VFD CONTROLLED MOTORS WITH GFCI

[ramsy]

How do you know that?

I know notzing!!

There is not much difference, other than the pickup point, between an RCD and a GFCI device.

There are several types of RCD, with plenty of differences:
Notice Type F is specifically designed for multi-frequency appliance / loads

Tesla's wall connector was the first to use 3-Phase Type B RCD, since no such xFCI junk did the job.

Electric Vehicle Charging Equipment is UL category FFWA
Tesla's equipment file is shown on Product IQ

 

[ramsy]

have never seen VFD capacitors with enough 'charge' to drive the electronics, much less operate the drive.

When I disconnect VSD's, the keypad display continues to operate, and record error codes, for several seconds.

 

[jim dungar]

When I disconnect VSD's, the keypad display continues to operate, and record error codes, for several seconds.

Yes, but the drive does not continue to power the load. The electronics are driven off of their own power supply with capacitors, not necessarily the same ones that power the load.

 

[jim dungar]

I know notzing!!

There are several types of RCD, with plenty of differences:
Notice Type F is specifically designed for multi-frequency appliance / loads

Tesla's wall connector was the first to use 3-Phase Type B RCD, since no such xFCI junk did the job.

Electric Vehicle Charging Equipment is UL category FFWA
Tesla's equipment file is shown on Product IQ

There are also GFCIs which are designed for high frequency loads. RCD and GFCI/GFPE are very close to each other both in design and basic function.

People like to think RCD are vastly different than GFCI because they use a different pickup level and time current curve.

I don't know of a single manufacturer that has tried to get US codes and standards changed to accept their RCD design instead of the GFCI/GFPE product they also make.

 

[garbo]

Retired a few years but all of the Danfoss & ABB VFD'S had built in ground fault protection. Danfoss called it earth fault when it tripped out. Probably over 90% of the Danfoss earth fault trips were false. Would make sure motor was at a compete stop then megger 480 motors with a 1,000 volt megger and restart. I'm all for ground fault protection for homes and even outdoor AC units but over 99% of the motors I wired up, trouble shoot or rep!aced were on a grounded metal plate and had copper ground wire to the motor and at my last jobs all runs from drives to motors were in steel EMT & a short section of Greenfield or sealtite with a metal core. If a motor was to ground out my thinking is it would not pose a large shock hazard. If the VFD had a bypass then an expensive three phase GFCI would be needed. I asked several drive techs how many milliamps to ground will trip out their drives but none knew. If drives have built in ground fault protection why would an expensive three phase GFCI have to be used ?

 

[jim dungar]

If drives have built in ground fault protection why would an expensive three phase GFCI have to be used ?

The NEC requires GFCI protection on the input. The GF protection provided by the VFD is always on the output.

 

[synchro]

Without the SPGFCI it doesn’t trip. The Vfd run perfectly without the GF device and will come up to speed fully 60hz

A suggestion is to disconnect the wires from the output terminals of the VFD, and then see if the VFD will run and ramp up to 60 Hz without the SPGFCI tripping. If it stops tripping, that shows that some of the VFD's common mode noise current from driving the capacitance of the motor and wiring is passing though the input rectifier diodes when they are conducting and causing the SPGFCI to trip. If the SPGFCI still trips when the wires at the VFD output are disconnected, that shows the noise current originating inside of the VFD is enough to trip the SPGFCI. And therefore any reactors or filters on the output of the VFD are not likely to prevent the SPGFCI from tripping.

 

[winnie]

Yes, but the drive does not continue to power the load. The electronics are driven off of their own power supply with capacitors, not necessarily the same ones that power the load.

I've used a couple of VFDs where the control electronics was powered by a DCC converter connected to the DC bus. The electronics will remain powered for a period of time as the DC bus voltage decays.

The DC bus does have significant capacitance, but compared to the usual load the energy stored is _tiny_. For example a 100kW drive might have 5000uF of bus capacitance. At a DC bus of 700V, this is perhaps 1300J of stored energy, but less than a cycle of operation if the drive is operating at 100kW output.

In the OP's case, if the drive is running a pump at low speed, (say the very beginning of a ramp up), then the power going to the actual load motor is pretty low and I'd expect the DC bus voltage to take several seconds to decay. So it is _plausible_ to me that the GFCI trips at the instant that the VFD starts to run the motor, but there is a delay before the drive notices that it doesn't have input power.



-Jonathan