VFD started tripping RCD

[ETforInd]

Hey Guys,

I'm troubleshooting a steel cable pusher and need some advice.

The 2.2kW motor is driven by a Lenze I550 VFD. It is powered via a 10a plug, as it is used in different locations. (Tripping RCD on different Locations)

The pusher was running fine for half a year it started tripping the rcd when either plugging it in or when the motor got too much mechanical pressure back. ( If I shut the lead off, plug in, and turn the lead live it's fine and will then only trip when too much mechanical pressure)

So far I disconnected the motor but rcd still tripping. I disconnected the Main Earth to determine how much voltage the VFD got against Earth and It's over 40V.

If I attach a second lead to increase the resistance against Earth rcd won't trip. Earth connections seem to be all right.

Was thinking EMF may be faulty? If so how can I determine that?


Thanks in advance for your input!!!

 

[KlausST]

Hi,

indeed I think this is a bit complicated.
(And to me it sounds as if there is something defective)

There is the HF current that will go via EARTH ...
And there will be mains frequency current.

The HF current may be caused by output filters, motor cables (especially shielded ones) and EMI reduction capacitors.

The best I can recommend is to put a 100 Ohms 0207 resistor in series with the EARTH line.
SAFETY FIRST: ALWAYS expect high voltage at any conductive part. Don´t touch directly. Don´t let any other people touch it.
In worst case the resistor may get hot, overheat or even explode !!! Put it on a non flammable surface.
Then try to switch ON the VFD.
If the RCD does not trip: use a voltmeter (AC, true RMS) to measure the voltage across the resistor.
If the voltage is above 5V --> Switch it off immediately I expect a technical failure inside. It needs to be repaired
If below 5V: you could use a scope to see the waveform and frequencies. Show us the scope picture (with all necessary informations)

Klaus

 

[cupoftea]

Could be a faulty Y capacitor in the VFD. Maybe its gone a little short, though they should fail open...but thats assuming the makers used a "proper" y capacitor, and not some cheap alternative.

Of course, maybe they used a too big value Y capacitor.
Maybe they didnt use a common mode choke, or not enough common mode choking inductance....that would pertain to high frequency currents flowing to earth though.

Maybe some or many of the installations where you work have other devices where they actually have too much y capacitance inside them...and your VFD is then adding to that and tripping out.......so that wouldnt necessarily be a problem with your vfd...but a problem with other peoples kit.

I lived in a flat block where the rcd kept tripping, and an electrician had to go through everybodys room to see who had the dodgy thing with too much y capacitance (or an insulation fault to earth) that was causing the overall earth current to be high anyway, and then often it tripped because it was so near the tripping threshold........

Also, could be the insulation from one of the power semiconductors has overheated or something and so is breaching its insulation to its earthed heatsink.

Something in that VFD is connecting, either directly , or capacitively, to the earth somehow.

But also, as discussed, insulation faults to earth are commonplace...so it might not actually be your VFD thats at fault....maybe the earth current in many installations is already well high, and then when you plug your thing in it just goes over that threshold....and trips the rcd.

Also, some places use cheap RCDs which trip at very low earth current....hooky RCDs from some cheap supplier.
Its no surprise you get dodgy RCDs because they are so expensive (at least the proper standards regulated ones are)...so some places skip out and use cheap ones that are trash.

 

[Easy peasy]

arching / tracking on the pcb - usually around the IGBT's or rectified mains to earthed heatsink or bracket - or other earth point . . .

 

[cupoftea]

Even with good RCDs and good VFDs...you can still trip an RCD if just too many items containing y capacitors are plugged into the phase overall.

I think somehow you are going to have to (like Klaus said) measure that earth current...so a sense resistor sounds good.

 

[FvM]

If I attach a second lead to increase the resistance against Earth rcd won't trip.

Sorry, I don't understand what you are doing here, please explain. How can an additional wire increase resistance?

My general thoughts about VFD tripping RCD: RCD is usually not sensitive to high frequent currents. Motor cable should always include a low inductance earth connection between motor chassis and VFD earth ground, preferably use a shielded cable with respective EMC cable glands at both ends. This way PWM frequent VFD output can't spread into mains power supply.

Most common RCD problem with VFD is combination of too sensitive RCD with large Y capacitors.

 

[cupoftea]

Most common RCD problem with VFD is combination of too sensitive RCD with large Y capacitors.

Very true, and its amazing because there is no way of knowing how much y capacitance is connected to any mains phase at any particular time.....with all the kit plugged into the mains. Nobody knows.....it could be well high earth current before you plug your vfd in....then you plug it in and the overall earth current has gone above 15 to 30mA or whatever is the threshold.

I worked in one place where they had made their own simple open loop fluxgate RCD's for a couple of quid in cheap parts.....(the "proper ones are £80 plus) .they installed hundreds of thousands of them and they were false tripping left right and centre.......we had literally tens of thousands of returns.

--- Updated Jan 12, 2025 ---

So it could also just be noise from the VFD....going down the mains cable and noise tripping the RCD circuitry.

So you could all told..

Measure earth current
Add Klaus's 100R resistor in series with the earth wire.
Measure Y capacitance at the plug from L-E and N-E
Wrap the earth wire round a torroid
Clasp a ferrite clamp round L,N of the VFD.
Even just a biggish X2 cap from L to N input wires can sometimes work. (as near to the vfd mains input terminals as poss)
Buy an RCD and power it through that and see if it trips. (on a mains cct where theres no other kit plugged in)

Others will think of more

 

[ETforInd]

Hi,

indeed I think this is a bit complicated.
(And to me it sounds as if there is something defective)

There is the HF current that will go via EARTH ...
And there will be mains frequency current.

The HF current may be caused by output filters, motor cables (especially shielded ones) and EMI reduction capacitors.

The best I can recommend is to put a 100 Ohms 0207 resistor in series with the EARTH line.
SAFETY FIRST: ALWAYS expect high voltage at any conductive part. Don´t touch directly. Don´t let any other people touch it.
In worst case the resistor may get hot, overheat or even explode !!! Put it on a non flammable surface.
Then try to switch ON the VFD.
If the RCD does not trip: use a voltmeter (AC, true RMS) to measure the voltage across the resistor.
If the voltage is above 5V --> Switch it off immediately I expect a technical failure inside. It needs to be repaired
If below 5V: you could use a scope to see the waveform and frequencies. Show us the scope picture (with all necessary informations)

Klaus

Hey Klaus,

Thanks for your answer.

Unfortunately, I don't have a scope. Would you still recommend checking the voltage on a 100ohm resistor?

Thanks, Jan

 

[KlausST]

Doing a VFD repair without scope? --> I recommend to lend one. Does not need to be fancy one. Even an USB scope would do.

Checking the voltage just with a meter ... does not give any information about the frequencies. (It does not hurt either)
And as written in post#2 .. I´d first find out whether the current is caused by high or low frequency.

Klaus

 

[FvM]

I think the only safe way to measure earth current is a current clamp/current transformer or under circumstances a low ohmic shunt keeping the earth wire cross section. Interrupting earth line is against all safety rules.

 

[KlausST]

I think the only safe way to measure earth current is a current clamp/current transformer or under circumstances a low ohmic shunt keeping the earth wire cross section. Interrupting earth line is against all safety rules.

I absolutely agree. Safety first.

But during development and repair in a laboratory .. sometimes there is no other option.

For example: How useful is a current clamp ... while the DUT can not be switched ON because it immediately trips the RCD?

The resitor has two jobs:
* to limit the current in a way the RCD does not trip
* to give a voltage that can be measured with a scope.

What value of resistor to use .. will depend. Is it too low: the RCD still will trip. Is it too high, the mesurement may be useless.

If the current is too high for the 100R .. then one may use a higher value resisitor .. but one needs to consider power dissipation.

In the end: cutting the EARTH wire permanently is a no go.
The person in the laboratory needs to be aware of the risks.

Klaus

 

[cupoftea]

If you dont have a scope...then you are limited.
Here is one thing you can do, but as FvM and Klaus say, be very very careful.
Dont leave it lying around with the earth lead cut.
Wear insulating gloves when you touch the VFD (ok i wouldnt bother but i must advise this)

Cut the earth wire, put a switch on it, so you can switch the earth wire in and out.......then you can see if it still trips if the earth wire is cut.

Earth is though, sometimes needed to make the vfd immune to noise....so you may do it an alternative way...ie, when you switch the earth out, have a 10n film cap in parallel with a say 10k resistor across the switch....so there is still a kind of earth connection to the vfd even though you have broken the earth wire connection.

Earth is needed to earth the enclosure in case somehow the electrical insides touch the earthed casing...so be massive careful with this.

Also, you could make up a jig which is just a mains plug with a Y capacitor from L-E and from N-E.....choose say 3n3 for both of them.
Plug this in and see if the RCD trips on a phase where your vfd is tripping.

You could make a couple, and plug 'em all in, and see when the rcd eventually trips.

 

[FvM]

It's not yet clear if RCD tripping is caused by VFD fault, inadequate wiring or unsuitable RCD. The reported results aren't yet consistent, please consider my question from post #6 about "additional earth wire".

 

[ETforInd]

Gonna try to measure frequency later via multimeter.
I took the VFD apart in the meantime to check the board all Capacitors seem to be alright and no damage visible.
So still can't determine where the 40V is coming from when the motor is disconnected.

Would the Y-Capacitor be just behind the input of VFD? I got L-N 1.90µF PE-L 376nF PE-N 326 nF

Update:

Got a second Machine which has the same problem with tripping RCD after running fine for months. Also here disconnected earth from VFD 35V VFD -> Earth with the Motor not running. (The motor runs fine when Earth from VFD disconnected, motor Earth still connected)

35V from VFD got 50hz so I guess main leaking somewhere internal on the PCB towards earth. Any idea where to look specific?

I attached pictures of the PCB.

Thanks
Jan

Doing a VFD repair without scope? --> I recommend to lend one. Does not need to be fancy one. Even an USB scope would do.

Checking the voltage just with a meter ... does not give any information about the frequencies. (It does not hurt either)
And as written in post#2 .. I´d first find out whether the current is caused by high or low frequency.

Klaus

 

[ETforInd]

Could be a faulty Y capacitor in the VFD. Maybe its gone a little short, though they should fail open...but thats assuming the makers used a "proper" y capacitor, and not some cheap alternative.

Of course, maybe they used a too big value Y capacitor.
Maybe they didnt use a common mode choke, or not enough common mode choking inductance....that would pertain to high frequency currents flowing to earth though.

Maybe some or many of the installations where you work have other devices where they actually have too much y capacitance inside them...and your VFD is then adding to that and tripping out.......so that wouldnt necessarily be a problem with your vfd...but a problem with other peoples kit.

I lived in a flat block where the rcd kept tripping, and an electrician had to go through everybodys room to see who had the dodgy thing with too much y capacitance (or an insulation fault to earth) that was causing the overall earth current to be high anyway, and then often it tripped because it was so near the tripping threshold........

Also, could be the insulation from one of the power semiconductors has overheated or something and so is breaching its insulation to its earthed heatsink.

Something in that VFD is connecting, either directly , or capacitively, to the earth somehow.

But also, as discussed, insulation faults to earth are commonplace...so it might not actually be your VFD thats at fault....maybe the earth current in many installations is already well high, and then when you plug your thing in it just goes over that threshold....and trips the rcd.

Also, some places use cheap RCDs which trip at very low earth current....hooky RCDs from some cheap supplier.
Its no surprise you get dodgy RCDs because they are so expensive (at least the proper standards regulated ones are)...so some places skip out and use cheap ones that are trash.

Thanks for your response,

What would be a '"Proper" y capacitor dimensioning for a 2.2Kw motor? I got on the Plug L-N 1.90µF PE-L 376nF PE-N 326 nF

Was running fine for months and two machines are start tripping now. They do not get treated as well and get moved quiet a bit around what might cause damage inside the VFD that's why I took it apart but couldn't determine any visual damage.

Only the VFD is on the circuit so won't be due to much y capacitance + Motor is disconnected and not running.

Could be a semiconductor do you have an idea where to look specifically on the PCB? I added some pictures in my response to Klaus.

 

[ETforInd]

arching / tracking on the pcb - usually around the IGBT's or rectified mains to earthed heatsink or bracket - or other earth point . . .

Thanks for your response,

It's 35V 50Hz leaking so could be one of the things.

 

[ETforInd]

Sorry, I don't understand what you are doing here, please explain. How can an additional wire increase resistance?

My general thoughts about VFD tripping RCD: RCD is usually not sensitive to high frequent currents. Motor cable should always include a low inductance earth connection between motor chassis and VFD earth ground, preferably use a shielded cable with respective EMC cable glands at both ends. This way PWM frequent VFD output can't spread into mains power supply.

Most common RCD problem with VFD is combination of too sensitive RCD with large Y capacitors.

Thanks for your response,

An additional 30m of cable will increase the resistance of the earth error loop,

I got 35V 50hz at the VFD with a disconnected motor, which trips the RCD as soon as I power the VFD.

It was running fine for months on various sites.

 

[cupoftea]

which trips the RCD as soon as I power the VFD.

Maybe the RCD is getting corrupted by the inrush, which is a "Loud" electromagnetic event, and happens when it is powered up.
So i am thinking. ..can you have the VFD on no load, and disconnected from motor, and then power it up without the inrush relays in circuit. Or maybe you can set up an external inrush resistor.......then the shorting relays wont be able to short that out.

Wear eye protection as inrush resistors can blow up.

What would be a '"Proper" y capacitor dimensioning for a 2.2Kw motor? I got on the Plug L-N 1.90µF PE-L 376nF PE-N 326 nF

300+nF sounds very unusually high indeed for a y capacitor on the mains....but you are saying the tripping happens when the motor isnt even connected.

 

[ETforInd]

Maybe the RCD is getting corrupted by the inrush, which is a "Loud" electromagnetic event, and happens when it is powered up.
So i am thinking. ..can you have the VFD on no load, and disconnected from motor, and then power it up without the inrush relays in circuit. Or maybe you can set up an external inrush resistor.......then the shorting relays wont be able to short that out.

Wear eye protection as inrush resistors can blow up.


300+nF sounds very unusually high indeed for a y capacitor on the mains....but you are saying the tripping happens when the motor isnt even connected.

Motor Connected or not I'm having the 35V at the VFD, not only in the startup phase.

Motor runs fine with Disconnected VFD-earth for testing.

How would an external inrush resistor help me with the Earthleak?

 

[cupoftea]

How would an external inrush resistor help me with the Earthleak?

The high dv/dt of the inrush could cause emission through stray capacitance and to earth like that.
Or just the general "electromagnetic loudness" of inrush could have gotten coupled into the mains wiring and then noise_bombed the electronics in the RCD....making it trip spuriously.

--- Updated Feb 11, 2025 ---

So you are saying you have significant 50Hz ripple on the VFD input..?
This 35v presumably comes from that switch mode transformer with the yellow tape round it.
So now i am wondering if it is overheating and has insulation damage inside it...

Taking it out to hipot test it doesnt sound like a great idea though.

I suppose the other way would be to get a separate isolated 35V supply and solder wire that in after the yellow transformer...having disconnected the yellow transformer from its ouput ...eg desolder the diodes what have you.

It does look a very small transformer for the power you mention. I dont see a PFC inductor either on the board you show.