Open Service Neutral GFCI/RCD Detector

[winnie]

I am again repeating myself.

You don't select trip level based on frequency of tripping. You select trip level based on protecting what you want to protect.

If you wish to use residual current detection as the prime protection from electric shock, then you must use a class A GFCI, at 5-6mA. 5-6mA is generally considered the level of no harm from shock.

See https://www.wikilectures.eu/w/Physiological_effect_of_electric_current to understand what happens when current levels go up.

30mA protection is suitable for equipment protection. Also it _may_ reduce shock hazard in the context of proper bonding of metallic equipment. If you have a properly grounded metallic frame appliance, and have a short in the appliance, then the frame gets energized until the OCPD opens. If you have a 30mA RCD, then the frame is only very slightly energized before the RCD opens.

But now imagine you have an _ungrounded_ appliance. Some sort of fault energizes the chassis, but since there is no grounding there is no significant fault current flow, and the RCD has nothing to detect. Now someone touches the frame and ground, the current flows through them. If the current through the person is sufficient, then the RCD opens. However the RCD will not open until the shock is rather severe.

So if you want to protect _people_ in the context of circuits without grounding, then use a class A GFCI, and if you get tripping during normal operation then you should consider the device unsafe. It may not be 'broken', but the device is not safe to be used without proper grounding.

If you want to enhance the protection of your _circuits_ and _appliances_, then go ahead and use 30mA RCD breaker. But this will not provide protection for _people_ from dangerous levels of electric shock. It will reduce the maximum severity of a shock, but would permit a painful shock that is severe enough that a person might not be able to 'let go'.

-Jon

 

[tersh]

I am again repeating myself.

You don't select trip level based on frequency of tripping. You select trip level based on protecting what you want to protect.

If you wish to use residual current detection as the prime protection from electric shock, then you must use a class A GFCI, at 5-6mA. 5-6mA is generally considered the level of no harm from shock.

See https://www.wikilectures.eu/w/Physiological_effect_of_electric_current to understand what happens when current levels go up.

30mA protection is suitable for equipment protection. Also it _may_ reduce shock hazard in the context of proper bonding of metallic equipment. If you have a properly grounded metallic frame appliance, and have a short in the appliance, then the frame gets energized until the OCPD opens. If you have a 30mA RCD, then the frame is only very slightly energized before the RCD opens.

But now imagine you have an _ungrounded_ appliance. Some sort of fault energizes the chassis, but since there is no grounding there is no significant fault current flow, and the RCD has nothing to detect. Now someone touches the frame and ground, the current flows through them. If the current through the person is sufficient, then the RCD opens. However the RCD will not open until the shock is rather severe.

So if you want to protect _people_ in the context of circuits without grounding, then use a class A GFCI, and if you get tripping during normal operation then you should consider the device unsafe. It may not be 'broken', but the device is not safe to be used without proper grounding.

If you want to enhance the protection of your _circuits_ and _appliances_, then go ahead and use 30mA RCD breaker. But this will not provide protection for _people_ from dangerous levels of electric shock. It will reduce the maximum severity of a shock, but would permit a painful shock that is severe enough that a person might not be able to 'let go'.

-Jon

In the USA. What GFCI do you use for your 240v appliances? What have you tried? Have you tried a 240v rice cooker on 5mA GFCI? Maybe somehow 240v rice cooker is too much for 5mA GFCI?

 

[romex jockey]

In the USA. What GFCI do you use for your 240v appliances? What have you tried? Have you tried a 240v rice cooker on 5mA GFCI? Maybe somehow 240v rice cooker is too much for 5mA GFCI?

Mostly hot tubs .....i don't own or even know what a rice cooker is,sorry....:ashamed1:~RJ~

 

[romex jockey]

He has explicitly stated that there are no EGCs in the wiring, and no neutral-earth bond in his service, and no neutral conductor to his service, and that these defects are not practical to correct.

He is essentially inventing band-aids to cover the poor installation practices used in his country.

-Jon

I'm wondering if simply bonding everything possible would be the best bandiad ??

~RJ~

 

[tersh]

Mostly hot tubs .....i don't own or even know what a rice cooker is,sorry....:ashamed1:~RJ~

Now I'm getting it why no one else has much experience on these 2-pole 240v GFCI breakers. Because you only use it on tubs and spas.

In the Philippines, rice is what we primary eat at lunch time and dinner. The poor eat rice and salt only. In the US, it's potatoes. Here is a rice cooker. Is it possible you haven't eaten rice before in your life?

It doesn't have any EGC (Equipment Grounding Conductor). All appliances don't have EGC.

Notice the flooring below. It's made of porcelain or marble. Are porcelain or marble insulator or can these conduct electricity (in ground fault)?

 

[mbrooke]

Maybe the Europeans are also doing band-aids for the TT system?

A properly installed TT system does not have any bandaids, unless one wants to debate the concept itself being a band-aid.

I just dropped by the home now and saw the plugs from the generic made in china Meiji GFCI receptacle transferred to another outlet. My parent told me it was because the GFCI kept tripping. It's connected to two rice cookers. Also I read this feedback from a European. He said something that made me wonder. Are 240v rice cooker prone to tripping that is why the Europe increases the threshold to 30mA? (Does your US rice cooker also keep tripping on the 5mA GFCI or not?) quoting him:

US rice cookers do not trip GFCIs. I personally can't see a 230 volt rice cooker tripping a 6ma socket GFCI either. I'll make it simple: if that rice cooker trips a 30ma, you know it is faulty and needs to be fixed or thrown out.


Europe uses 30ma because its common to protect multiple circuits which means cables leakage + appliance leakage is often great enough to trip 5ma GFCIs.

"Here in Europe, RCDs are usually installed to protect entire circuits, not just single socket outlets. Each appliance may leak a new mA to ground, even if it is working correctly. Electronic devices with suppression capacitors, and devices with mineral insulated heating elements (such as cookers and water heaters) are usually the worst for this.

This is correct- electronic suppression caps do leak. Electric showers are often the worst offenders by far.

So using a 5mA or 10mA RCD will probably result in nuisance tripping.

So for protecting people from electrocution, 30mA was chosen as a compromise. The 30mA is the maximum at which the RCD must trip. Manufacturing variances mean that some will trip at lower currents.

Earthing (grounding) arrangements vary wildly across Europe. Here in the UK we still use three different arrangements - TT, TN-S and TN-C-S.

Correct

If it is a TT installation, the supplier provides no ground, and the customer must use an earth rod, which is not connected to the supply neutral. Even a dead short between live and the consumer's earth will not pass enough current to blow the supply fuse. So it is common to fit a 100mA time-delayed RCD to the incoming supply, which will trip instead. In large commercial installations, a 100mA RCD may still nuisance trip. So it may be necessary to increase the RCD to 300mA or more. Note that these RCDs are there to detect and clear faults, not to protect people from electrocution.

100mA or more RCDs are sometimes also used where long cable runs are necessary, and the time to clear the fault by a fuse blowing or circuit breaker tripping may be too long.

Correct- there is usually a main time delayed RCD and 30ma sub main RCDs or individual RCBOs. The 30ma provides people protection, the main time delayed backup is to provide backup protection for clearing faults should the branch RCD protection fail. The 2x concept is not mandated by IEC 60364, but often considered good practice because a fault left standing on TT is considered very dangerous.

European 230V RCDs will have four terminals - live in, live out, neutral in and neutral out. There is no provision for two phases, plus neutral. So there is no guarantee that an RCD connected only across the two phases would detect a fault current leaking to neutral or ground."

I could be wrong, but I can't see why 2 phases and a neutral would not work if run through a 3 or 4 pole RCD. A torrid summation is a torrid summation. As long as everything cancels the torriod will not care how the phase angles and magnitudes sigma to zero.

 

[mbrooke]

I am again repeating myself.

You don't select trip level based on frequency of tripping. You select trip level based on protecting what you want to protect.

If you wish to use residual current detection as the prime protection from electric shock, then you must use a class A GFCI, at 5-6mA. 5-6mA is generally considered the level of no harm from shock.

See https://www.wikilectures.eu/w/Physiological_effect_of_electric_current to understand what happens when current levels go up.

30mA protection is suitable for equipment protection. Also it _may_ reduce shock hazard in the context of proper bonding of metallic equipment. If you have a properly grounded metallic frame appliance, and have a short in the appliance, then the frame gets energized until the OCPD opens. If you have a 30mA RCD, then the frame is only very slightly energized before the RCD opens.

Correct from a UL perspective- but internationally 30ma is the value considered acceptable people protection.


http://www.electrical-installation.org/enwiki/Protection_against_electric_shock

But now imagine you have an _ungrounded_ appliance. Some sort of fault energizes the chassis, but since there is no grounding there is no significant fault current flow, and the RCD has nothing to detect. Now someone touches the frame and ground, the current flows through them. If the current through the person is sufficient, then the RCD opens. However the RCD will not open until the shock is rather severe.

So if you want to protect _people_ in the context of circuits without grounding, then use a class A GFCI, and if you get tripping during normal operation then you should consider the device unsafe. It may not be 'broken', but the device is not safe to be used without proper grounding.

Not if it trips fast enough. Its not only magnitude, but duration as well.

 

[tersh]

I let the electrician removed from the wall the universal made in china 240v GFCI (with 6mA tripping) receptacle http://www.internationalconfig.com/icc6.asp?item=74900-RCDS (these are OEM in many other brands) because my mom doesn't like to use it, and brought it to the Meiji office to ask them how they would know it has self test and if it's really 5mA tripping. They have an instrument to test it:

The Meiji technician was holding the receptacle. He said according to the china factory. You can test if it has auto maintain self-test feature in UL 2015 by lowering the voltage to 20 volts and seeing if there is a red light that would appear in the unit. It appeared at about 20 volts. You can't press any button.. but neither when the voltage is 150 volts or below (but the weak red light only appearing at 20volts).

Have you heard anything like it where this is used to test the auto maintain self-test feature in UL 2015. Do you understand the logic of it?

Second. I told them how to know it really trips at 5mA. They have another test for it.

Here when the load is adjusted above 5mA leakage. It trips. So the 6mA rating may be real. But I'm still not sure about the auto maintain self test.

I need to know something. How difficult it is to build GFCI or the toroid with differential sensor and tripper? Can China really build such with quality (the same model is OEM in many brands worldwide not just to Meiji)? Is it like rocket science, or requiring very high tech stuff like Intel factory or something. I just want to know the long term reliability of china made 240v GFCI with 5mA tripping.

For those who are not familiar with the auto-maintain self test. It is NOT the white button you press to test but something integrated in the circuitry within the unit:

"New Self-Test Requirements:n Underwriters Laboratories (UL) made new revisions to the UL 943 (GFCI) standard that will take effect onJune 29th, 2015.n UL REVISION 943 STATES:1. All GFCI receptacles must have an auto-monitoring (self-testing) feature.2. If auto-monitoring determines the GFCI can no longer offer protection, one or more of the following will happen1) Unit is no longer able to reset and denies power.(2) Unit can reset and is subject tothe next auto-monitoring testcycle within 5 seconds of powerto the device.(3) All GFCIs must provide audible or visual indication if it does NOTgo into power denial.3. Provisions to ensure receptacle will not reset if miswired during installation aswell as reinstallation."​

 

[tersh]

I let the electrician removed from the wall the universal made in china 240v GFCI (with 6mA tripping) receptacle http://www.internationalconfig.com/icc6.asp?item=74900-RCDS (these are OEM in many other brands) because my mom doesn't like to use it, and brought it to the Meiji office to ask them how they would know it has self test and if it's really 5mA tripping. They have an instrument to test it:

The Meiji technician was holding the receptacle. He said according to the china factory. You can test if it has auto maintain self-test feature in UL 2015 by lowering the voltage to 20 volts and seeing if there is a red light that would appear in the unit. It appeared at about 20 volts. You can't press any button.. but neither when the voltage is 150 volts or below (but the weak red light only appearing at 20volts).

Have you heard anything like it where this is used to test the auto maintain self-test feature in UL 2015. Do you understand the logic of it?

Second. I told them how to know it really trips at 5mA. They have another test for it.

Here when the load is adjusted above 5mA leakage. It trips. So the 6mA rating may be real. But I'm still not sure about the auto maintain self test.

I need to know something. How difficult it is to build GFCI or the toroid with differential sensor and tripper? Can China really build such with quality (the same model is OEM in many brands worldwide not just to Meiji)? Is it like rocket science, or requiring very high tech stuff like Intel factory or something. I just want to know the long term reliability of china made 240v GFCI with 5mA tripping.

For those who are not familiar with the auto-maintain self test. It is NOT the white button you press to test but something integrated in the circuitry within the unit:

"New Self-Test Requirements:n Underwriters Laboratories (UL) made new revisions to the UL 943 (GFCI) standard that will take effect onJune 29th, 2015.n UL REVISION 943 STATES:1. All GFCI receptacles must have an auto-monitoring (self-testing) feature.2. If auto-monitoring determines the GFCI can no longer offer protection, one or more of the following will happen1) Unit is no longer able to reset and denies power.(2) Unit can reset and is subject tothe next auto-monitoring testcycle within 5 seconds of powerto the device.(3) All GFCIs must provide audible or visual indication if it does NOTgo into power denial.3. Provisions to ensure receptacle will not reset if miswired during installation aswell as reinstallation."​

Click here for the full size image of the above so you can see all the numbers in the knobs (It can test for leakage current from 0 to 500mA but I still don't understand the logic of lowering it to 20v to test for auto-monitoring self test capability). or click this.. https://imageshack.com/a/img921/2604/o4JNc2.jpg

 

[mivey]

Yes, I've been following. They are all in the realm of DIY.

Pretty much sums up the threads.

 

[tersh]

Pretty much sums up the threads.

It's not a DIY. But studying the different standards in different countries for an understanding of the importance and implementation of grounding.

Also in a country with something less than a TT. We need superior GFCI/RCD technology. So there must be additional vendors that must be familiar with them.

I'm thinking whether it's feasible to be a vendor. Hence exploring all sorts of combination that can give better protection to a country where almost all appliances and homes don't have any EGC and city hall inspectors don't even mind.

 

[mbrooke]

Pretty much sums up the threads.

But its very educational. And shows you how lucky we are- not every country is blessed with top notch reasonably safe electrical systems. Has you racking your mind on how to improve on it with whats all ready there.

 

[tersh]

First. Meiji just buys it from this OEM distributor http://www.internationalconfig.com/icc6.asp?item=74900-RCDS
and label it as Meiji. Anyway supplier shows me the internal parts of it. Anyone recognize if it has an auto-monitoring self test IC (or circuit pattern) or something?

This is the front of it.

​

Here is the back of it:

Here is the side of it (other side is symmetric to it):

Any internal pictures of any US 120v GFCI 5mA receptacles?​

 

[tersh]

I found the datasheet of the Meiji FM2141 GFCI chipset.

http://elcodis.com/parts/6008223/FM2141_p3.html#datasheet

It looks like there is no self test which is clearly emphasized in the Fairchild FAN4149

http://www.mouser.com/ds/2/149/FAN4149-365238.pdf


Product Overview
Introduction
FM2141 is a low power controller for AC receptacle ground fault circuit interrupters. The devices
detect hazardous current paths to ground and ground to neutral faults. The circuit interrupter then
disconnects the load from the line before a harmful or lethal shock occurs.
FM2141 contains a diode rectifier, shunt regulator, precision sense amplifier, current reference, time
delay circuit and SCR driver internally. Two sense transformers, SCR, solenoid, three resistors and
four capacitors complete the design of the basic circuit interrupter.
The simple layout and minimum components insure ease of application and long term reliability.
Features not found in other GFCI controllers include a low offset voltage sense amplifier eliminating
the need for a coupling capacitor between the sense transformer and sense amplifier, and an
internal rectifier to eliminate high voltage rectifying diodes.
FM2141 is powered only during the positive half period of the line voltage, but can sense current
faults independent of its phase relative to the line voltage. The gate of the SCR is driven only during
the positive half cycle of the line voltage.
Features
Powered from the AC line
Built-in rectifier
Direct interface to SCR
500
A quiescent current
µ
Precision sense amplifier
Adjustable time delay
Minimum external components
Meets UL 943 requirements
For use with 110V or 220V systems
Available in 8 pin SOP or DIP package

btw.. forgot to show to the front toroid pic: The wire in the winding seems a bit small

20A, 250V. Ain't it. ​

​

 

[Adamjamma]

<sub>hTersh, notice what the European said... 30 was chosen as a compromise because of the tripping in Europe of certain appliances. Because of the use of the RCD to cover several circuits...
US gfci covers one circuit. It also covers one appliance sometimes, in the case of certain wired in or cabinet covered appliances. Wiring practices, especially in the kitchen, are different. NEC requires two circuits minimum for a kitchen... UK requires two outlet circuits basically for the whole house...

Also, you refer to the TTsystem and yes, that is closest to what your home has... but it still would have grounds run to all outlets and neutrals run as well.
the difference between the TT system and the normally used other systems comes down to theno ground run from the transformers... and a ground rod required then at the residence which you have said was neglected at your parents home as well as no neutral.

Now, there is argument over the need for such and wether the neutral is supplied, and if there is any difference between the TT system and what the USA uses...my own instructor here has argued that the USA uses a TT system because they only supply two live and a neutral, and the ground is supplied at the home, while in the UK the ground and the neutral are supplied by the utility in the PEN conductor, and a ground rod at the home is not necessary but is a safety precaution.
again, I am just a student, but, if it was my home or my parents home then I would be trying to rewire the home using a good electrician so that every outlet had the live wires, a neutral wire, and a ground wire, using at least NMB 4 wire or is done in conduits, four wires... and I would get the neutral wire run by the electrical provider... but I am not you.</sub>

 

[Adamjamma]

Oh, and the torrid is not wound to small because all it is doing is picking up the current from the outside of the wires.
It is not transforming any current actually but only getting enough to sense the differences

 

[tersh]

<sub>hTersh, notice what the European said... 30 was chosen as a compromise because of the tripping in Europe of certain appliances. Because of the use of the RCD to cover several circuits...
US gfci covers one circuit. It also covers one appliance sometimes, in the case of certain wired in or cabinet covered appliances. Wiring practices, especially in the kitchen, are different. NEC requires two circuits minimum for a kitchen... UK requires two outlet circuits basically for the whole house...

Also, you refer to the TTsystem and yes, that is closest to what your home has... but it still would have grounds run to all outlets and neutrals run as well.
the difference between the TT system and the normally used other systems comes down to theno ground run from the transformers... and a ground rod required then at the residence which you have said was neglected at your parents home as well as no neutral.

Now, there is argument over the need for such and wether the neutral is supplied, and if there is any difference between the TT system and what the USA uses...my own instructor here has argued that the USA uses a TT system because they only supply two live and a neutral, and the ground is supplied at the home, while in the UK the ground and the neutral are supplied by the utility in the PEN conductor, and a ground rod at the home is not necessary but is a safety precaution.
again, I am just a student, but, if it was my home or my parents home then I would be trying to rewire the home using a good electrician so that every outlet had the live wires, a neutral wire, and a ground wire, using at least NMB 4 wire or is done in conduits, four wires... and I would get the neutral wire run by the electrical provider... but I am not you.</sub>

So in short for the Siemens subpanel with 6 GFCIs 5mA. A single 30mA RCD would be good for it just to protect the Siemens panel shorting to any hot wire from main feed or melting bus bar, right?

 

[Adamjamma]

Think you are trying to protect from arc fault rather than ground fault here... not sure you really need another ground fault here. But have to let others chime in. The grounds and neutrals being run in my mind are more important.

 

[tersh]

Consider this Siemens subpanel with all 6 pcs GFCI 5mA (only 2 installed for presentation).

If the load side of the GFCI touches panel chassis, the GFCI would trip. But what would protect the main feeder wires from main panel itself (not yet put)? It has to be an RCD connected before the subpanel.. isn't it? I was asking if 30mA RCD would be sufficient.

 

[tersh]

_{hTersh, notice what the European said... 30 was chosen as a compromise because of the tripping in Europe of certain appliances. Because of the use of the RCD to cover several circuits...
US gfci covers one circuit. It also covers one appliance sometimes, in the case of certain wired in or cabinet covered appliances. Wiring practices, especially in the kitchen, are different. NEC requires two circuits minimum for a kitchen... UK requires two outlet circuits basically for the whole house...}

What would happen if a single 30A GFCI breaker (5mA) is supplying 5 outlets. And you have 5 appliances connected to each. Would each appliance leak 1mA? Because this is what I intend to use it per room. So the 5 outlets in each room will power say 1 electric fan, 1 TV, 1 iPhone charging, 1 hairdryer, one music stereo.. can each leak 1mA? If it does, then the plan is in trouble.

Also, you refer to the TTsystem and yes, that is closest to what your home has... but it still would have grounds run to all outlets and neutrals run as well.
the difference between the TT system and the normally used other systems comes down to theno ground run from the transformers... and a ground rod required then at the residence which you have said was neglected at your parents home as well as no neutral.

Now, there is argument over the need for such and wether the neutral is supplied, and if there is any difference between the TT system and what the USA uses...my own instructor here has argued that the USA uses a TT system because they only supply two live and a neutral, and the ground is supplied at the home, while in the UK the ground and the neutral are supplied by the utility in the PEN conductor, and a ground rod at the home is not necessary but is a safety precaution.
again, I am just a student, but, if it was my home or my parents home then I would be trying to rewire the home using a good electrician so that every outlet had the live wires, a neutral wire, and a ground wire, using at least NMB 4 wire or is done in conduits, four wires... and I would get the neutral wire run by the electrical provider... but I am not you.

USA is not TT because ground current passes through the neutral and to centertap which can trip breaker. In pure TT, since no conductor to transformer neutral, then breaker won't trip. Hence you use RCD. Let's assume this is my scenario where there is only ground connected to the local electode, then an RCD of 30mA before the Siemens subpanel with 6 GFCIs would work, isn't it?