Leviton GFCI nuisance tripping and circuit analysis

[tersh]

I made a series of tests.

I inserted a 240v surge protection between the waterpik shade pole motor and Leviton chipped GFCI. It only trips rarely, like once in 20 switchings!

Here I inserted a 1:1 medical grade isolation transformer (with dielectric shielding between primary & secondary) between the waterpik shaded pole motor and GFCI.

It doesn't trip at all. But is it not transients still pass through an isolated transformer and only the common mode surges were suppressed? But noticed there is no ground, the surface is glass, so could the 240v 1:1 isolation transformer filtering some line to line transients too? The model of it is: http://catalog.triadmagnetics.com/Asset/VPM240-2080.pdf

I bought the 240v 1:1 isolation transformer supposedly to put it on the refrigerator which I tested to trip last year on the GFCI outlet. I thought it trips on all GFCIs. But when I tested the Siemens GFCI breaker lately, the refrigerator no longer trips!

The meter at the left side is a flux meter. I'm studying the flux properties of the shaded pole motors. Note the behavior is the same whether the rotor is put in the stator or not. The flux meter is also to verify if the motor is turned on or off. The needle jumps above 100mG when it is turned on.

Here the shaded pole motor is connected directly to the GFCI. The tripping is 80-90% of the time.

Do you have any flux mapping or plot of a shaded pole motor? Why does it produce more inductive kick than other motors. I noticed the magnetic flux is very strong, more than a 500va isolation transformer or autotransformer. Does motor with very large magnetic flux have more inductive kick?

Btw.. there is no other 240v GFCI receptacle with 5mA tripping in the entire world. Only one brand carries it. It doesn't have the auto-monitoring self test in UL 2015. So all using 240v are stuck with it and have to deal with shaded pole motors (in refrigerator, etc.) that can trip it.

 

[tersh]

I made a series of tests.

I inserted a 240v surge protection between the waterpik shade pole motor and Leviton chipped GFCI. It only trips rarely, like once in 20 switchings!

Here I inserted a 1:1 medical grade isolation transformer (with dielectric shielding between primary & secondary) between the waterpik shaded pole motor and GFCI.

It doesn't trip at all. But is it not transients still pass through an isolated transformer and only the common mode surges were suppressed? But noticed there is no ground, the surface is glass, so could the 240v 1:1 isolation transformer filtering some line to line transients too? The model of it is: http://catalog.triadmagnetics.com/Asset/VPM240-2080.pdf

I bought the 240v 1:1 isolation transformer supposedly to put it on the refrigerator which I tested to trip last year on the GFCI outlet. I thought it trips on all GFCIs. But when I tested the Siemens GFCI breaker lately, the refrigerator no longer trips!

The meter at the left side is a flux meter. I'm studying the flux properties of the shaded pole motors. Note the behavior is the same whether the rotor is put in the stator or not. The flux meter is also to verify if the motor is turned on or off. The needle jumps above 100mG when it is turned on.

Here the shaded pole motor is connected directly to the GFCI. The tripping is 80-90% of the time.

Do you have any flux mapping or plot of a shaded pole motor? Why does it produce more inductive kick than other motors. I noticed the magnetic flux is very strong, more than a 500va isolation transformer or autotransformer. Does motor with very large magnetic flux have more inductive kick?

Btw.. there is no other 240v GFCI receptacle with 5mA tripping in the entire world. Only one brand carries it. It doesn't have the auto-monitoring self test in UL 2015. So all using 240v are stuck with it and have to deal with shaded pole motors (in refrigerator, etc.) that can trip it.

This is inside the CD-R king surge protector:

It's from file photo I took about 6 months ago. Notice the 3 MOVs (Metal Oxide Varistors). It's only line to line only (made in china) so they are all connected line to line. Our electricity is equivalent to the 240v red and black hot lines in the US so we don't have any line to neutral, or neutral to ground. So there is no protection modes for them unlike in US manufactured surge protectors.

Now my question about it. Inductive kicks are not surges, are they?! Why did the surge protection strip eliminate the tripping to almost 5% only? It trips once in every 20 switchings. Without the surge protection strip, the shaded pole motor trips 4 out of 5 switchings in the Leviton based GFCIs (never in the Siemens GFCI which perhaps contain more MOVs)?

 

[junkhound]

Inductive kicks are not surges, are they?!

Yest they are.

The MOVs with clip the top off the inductive voltage kicks, by providing a low impedance path. Same type peak voltge limiting as the capacitor across the device as mentioned in an earlier EU appliance comment.

 

[tersh]

Inductive kicks are not surges, are they?!

Yest they are.

The MOVs with clip the top off the inductive voltage kicks, by providing a low impedance path. Same type peak voltge limiting as the capacitor across the device as mentioned in an earlier EU appliance comment.

Ok.

Here is a big washing machine with big motor that can run for half a day without tripping the same GFCI once.

But a tiny shaded pole motor can trip it so repetitively.

What's in a shaded pole motor that can repetitively cause inductive kick? any clue anyone?

 

[ptonsparky]

Ok.

Here is a big washing machine with big motor that can run for half a day without tripping the same GFCI once.

But a tiny shaded pole motor can trip it so repetitively.

What's in a shaded pole motor that can repetitively cause inductive kick? any clue anyone?

The coil.

 

[tersh]

The coil.

An electric fan also has coil but no matter how many times it is turned on and off.. it doesn't trip the Leviton based GFCI, why?

 

[junkhound]

Different inductances.

BTW, the "Levitoncircuit' in yoaur original post is simly a copy of the Fairchild spec typical circuit.

Have you actually traced out the real circuit in the Leviton gfci? Differences are likely.

You (or Gar) could hook up a scope to pin1 of the Fairchild chip with different motors and see the different responses.

Also, in your photos, do not see a switch except on the outlet strip. I'm not clear on where the switch is, where is it in the circuit ?

 

[tersh]

Different inductances.

BTW, the "Levitoncircuit' in yoaur original post is simly a copy of the Fairchild spec typical circuit.

Have you actually traced out the real circuit in the Leviton gfci? Differences are likely.

You (or Gar) could hook up a scope to pin1 of the Fairchild chip with different motors and see the different responses.

Also, in your photos, do not see a switch except on the outlet strip. I'm not clear on where the switch is, where is it in the circuit ?

Here is the switch.

Note the GFCI is not made by Leviton. It just used the same chipset used by Leviton. It's not even the same chipset. That is. A china company copied the Fairchild chipset 100%. It has even same wordings in the manual. I don't know if it's under license or illegally. The Fairchild chipset copied is the N4141A (diagram and manual in the original post) produced about 2008. The Leviton video bellow detailed it.

It's made by this company in year 2007: Shanghai Fudan Microelectronics Group Co. Ltd

http://www.chip-memory.com/4-1-2-integrated-circuit.html

It's the only 240v GFCI with 5mA tripping model in the whole world. Distributed here OEM to many companies.

http://www.internationalconfig.com/icc6.asp?item=74900-RCDS

The 100% copied chipset is called FM2141 with diagram here:

Noticed it is 100% identical to the Fairchild N4141A:

In this Leviton video. You can see the above R4141A chipset exactly mentioned and even the second coil explained in detail.

The Shanghai company never copied other models. So all 240v GFCI receptacle users worldwide were stuck with their 2007 model.

In current Leviton GFCIs, it no longer trips on inductive kick right? They have fixed all the problems and it is only the old stock at homes which experienced it?

 

[al hildenbrand]

What's in a shaded pole motor that can repetitively cause inductive kick? any clue anyone?

The coil.

Exactly. The coil.

Examine the wire gauge and number of windings in the different types of motors your are raising in you question. You will find that the tiny shaded pole fractional horsepower motor that you have in your photos has a coil that has a large number of turns (windings) of very fine gauge wire.

That right there is what sets your little motor apart from the rest.

When the supply to the motor is switched off at an instant of maximum magnetic field density in the laminated core, the field will collapse. The collapsing magnetic field, moving by the coil windings, induces the "kick". Strength and speed of the collapsing magnetic field and the turns ratio of the windings in the coil do the rest.

 

[tersh]

Exactly. The coil.

Examine the wire gauge and number of windings in the different types of motors your are raising in you question. You will find that the tiny shaded pole fractional horsepower motor that you have in your photos has a coil that has a large number of turns (windings) of very fine gauge wire.

That right there is what sets your little motor apart from the rest.

When the supply to the motor is switched off at an instant of maximum magnetic field density in the laminated core, the field will collapse. The collapsing magnetic field, moving by the coil windings, induces the "kick". Strength and speed of the collapsing magnetic field and the turns ratio of the windings in the coil do the rest.

You have a point. The shaded pole motor has very strong magnetic field even a meter away, more than a 500va transformer. My EMF meter range is only up to 100mG. Does anyone know of the best quality EMF/flux meter that has range up to say 10 Gauss and accurate. I'd like to measure the leaking flux half inch away from the coil and compare it to other transformers.

 

[tersh]

Here is the switch.

Note the GFCI is not made by Leviton. It just used the same chipset used by Leviton. It's not even the same chipset. That is. A china company copied the Fairchild chipset 100%. It has even same wordings in the manual. I don't know if it's under license or illegally. The Fairchild chipset copied is the N4141A (diagram and manual in the original post) produced about 2008. The Leviton video bellow detailed it.

It's made by this company in year 2007: Shanghai Fudan Microelectronics Group Co. Ltd

http://www.chip-memory.com/4-1-2-integrated-circuit.html

It's the only 240v GFCI with 5mA tripping model in the whole world. Distributed here OEM to many companies.

http://www.internationalconfig.com/icc6.asp?item=74900-RCDS

The 100% copied chipset is called FM2141 with diagram here:

Noticed it is 100% identical to the Fairchild N4141A:

In this Leviton video. You can see the above R4141A chipset exactly mentioned and even the second coil explained in detail.

The Shanghai company never copied other models. So all 240v GFCI receptacle users worldwide were stuck with their 2007 model.

In current Leviton GFCIs, it no longer trips on inductive kick right? They have fixed all the problems and it is only the old stock at homes which experienced it?

In summary. The only 240v GFCI model in the world is a perfect clone of the Leviton Fairchild chip R4141A described in the youtube video above. In the overall circuit, what changed are only values of the rectifiers to make it 240v instead of 120v. The shanghai company may have licensed it because all the manual wordings are exactly the same as the Fairchild. And of course they won't change other circuits since they just copied it from Fairchild. Also Fairchild never make any 240v version of their GFCI receptacles. So maybe they gave license to Shanghai Fudan to produce 220v-240v versions of GFCI with 5mA tripping? GFCI is only a USA label. But somehow the china company was able to produce 240v version of it with 5mA tripping. I am able to verify the tripping is 5mA or so with GFCI tester machine.


Whatever, it inherits the Leviton tripping problem. This is zoom of the Shanghai Fudan FM2141 circuit itself:

This is zoom showing the chipset marked FM2141 by Shanghai Fudun Microelectronics Co, Ltd that is completely identical to the Fairchild R4141A. Do you think it's license agreement or illegally copied?

This is the other side (opposite surface) of the PCB above.

Any electronics specialist here with software that can track all the components in the PCB circuit and create circuit diagram out of it?

Here's proof showing the Fairchild R4141A and Shanghai Fudan FM2141 is similar even in the wordings in the data sheet. This is Fairchild R4141A data:

Full datasheet at http://datasheet.elcodis.com/pdf2/81/73/817329/rv4141a.pdf

This is Shanghai Fudan FM2141 (in my 240v GFCI) description identical to the Fairchild R4141A above:

Full datasheet at http://datasheet.elcodis.com/pdf2/90/42/904285/fm2141.pdf

This is in addition to the identical circuit diagrams of the Fairchild R4141A and Shanghai Fudan FM2141.

This is just to show that the FM2141 inherits all the tripping nuisance problems of the Fairchild R4141A used by Leviton.

How about latest models of Leviton GFCIs that used the auto-monitoring self test chipsets. Does it have the same nuisance tripping problem? Or solved? We don't have any 240v GFCI with auto-monitoring self test versions.

 

[LarryFine]

Any electronics specialist here with software that can track all the components in the PCB circuit and create circuit diagram out of it?

I could do that; it's not difficult. I have designed PC boards from schematics.

 

[gar]

189198-1035 EST

tersh:

You need an air core coil of wire and an AC voltmeter.

I have a calibrated coil that is 20 mV/gauss at 60 Hz 1.3" in diameter. Double the frequency and the voltage doubles as you know from df/dt. I also have an uncalibrated coil of 7000 turns of #43 wire about 1.4" square.

About 18" from a small Sola constant voltage transformer I read 0.4 mV from the calibrated coil and about 3 mV from the 7000 turns. My guess is the calibrated coil is in the 1000 turn range.

The near field flux is not likely your problem. The e = N*df/dt from opening the coil with current flowing and the transient voltage placed on the power wire is the likely cause of your noise triggering. Note that closing a switch can actually including opening while closing. Contact bounce might be a description.

.

 

[tersh]

I could do that; it's not difficult. I have designed PC boards from schematics.

Some resistor values not visible in the picture (like C5, C6). Btw.. Shanghai Fudan only cloned the Fairchild R4141A chipset used by Leviton with inductive kick tripping problem. Shanghai Fudan didn't build the entire circuit. It's made by an unknown OEM company. Of course the unknown china company should copy the circuit diagram 100% (except the rectifier for 240v instead of 120v). Notice the PCB is dirty, unlike the Leviton PCB in the youtube video shared earlier. They won't modify a thing to avoid unwanted problems. So I guess the following diagram should match the actual PCB.

The circuit diagram in the Shanghai Fudan datasheet is more blurred. The one in the Fairchild is clearer, the circuit diagram of both is 100% identical:

 

[GoldDigger]

A chip with the same specifications, designed to be pin compatible in the same circuit, will not necessarily use the same internal circuitry (reverse engineered or licensed). It may instead be freshly designed as a pin compatible replacement and so have its own totally independent set of quirks.

 

[tersh]

189198-1035 EST

tersh:

You need an air core coil of wire and an AC voltmeter.

I have a calibrated coil that is 20 mV/gauss at 60 Hz 1.3" in diameter. Double the frequency and the voltage doubles as you know from df/dt. I also have an uncalibrated coil of 7000 turns of #43 wire about 1.4" square.

About 18" from a small Sola constant voltage transformer I read 0.4 mV from the calibrated coil and about 3 mV from the 7000 turns. My guess is the calibrated coil is in the 1000 turn range.

The near field flux is not likely your problem. The e = N*df/dt from opening the coil with current flowing and the transient voltage placed on the power wire is the likely cause of your noise triggering. Note that closing a switch can actually including opening while closing. Contact bounce might be a description.

.

What kind of EMF or flux meter are you using? I only have something similar to this for the past 25 years.

It has range of up to 100mG only. I need something accurate to 10 Gauss or more.

What is the highest gauss can you read when you use an EMF/flux meter a few millimeter away from the coil or transformer laminated cover?

There are many Gaussmeters at amazon and I need to order one that is most accurate and professional with accuracy at high gauss or flux.

I'm interested in all this to get the hang of Maxwell equations because ultimately my interest is understanding physics beyond the standard model, and figuring it out.

 

[tersh]

A chip with the same specifications, designed to be pin compatible in the same circuit, will not necessarily use the same internal circuitry (reverse engineered or licensed). It may instead be freshly designed as a pin compatible replacement and so have its own totally independent set of quirks.

You mean this regularly happens and not illegal?

So Shanghai Fundan Microelectronics Ltd Co can just manufacture a clone of the Fairchild R4141A and create pin compatible chip to the same circuit without even talking to Fairchild and the latter not able to sue the former?

For example, if a company create identical Intel Core processor. It's not illegal and Intel can't sue them?

If so, then it's unlikely Shanghai Fundan would pay Fairchild huge sum of money just to license it. And remember Trump hates china precisely because of all widespread intellectual property infringement.

Also the Fairchild R4141A chip internal circuitry is never shared publicly and a company secret? Or can they share it at public domain after certain years?

 

[tersh]

I have something to ask about the GFCI with auto-monitoring self test required in UL 2015.

1. This is the Fairchild version with the additional chip and circuitry for the auto-monitoring self test.

The Fairchild FAN41501 is the additional chip for the self test. Data sheet at: http://www.mouser.com/ds/2/149/FAN4149-365238.pdf

First question.

1. Leviton used the chip for their auto-monitoring self test GFCI models required in UL 2015. Does it inherit the same inductive kick nuisance tripping as the older N4141A chip brother?

2. The Fairchild FAN41501 auto-monitoring self test chipset seems quite sophisticated. Theoretically what components can't it check or miss? Only missing power to the solenoid coil? What is the usual components that can break down in relays. One the FAN41501 Chipset can miss.

​

http://www.mouser.com/ds/2/149/FAN41501-348130.pdf

"Periodic Functional Testing for Key GFIC Components: GFCI Controller, Solenoid, Sense Transformer, And Silicon-Controlled Rectifier (SCR)"


Is this the same chipset used in Siemens, Leviton and all major GFCI makers or do they have their own chipset design?


3. Major hotels in the Philippines like Holiday Inn installed all Meiji GFCIs (with the Shanghai Fundan FM2141 chip without auto-monitoring self test.. my GFCI teardown is that of Meiji brand) in hundreds of their receptacles (see the url below). Meiji widely advertised it as having auto-monitoring self-test. Just want to know what would happen if 8 years later, some didn't function and someone got electrocuted. Could that person or the hotel sue Meiji for advertising it as having auto-maintain self-test when it didn't? What do you think? If you were the vendor or electrician who sell or installed it. I guess only the Meiji company is liable for false advertisement?

​

https://meijielectric.ph/blog/resort...c-gfci-mgf250/​

 

[winnie]

I am pondering the following:

The GFCI chip is powered by connection to the AC line through a resistor, a diode, and the 'trip solenoid' (which I presume is supposed to do double duty as a filter choke). The capacitor from pin 6 is a decoupling capacitor to filter the rectified AC.

The circuit has a residual current detection circuit that uses an op-amp connected to a sense coil, using standard op-amp feedback through Rset. The output of this amplifier connects to a pair of comparitors so that when the output of the RCD is high enough the system is triggered.

Also connected to the _output_ of the op-amp is the 'grounded neutral detection' circuit. The way this works is that if the neutral is grounded upstream of the GFCI, the combination of the ground-neutral loop, the various capacitors and the second toroid creates a feedback path that makes the sense op-amp oscillate and drive a detectable current through the ground-neutral loop.

I am wondering if particular inductive loads can interact with the grounded neutral detection, and dump current into pin 1 via the power supply connection, and trip the circuit not by creating an actual residual current, but rather by creating a current spike at pin 1 of the chip.

Tersh, if you want to continue to research this, I think that the next step would be to build a proper spice model of the circuit, and simulate what happens with various inductive loads, _or_ probe the circuit using a good oscilloscope. You might also hire someone to do this research.

Very, very interesting, since you have clearly demonstrated a case where there is no 'residual current', but you are consistently tripping this circuit.

-Jon

 

[tersh]

I am pondering the following:

The GFCI chip is powered by connection to the AC line through a resistor, a diode, and the 'trip solenoid' (which I presume is supposed to do double duty as a filter choke). The capacitor from pin 6 is a decoupling capacitor to filter the rectified AC.

The circuit has a residual current detection circuit that uses an op-amp connected to a sense coil, using standard op-amp feedback through Rset. The output of this amplifier connects to a pair of comparitors so that when the output of the RCD is high enough the system is triggered.

Also connected to the _output_ of the op-amp is the 'grounded neutral detection' circuit. The way this works is that if the neutral is grounded upstream of the GFCI, the combination of the ground-neutral loop, the various capacitors and the second toroid creates a feedback path that makes the sense op-amp oscillate and drive a detectable current through the ground-neutral loop.

I am wondering if particular inductive loads can interact with the grounded neutral detection, and dump current into pin 1 via the power supply connection, and trip the circuit not by creating an actual residual current, but rather by creating a current spike at pin 1 of the chip.

Tersh, if you want to continue to research this, I think that the next step would be to build a proper spice model of the circuit, and simulate what happens with various inductive loads, _or_ probe the circuit using a good oscilloscope. You might also hire someone to do this research.

Very, very interesting, since you have clearly demonstrated a case where there is no 'residual current', but you are consistently tripping this circuit.

-Jon

It was over 25 years ago when I handled oscilloscopes. We used this a lot in our electronic classes. So the surges from the shaded pole motors can show up in the oscilloscope? I'd like to know how it looks like. What kind of oscilloscope must I buy? There are some in the local electronics shop. At least I can review my lessons 25 years ago. Lol.

This is especially so when I'd be stuck with the FM2141 based chip GFCIs for the next 25 years as this is the only 240v GFCI in the world. And the only GFCI outlet model in our country.