More GFCI trouble!

[MikeWi]

I have a customer (only one thankfully) where the jumper trick for the Delta MS300 VFD stopped being enough to prevent his GFCI from tripping.
In talking to Delta, I was given a spec required for a GFCI to be used with it:
select a current sensor with sensitivity of 200mA or above, and the operation time should not be less than .1 second to avoid nuisance tripping.

Not sure this even exists at an affordable price. Even the hospital grade Leviton that has been recommended before doesn't meet that. In fact, Leviton says they don't have anything like that.

Anyone have recommendations?

 

[markba633csi]

Do you have to use a GFCI? Is this installation in a garage or shop where the VFD could be hardwired in?

 

[MikeWi]

in a garage. Hardwiring is an idea, but from what I've heard, I'm not sure it's an option for him.

 

[mksj]

Hardwire is the primary option. NEC code change now requires GFCI for both 120 and 240VAC circuits in many locations including the garage/basements. I mostly have seen a problem with 120VAC GFCI's and VFD's, almost guaranteed to trip. Options vary such as using an isolation transformer, some EMC/RFI low leakage filters just before the VFD may help, it should be specific for VFD's. I am not aware of any GFCI's for residential use that have a higher trip current of more than ~5mA, there are expensive commercial/industrial GFCI's for use with VFD's, but if I recall they trip at around 30mA but some are adjustable.

GFCI breaker with Hitachi VFD?

Searched the forum and didn't see anything so here's the question. I am wiring up my VFD for my shiny new 1236T, and I have to wire in a new 30a circuit into my shop subpanel. Should I use a regular breaker or GFCI? Hitachi manual lists both as possibilities, but as expected says "check...

www.hobby-machinist.com

If 120VAC GFCI, the one below (Leviton GFPL2-PLT) was recommended to work better with VFD's, I find the normal light duty ones will trip no matter what and also seem to fail over time. The industrial grade/heavy duty do better in my experience.

https://www.amazon.com/dp/B019CK9NTW/

 

[MikeWi]

I emailed Leviton because the Leviton GFNT2-HGW that had been recommended before did not work. They said that they do not make a GFCI with those specs, which seems suspect. Maybe they just meant they don't have a class A that will work which would go without saying. Not very helpful.

 

[mksj]

GFCI's for residential use must meet class A requirements with a trip current of between 4-6mA, anything other (i.e. industrial, commercial or VFD specific) would not be to code, and in that case eliminating the GFCI would be pretty much the same thing. Other forms of mitigation may be required as mentioned if they cannot hardwired the power as opposed to using a plug. Having the fault current at 200mA is not an option for a residential GFCI, as it does not comply with the class A requirement. If this is a 120VAC application, they are almost guaranteed to trip the GFCI, a few manufactures like KB Electronics do have VFD's that may be compatible with GFCI's but also depends on the wiring install, etc.

 

[B2]

@MikeWi

Code was not your question, however, Municipalities commonly do not adapt or accept all codes from the NEC standards book. These books describe codes and the local municipalities, especially big ones with resources to do so select sections, avoid some, and sometimes write their own. So just because it is in the book does not mean you have to, or should, follow it.

I would be curious to know what is causing the GFI to trip. If you want to diagnose what is going on you could measure the voltage between the ground and neutral at the lathe. If you want to know the current flowing you might wire in a fuse between the ground and the neutral and observe the voltage drop, DC and AC. It there are spikes (AC) you will need a scope. Most simple current meters do not provide mush information regarding fast events. So, as a rule of thumb metal bus fuses typically burn out, when there is about 0.1 volt drop across them. (This is my rule of thumb, from DC measurements I made ages ago when I was designing power supplies.) Ideally there is no current in the ground line so there should be no voltage between them at the appliance. At the breaker box there will be as current is flowing in the neutral. If there are other appliances in the circuit they could confuse the values so turn them all off and do the measurement at the VFD end of the circuit, not the breaker box. So a 10Amp fuse showing a 0.1 volt drop would imply , very roughly, a 10Amp leak. Which would be very dramatic anyway. So if you think there is a 100ma leak (0.1 amp) you might want to use a 100ma fuse, or a little larger, so that it does not blow out to often and yet there is a nice size signal.

So as an aside, a fuse can be a very handy small resistor for measurements of current. Just measure the voltage across one to determine (roughly) the current flow in a wire. Of course, the resistivity of a metal as it gets hot enough to melt is not linear and so the measurement is only an approximation. If you have a four point probe impedance meter you can measure the resistance of what would normally be considered almost a short, or a very low, resistance.

Dave L.

 

[Reddinr]

The cynic in my thinks the CODE will not be happy until all electricity is removed from all residences in order to be 100% electrically safe.
The other cynic in me thinks that the CODE committees are only happy when forcing the sale of more expensive breakers etc.

Our local codes are usually at least one code version behind the latest one.

 

[den-den]

@MikeWi

Code was not your question, however, Municipalities commonly do not adapt or accept all codes from the NEC standards book. These books describe codes and the local municipalities, especially big ones with resources to do so select sections, avoid some, and sometimes write their own. So just because it is in the book does not mean you have to, or should, follow it.

I would be curious to know what is causing the GFI to trip. If you want to diagnose what is going on you could measure the voltage between the ground and neutral at the lathe. If you want to know the current flowing you might wire in a fuse between the ground and the neutral and observe the voltage drop, DC and AC. It there are spikes (AC) you will need a scope. Most simple current meters do not provide mush information regarding fast events. So, as a rule of thumb metal bus fuses typically burn out, when there is about 0.1 volt drop across them. (This is my rule of thumb, from DC measurements I made ages ago when I was designing power supplies.) Ideally there is no current in the ground line so there should be no voltage between them at the appliance. At the breaker box there will be as current is flowing in the neutral. If there are other appliances in the circuit they could confuse the values so turn them all off and do the measurement at the VFD end of the circuit, not the breaker box. So a 10Amp fuse showing a 0.1 volt drop would imply , very roughly, a 10Amp leak. Which would be very dramatic anyway. So if you think there is a 100ma leak (0.1 amp) you might want to use a 100ma fuse, or a little larger, so that it does not blow out to often and yet there is a nice size signal.

So as an aside, a fuse can be a very handy small resistor for measurements of current. Just measure the voltage across one to determine (roughly) the current flow in a wire. Of course, the resistivity of a metal as it gets hot enough to melt is not linear and so the measurement is only an approximation. If you have a four point probe impedance meter you can measure the resistance of what would normally be considered almost a short, or a very low, resistance.

Dave L.

I can't claim to "know" but suspect that the trips are caused by charging the large capacitors in a VFD. GFCIs compare the current in the hot line to the current in the neutral and assume any difference is a fault. When the VFD is energized (not when it tries to start a motor) the capacitors store quite a bit of energy which looks like a fault to the GFCI.

 

[Philzy]

Hey everyone. I’m the customer Mike is dealing with on this problem and coming here was good thinking. I checked with Mike before posting this. I didn’t want to step on his toes. Mike’s been a great help.

I purchased a PM-1130V back in October of ’21. Being new to VFDs I quickly learned they don’t play well with GFCI circuits. My new lathe kept tripping the GFCI outlet. After doing a Google search I tried the usual “fixes”. No joy. I called PM and spoke with Mike. Mike did a little research on the Delta MS300 VFD and found that if a jumper was removed my issue should be fixed. He was right. I pulled the jumper and the lathe worked perfectly for two years. I mean perfect. It never failed to start or run and never again tripped the GFCI outlet. Until………

The first week of this past October I went to fire up the lathe. I powered it up and heard the familiar cooling fan on the motor spinning up. Turned the direction switch to forward and hit the start button. As soon as I hit the button….“pop”. I tripped the GFCI. I reset the outlet, fired it back up, hit the start button and …“pop”. That’s not good. I unplugged everything on the circuit and tried again. Pop. I replaced the GFCI outlet and it still popped.

I next ran an extension cord into house and plugged into a different circuit and tripped the GFCI outlet there too. I then ran the cord to a circuit in my living room that did not have a GFCI outlet but it did have a GFCI circuit breaker in the power panel. I fired up the lathe and it ran. I cranked the speed from the slowest the fastest speed with no problem. When I shut off the motor the GFCI circuit breaker in the power panel popped as the motor decelerated.

I tested numerous circuits in the house and got the same results every time in both forward and reverse:

• If the circuit was equipped with a GFCI “outlet”, it would trip on start-up.
• If the circuit was equipped with a GFCI “circuit breaker”, it would trip when the motor decelerated/stopped.

Something has changed on the lathe. The jumper fix was no longer working. There is no new equipment in my garage and no modifications have been made to the lathe. I called PM and talked to Mike. After a few calls and emails he brought the vendor into the loop. After sending videos of the problem and pics of the wiring, the vendor suggested replacing the direction control switch. This is a 3 position rotary switch: forward-stop-reverse. I received and installed the new switch with no change. The vendor then suggested a few changes to the VFD parameters. After a few missteps on my part, not being too familiar with programming the VFD, I got the parameters changed and saved. Still, no change/improvement.

This is a time consuming process through no fault of PM or Mike. I’m on the west coast, PM is on the east coast and China is…..well China. So messages/emails took some time. I’m retired so I got plenty of that.

While waiting for the vendor to troubleshoot and suggest the next option I went ahead and replaced the old GFCI outlet with a heavy duty hospital grade 20amp GFCI outlet and for good measure, I replaced the circuit breaker in the power panel. The new heavy duty GFCI did improve things a bit. The lathe now starts every time but it is now tripping the GFCI outlet on shutdown/deceleration every time.

The vendor came back and suggested I replace that huge Siemens power relay. It took 2 weeks to get the part…remember….China. I installed it and no change.

So that is where this sits as of today. I know I can direct wire it and eliminate the GFCI altogether. I know I could just pull the GFCI outlet out and replace it with a standard one. That’s not the point. The point is that the lathe was working fine for two years and now all of a sudden it is not. Something within the four corners of the lathe has changed/failed. If the lathe was not still under warranty I would most likely choose one of those other options. But since it is under warranty for another year, I would prefer to have it fixed if possible.

Thanks for any help you guys can give.

Phil