes on the power feeding the house…surges/spikes in the millisecond ballpark which are too short for a human to notice and probably even electronic devices to notice (except those surges/spikes are probably slowly killing the devices like high blood pressure slowly kills a person). I know RCBOs are suppose to be built to not trip on such surges/spikes especially the really brief ones. Any ideas?

First it would appear that your RCBOs are genuine. It would also appear that your tripping of 10mA units is being caused by leakage currents from the heater. On this case your water heater circuit 32A/10mA.

One would recommend replacing this RCBO with a 32A/30mA one.

10mA RCDs need only to be used to protect socket outlets in high risk areas, eg a socket outlet may be a 10mA SPRCD ( socket protected rcd) and the circuit inself protected by a 30mA RCD at the switchboard.

A 10mA RCD must trip at between 5 and 10mA and a 30 mA one between 15 and 30mA. the level of leakage current is higher with the 30mA unit and will operate within the required disconnection time without compromising safety.

To corrrectly measure leakage current one must measure the leakage current using a clamp current tester with a low eg. 0 -10mA range, clamped onto the earth wire and the reading taken with the equipment on and then with the equipment off.

Note: The integral switch on the water heater may not isolate the line and neutral terminals of the heater itself.

[Crossy]

The problem seems to be actually getting the 30mA RCBO's, I know Schneider make them but the big outlets don't seem to stock them

As far as measuring the earth leakage is concerned. The IEE recommend clamping the L and N lines with your low amp meter, it will read any imbalance in the same way as your RCD would detect it. You get the 'real' leakage which may not necessarily be flowing in an earth cable (I know it should flow in the earth cable but it may have decided to take a detour).

[electau]

The problem seems to be actually getting the 30mA RCBO's, I know Schneider make them but the big outlets don't seem to stock them

As far as measuring the earth leakage is concerned. The IEE recommend clamping the L and N lines with your low amp meter, it will read any imbalance in the same way as your RCD would detect it. You get the 'real' leakage which may not necessarily be flowing in an earth cable (I know it should flow in the earth cable but it may have decided to take a detour).

Yes, that is the other method of checking leakage currents. With portable appliances use a short three (3) core lead with the Line and Neutral conductors separated from the earth so the clamp meter can access them. Source: Clipsal.

An insulation test should be carried out first on the fixed wiring first. Leakage current is additive over a circuit where more than one piece of equipment is plugged into socket outlets. This is especially true with computers etc.

[Pib]

To correctly measure leakage current one must measure the leakage current using a clamp current tester with a low eg. 0 -10mA range, clamped onto the earth wire and the reading taken with the equipment on and then with the equipment off.

While I have a 2 to 500A clamp-on ammeter, I don't have one like you describe which I expect only those in the electrical professional may have. But since a RCBO operates on the principle of an imbalance between the current flows in the L and N lines, it seems hooking a milliamp meter inline/in series with both the L and N lines (right at the RCBO L & N Load connections) as I did would provide the same results when the heater is off, especially since the RCBOs have only tripped when the heaters were off...off for hours. I guess I could do an inline milliamp measurement of the L to E, but I would need to remove the cover of my heater (easy enough as I've done it numerous times to include during this RCBO tripping problem) and put my ampmeter in series with the short ground braid wire to the E connector....I may try that later on as I could actually test for earth leakage with the heater turned on...but as mentioned, the RCBOs have never tripped when the heaters are in operation...just hours/days after they have been turned off. But if there is indeed leakage occurring and it's occurring on an intermittent basis, all the testers in the world may not find the problem....I hate intermittent problems...they can be so, so hard to find...and sometimes you can't find the problem until it completely, finally fails.

It's now been 2 days since the last trip noticed on the morning of 16 Apr as mentioned in my 16 Apr post...a trip on the heater which hadn't been in use for days....the RCBO was not tripped before going to bed as I looked...heater was not used during the night....next morning the RCBO is tripped.

Just to satisfy myself on one thing I again confirmed the RCBO L & N connectors where nice-and-tight because the heater wires are multi-strand, heavy gage wire. With heavy gage, multi-strand wires a person needs to pay extra attention to ensure the connection is tight...and you may need to recheck them after a few days. If a person just initially tightens the RCBO connector screws he can think the connection is tight...wires very secure...but wait a few minutes/hours and/or wiggle the wire connection to the RCBO L & N connectors after initially tightening the connection and the connection may not feel as tight as it originally was...this can be caused by the multi-strand wires kinda moving/settling into position under connection pressure. Best to wiggle, tug, push on the wires right where they connect to the RCBO during the tightening process...then even unscrew the RCBO L & N screws and then re-tighten to allow the multi-strand wire to settle into a more secure/final position.

This is what I did from day one installing the RCBOs to help ensure the best possible, tight connection with heavy gage multistrand wire....I don't have a special torque screwdriver to ensure the recommended 2 N.m torque...all I know the connections are good and tight. When looking into the RCBO connector you can see that the two surfaces that end up squeezing the multi-stand wires are basically flat like a vice...one of the surfaces does have a slight curve but it's still pretty much two flat surfaces being squeezed against the wire...just like a vice. Just bringing this up for others listening in...as with any electrical connection, ensure your RCBO connections are good and tight....preaching to the choir I know.

Thanks for the recommendations and info. Cheers.

Edited April 18, 2012 by Pib

[Pib]

[T_Dog, These SquareD RCBOs are indeed pricey. And I could be wrong but when I went back to HomePro to get the replacement RCBO I actually went to the electrical section to grab the replacement myself and take it back to Customer Exchanges & Refunds, and it looked like the price had increased from 1,595 baht to 1,795 baht since I bought mine on 4 Apr.

I went back to HomePro today to get a variety of stuff and visited the electrical section again...it was less busy today as on 13 Apr and they had finished relabeling/repricing components. Where I mentioned earlier the new & improved price was 1,795 baht, we'll it's actually 1,746 baht for a Square D 10ma 32A, 20A, or 16A RCBO...they had all three in stock. No 30ma Square D RCBOs available nor was there any location marked for them in case they were out each day I have bought or looked at RCBOs at this HomePro. Also, when I looked at Global in Nakhon Pathom a few weeks ago they had the Square D 10ma RCBO (no 30ma version in stock or carried apparently) for a little under 1,700 baht. The only other standalone RCBO this western Bangkok HomePro had was a Haco 40A/30ma for approx 1,300 baht but it was designed for Haco boxes not Square D boxes. Everything else RCD-wise was those Safe-T-Cut combo boxes consisting of circuit breakers and an adjustable RCD from 6 to 30ma...and these boxes were way more pricely than the standalone RCBOs.

[Pib]

I'll probably jinx myself in providing this update but here it goes. As mentioned in my above posts one of the two Square D 10ma RCBOs I bought on 4 Apr 12 for installation on my two water heater circuits had a nuisance tripping problem...it would occasionally trip when the heater was not even in use...it could trip during the day or the middle of the night...no one at home, it just tripped when it felt like it.

When I swapped RCBOs between heater circuits the tripping problem would follow the RCBO. That sure seemed to indicate a faulty RCBO...especially since my circuit testing mentioned in my posts couldn't detect any fault/leakage current....of course the fault/leakage current could be intermittent and just not occuring when I did the test.

Anyway, took the RCBO back to HomePro and they exchanged it no problem. Installed the new one 12 Apr...no trip for for 4 days...then on 16 Apr I noticed it had tripped and the heater hadn't even been in use. On that day I did more fault/leakage current testing...couldn't find/detect any leakage current at all....also checked the RCBO Test function and even used my 7W light bulb hook up between L & E just to reconfirm the RCBO was tripping with fault current (an external test of the Test buttion) and the RCBO passed OK-fine.

So, this time before forgetting about the tripping until the next trip occurred, I pressed Test and reset the RCBO "ten" times and the RCBO has not had a nuisance trip since 16 Apr...which is around 34 days ago. Actually I did this on both RCBOs which included the one not displaying any nuisance tripping. I had even "exercised" the one I took back to HomePro but only did a Test and reset 3 consecutive times over five seconds or so. And just before finishing up this updated, I pressed the Test button to check both RCBOs...they tripped as required...turned the heaters on for about 15 seconds with no RCBO trip.

OK, the replaced RCBO has not tripped for 34 days and the other RCBO has not had any tripping problem regardless of which heater circuit it was in. With all my testing and since I've had no RCBO trip since "exercising" the tripping/replaced RCBO 10 times by pressing the Test button and reseting the RCBO, I wonder if the problem with the returned and replaced RCBOs was a mechanical issue which the "exercising" the RCBO 10 consective times fixed....maybe cleared out/moved some internal manufacturing residue or non-conductive grease, or just something in the RCBO. I've never heard of any RCBO manufacturer saying be sure to exercise a new RCBO X-times so I seriously doubt there is such a need. Hard telling...still could have been an intermittent fault current that will raise its head again...I hate intermittent problems...give me a hard break problem almost anytime (except times like when flying ) so the problem can be specifically located.

Time will tell...knock on wood....haven't had any nuisance trip in over a month on the replacement RCBO and never a nuisance trip on the other RCBO which has been in use for around 46 days...have had one thunderstorm come over...and the power company periodically switching balancing power power loads around my western Bangkok area which causes a split second power loss continues to occur several times a week. Yeap, sure seems to indicate I did get a bad Square D RCBO....more time will tell....knock on wood....just wanted to provide an update. Cheers.

[T_Dog]

Pib... Glad to hear things are working there, and I certainly would not think you need to "break-in" your breaker! After I culled out the bad ones, the five I have in my panel are doing fine. Only one trip when a worker brought over an extension cable with bare wires and dragged it through the mud. We have had a lot of lightning, power outages and brown-outs here as well. I still would not recommend these for anyone to purchase and install with the inconsistencies you and I have seen. It is either a design or manufacturing problem with the defective units.

[Litlos]

Well to throw a couple of different things into the mix from recent experience.

Where I used to work the facilities were designed to US Standards (?). As such we had 220V single phase distribution boards supplied by delta/delta 3 phase transformers, so no real neutral to earth. When I questioned the single phase bit as was obviously 2 phase the local US personnel sort of looked at the floor and said is single phase, sort of? Not in a bulls roar is it single phase is a 2 phase setup, well we call it single phase in the USA, enough said. Anyway we had some ELCB’s installed and requested from CH what to do with the white pig tail wire, nothing was the answer as the unit will work out current balance and what is going “elsewhere”. How does elsewhere even work from our project department, easy through leakage it ends up with a semi earth (imbalance) reference point. Ok, so it appears the units will detect the current imbalance between the legs and not really care where it goes.

The second one is more interesting as happened to a friend while I was away offshore and he was trying to work it out. Post floods he was renovating his house and got a boot when touching the curtain rail, turns out he had taken my advice to put in a Safety Cut, but was set to override by installing sparkie, about as much use as a concrete parachute. Long story cut short he worked out a screw for the curtain rail had touched a wire for a power outlet below. Got another sparkie in who filled hole with silastic and all was good. I asked if he could smell vinegar during the repair, yep sure. Well dig it out as Acetic acid will eat out the copper and that was what you could smell. Oh <deleted>.

The point of the above is that when there are problems with RCBO’s etc tripping it might be real and something not obvious. But a western standard set of tests would have quickly identified a damaged cable insulation to earth leakage. If the RCBO’s cannot be rest due to mechanical failure then this is obviously different.

Cheers

Edited May 20, 2012 by Litlos

[electau]

Well to throw a couple of different things into the mix from recent experience.

Where I used to work the facilities were designed to US Standards (?). As such we had 220V single phase distribution boards supplied by delta/delta 3 phase transformers, so no real neutral to earth. When I questioned the single phase bit as was obviously 2 phase the local US personnel sort of looked at the floor and said is single phase, sort of? Not in a bulls roar is it single phase is a 2 phase setup, well we call it single phase in the USA, enough said. Anyway we had some ELCB’s installed and requested from CH what to do with the white pig tail wire, nothing was the answer as the unit will work out current balance and what is going “elsewhere”. How does elsewhere even work from our project department, easy through leakage it ends up with a semi earth (imbalance) reference point. Ok, so it appears the units will detect the current imbalance between the legs and not really care where it goes.

The second one is more interesting as happened to a friend while I was away offshore and he was trying to work it out. Post floods he was renovating his house and got a boot when touching the curtain rail, turns out he had taken my advice to put in a Safety Cut, but was set to override by installing sparkie, about as much use as a concrete parachute. Long story cut short he worked out a screw for the curtain rail had touched a wire for a power outlet below. Got another sparkie in who filled hole with silastic and all was good. I asked if he could smell vinegar during the repair, yep sure. Well dig it out as Acetic acid will eat out the copper and that was what you could smell. Oh <deleted>.

The point of the above is that when there are problems with RCBO’s etc tripping it might be real and something not obvious. But a western standard set of tests would have quickly identified a damaged cable insulation to earth leakage. If the RCBO’s cannot be rest due to mechanical failure then this is obviously different.

Cheers

Two phases of a three phase system operates as 2 wire single phase. The system you describe if it had no neutral it must be an IT system. RCDs are not used on IT systems they use LIMs line insulation monitors that will alarm only when earth leakage current exceed 5mA, They do not disconnect the supply. A neutral is earthed or it is not. I think you will find that the neutral was earthed. The US uses some strange transformer connections. A delta system can be earthed using a certain type of connection. This is the neutral connection.

IT would be interesting to know what tests where carried out on the RCDs after installation. Strange that the electricians installing the RCDs did not know what the small white pigtail was for. It is connected to the earth bar and it is an FE (functional earth) and ensures operation of the RCD if the neutral fails.

Setting an adjustable RCBO (Thai Safe-T -Cut) to the direct postion will isolate the RCD function so it will not operate on earth leakage. This is one of the reasons that they do not comply with IEC, BS requirements.

[electau]

RCDs and RCBOs are very reliable and the fallure rate is very low. From personal experience is that an example of 2 defective units in a sample of over 2000 on a five year basis, the RCDs where tested when installed and at regular intervals then at 6 month and 2 year intervals. The six month tests were operation by the test buttom only, the 2 year tests were time /current tests in accordance with AS/NZ standards.

Nuisance tripping is not indication of a faulty RCD, it is generally doing its job.

RCDs will only protect a circuit and what is connected to it on the load side of the RCD.

Insulation resistance should be a minimum of 1.0 Megohms, Line to Earth and Neutral to Earth.

Standing leakage current to earth on a circuit protected by an RCD should not exceed 3 mA for a 10mA RCD and 10mA for a 30mA RCD.

An RCD suspected of being faulty must be tested with the required test equipment according to a standard.

RCDs if tested faulty must be replaced they are not repairable.Opening a RCD will void any warranty.

Most RCD failure are failure to operate when the test button is used and failure to operate when a time/current test is used. Socket outlets must be earthed for this test ( time/current) to be carried out.

Tripping may be caused by 1, a defective item of equipment or 2. failure of insulation on circuit wiring to earth. 3. moisture in equipment or in a junction box etc.

RCDs must be installed in accordance with a recognised standard, IEC, BS or AS/NZ etc.

RCDs in an residential type environment should be tested using the test button at a minimum of every 3 months.

Some RCDs have a white pigtail this is a FE (functional earth) and it is connected to the earth bar, it ensures that the RCD will operate on a neutral failure.

The correct use of RCDs in accordance with a standard require that the supply neutral is earthed/grounded ie TT , TN-S or TN-C-S. Equipment and socket outlets should be earthed with a protective earth(PE).

Where surges or spikes are prevalent on the supply network installing SPDs between the main circuit breaker and the line side of the RCDs may solve this problem.

[T_Dog]

Again, all good information but this thread is about a particular model that either has a design problem or manufacturing problem. I have used various protective devices like these in the past and I have never experienced a 25 % failure rate such as with these RCBO's. They do indeed fail, as the problem follows the breaker and the new ones work fine when replaced. If I could do it over again, I would not use them, but would have implemented a different solution.

[Pib]

Indeed electau's info is good info indeed (thanks electau).

However, I pretty much feel like T_Dog regarding the Square D RCBOs I got--or at least that one which I had to return and its replacement especially with all the earth leakage and other testing I did to try to ensure it was just a faulty RCBO versus actual earth leakage on the circuit. Now I don't have the high end equipment a professional electrican would have to do fault testing, but I feel my multimeter, swap-out, and visual testing was about the best a layman could do.

RCBO installation was easy enough and fortunately I have dedicated L, N, & E wires running to my heaters with no tag-on outlets/switches. And my house is only 4 years old so hopefully PVC insulated 750V 70C rated wiring is still in good shape and the heaters looked & ohmed-out fine on the inside...of course I can't track the circuit wiring for every inch from service box to the heaters since the wire is enclosed in PVC pipe/metal conduit for about 98% of the run. When moving the suspected bad RCBO to the other heater circuit and it trips within minutes and the other testing I mentioned in my posts while the other RCBO continues to operate fine regardless of which heater circuit it's placed in convinces me in my case the RCBO was just defective. And when the replacement tripped within days but has been working fine for over a month now after doing that ten times exercise thing.....like that exercise may have resolved a RCBO mechanical issue. Right now, I'm just gun-shy and watchful of these two Square D RCBOs I now have.

Approx monthly I'll continue to Press the RCBO Test button and also do a separate test of touching a 7W light bulb between L & E lines on the heater circuits to help confirm both RCBOs continue to work properly....I know 7W bulb pulls approx 3-4 times the trip current of these 10ma RCBOs but at least that 7W current pull ensures the RCBO is at least tripping at approx 30ma which is in the safe ballpark for impacts on the ol' body....heck, I may see if I can I pickup a 3-5W bulb today at Global which would test the RCBOs in the 20ma ballpark...I might even get lucky and actually get a 3-5W bulb that only pulls around the 3W area and then I'll be testing the 10ma RCBOs at around 14ma asuming a low wattage bulb's initial high surge current (higher than operating current) for a few milliseconds don't cause a false trip (even if I had a 1W bulb which normally pulls a little less than 5ma on a 220V circuit) and is just a bad way to do an easy & cheap test without having to solely rely on the RCBO's Test button function.

Any comments on using a low wattage 220V bulb in the 3 to 7 watt ballpark in doing a RCD trip test? I think I've seen other posts where people use 15W bulbs which would be pulling around 68ma.

Knock on wood....both current RCBOs continue to work.

[electau]

A 10W incandesant lamp will trip an RCD. The resistance is very low when cold. This proves that the RCD will trip. Along with the test button this is all that is required.

Portable RCD units should be tested by way of the test button each time before use.

The exact tripping current in mA is not important , but the RCD must trip using the test button.

Edited May 23, 2012 by electau

[electau]

Again, all good information but this thread is about a particular model that either has a design problem or manufacturing problem. I have used various protective devices like these in the past and I have never experienced a 25 % failure rate such as with these RCBO's. They do indeed fail, as the problem follows the breaker and the new ones work fine when replaced. If I could do it over again, I would not use them, but would have implemented a different solution.

Automatic disconection of supply is the most common method of protection against indirect contact in the case of an insulation fault.

Use of RCDs comply with this requirement, they are also used when the earth fault impedance is too high for a MCB to operate in the required time (less than 0.4secs).

[Pib]

A 10W incandesant lamp will trip an RCD. The resistance is very low when cold. This proves that the RCD will trip. Along with the test button this is all that is required.

Portable RCD units should be tested by way of the test button each time before use.

The exact tripping current in mA is not important , but the RCD must trip using the test button.

Yesterday I picked up a Daiichi bulb rate 3-5W at Global for 6 baht. When cold an ohmeter says it has a 460 ohm resistance which would pull approx 490ma on my 225V circuit for a few milliseconds (don't know how many) until it heated up. With the light turned on it draws approx 43ma (pretty steady also from the very first second the milliammter gives a reading) which equates to a little under 10W...basically the same wattage the other bulb rated at 7W pulled. Thought for second maybe my AC milliamp function on my multimter was way off, but I pulled out a Neon indicator lamp which is suppose to pull around 1ma (also depends on assocated current limiting resistor) and my AC milliamp function measured 1.5ma. So, my milliammeter is working fine. Even checked a current pull on a 60W bulb and it caculated out to 55W. So, it appears low wattage bulbs in the 3 to 7W ballpark have a W.....I......D......E manufacturing tolerance but a higher wattage bulb will be closer to its actual wattage rating based on my limited tests.

I did put the Neon lamp pulling 1.5ma across L-E on both my Square D 10ma RCBO and neither tripped. I've got seven of these Neon lamps...maybe someday when I really bored I'll put some in parallel to increase L-E current draw by 1.5ma per bulb.....4 in parallel would pull 6ma...6 in parallel would pull 9ma....7 in parallel would pull 10.5ma......but I'm not bored enough yet to do that. I'll continue to keep my eyes open for a dirt cheap variable resistor or standard resistors of high enough wattage rating which I could use to do much more accurate RCD testing, at least tripping current-wise. Cheers.

[electau]

The RCD will trip with an incandesent lamp long before the lamp will even heat up, as the RCD will most probably trip in 20 to 30mS.

The standard test is the by the test button on the device. the other tests are time/current tests and these can only be carried out using the correct test equipment in accordance with a recognised standard (eg AS3670).

A 30mA RCD must operate in less than 300mS and a 10mA in less than 40mS.

The use of the test button complies with the requirements of electrical regulatory bodies including AS3000. It complies with the 3 month test as carried out by a consumer.

Neon indicators are not used to trip an RCD as they will give indication of power available (which is what one wants to as to prove supply)

Simple test devices are available with a switched range of resistors and 3 neon indicators for polarity. Tests they cover from 5mA/10mA/15mA/30mA. Fitted with a 3 pin plug top and short flex 3 core lead. They are not expensive.

However devices for time current and sensitivity testing are expensive.

The test resistance ( fixed value resistor) for a 30mA RCD is 7666 ohms, and for a 10mA RCD it is 23000 ohms.

When a device is removed for additional testing if suspected of being faulty it is given three tests in succession, the test of the test button, and two time tests at rated mA , at 0 degrees and 180 degrees of the AC waveform and it must operate in less than the specified time in mS.

[electau]

For anyone interested in the reliability of RCDs this imformation from the website below is interesting. There are PDF documents on the subject which may be of interest.

http://www.justice.qld.gov.au/fair-and-safe-work/electrical-safety/for-householders-and-consumers/safety-switches/safety-switch-reliability-research-project

Edited May 28, 2012 by electau

[Pib]

I scanned the research project quickly...some interesting info. Based on the limited scope of this particular research/survey the reliability data shows a 96 to 99% reliability rate based on RCDs that have been in use for a while, but it does not appear to provide reliability data on brand new RCDs which failed out of the box/returned under warranty...and we all know X-percentage do fail from the get-go for any product we buy. This failure in the early stage of a product's life is commonly referred to as infant mortality and it's even addressed in the report. And the report said only 1.7% of the RCD tested were less than a year old. So, there appears be little to no reliability data on RCDs that are brand new or within their usually warranty period of 1 year....that infant mortality barkpark.

When it comes to new RCDs returned to the store for a replacement/refund within days/weeks of purchase it would be interesting to know what that infant mortality failure rate is, but that would require RCDs to get returned all the way to the manufacturer and the manufacturer collecting real world failure data.....or the manufacturer to provide manufacturing quality assurance sampling results. But that kind of manufacturer data would probably be considered close-hold/proprietary in nature. I don't know of any manufacturers of any product that posts results on what percentage of their products failed out of the box/within the warranty period.

I think the partial quote below from the report does a good job of summing up the report's success in getting failure data from most manufacturers (but apparently it did get a few to record/provide some data for a 90 day period) :

It was also found that relying on industry goodwill to volunteer information about safety switch reliability is problematic. Feedback

received suggested that organizations generally found recording test results for this study a burden.

When it comes to returning a brand new item (regardless of what it was) to the store for a replacement/refund because that new item didn't work properly, I think we have all done that numerous times over our lives. I'm still knocking on wood...the replacement RCBO I got is still working fine (i.e., passes the Test button check and no nuisance trips compared to the one I took back which was tripping every few hours to days).

Another area of the report identified something I've seen in other similar reports, that RCD regular testing maintains reliability (i.e., pushing the Test button). Beyond the fact the Test button helps to confirm the RCD is still working, maybe this testing "exercise" of the RCD so to say helps to keep the mechanical portion of the RCD free to move/in better condition/helps clear any contaminants.

I have no doubt that RCDs on the whole are very reliable, but I'm never surprised anymore (but maybe a little mad) when I have to take a brand new item (any product) back for a replacement/refund because it didn't work properly out of the box. Oh well, stuff happens. Cheers.