Thanks for any help you guys can give.
Phil
Wow! Nice complete description. So I agree with Mark that it seems unreasonable that it might be anything mechanical on the lathe or a major relay. However, the VFD's manuals seem to qualify the VFD wrt aging. Especially the capacitors. However, I would expect the power capacitance to decrease with time or to even become a short. But this does not seem to be consistent with an increase in surges to increase currents. But since your GFI (GFI-GFCI) seems to trip with surges then what you might need is a choke or EMI filter on the incoming line.... to slow things down and spread the current surge out in time. If you look at most PC power supplies and even the lines to a monitor you see chokes. On digital electronics boards you also commonly see them. However, these are small and are suppose to kill the very high frequencies. A choke of this sort is just an inductor, with wire that is large enough to not represent much resistance to DC (low frequency) currents.
Edited this in the light of day.... Below I will provide a few thoughts about how I think these things might happen. But there are probably other reasons. From this maybe you will find a concept that will help you solve the problem.
In this case it should be an inductor that effectively does not have much inductance at 60Hz, but would at higher frequencies. Even having inductance at 60 Hz would not be much of a problem. It seems kind of expensive to replace the VFD, unless it is under warranty. EMI filters are usually suppose to stop the higher frequencies, but commonly much higher than you expect that would trip an GFI.
Old GFCI contained thermal measurement such that the measurement of the difference in current out the hot line and back the neutral would trip the GFCI if differences were around too long, and have trip sensing times of many many milliseconds. So that high frequencies did not affect them.
Modern GFCIs function using far more electronics and work differently.
This video provides some insight to one device and maybe useful.
So we have to ask why your VFD your system used to work and now does not. Something has changed such that the current surges now trip the GFCI. There must be something about your set up or the VFD that now allows to current to flow in a different path than it used to. Perhaps a point in the circuit or the VFD now has less impedance, or more than it used to.
A simple example of how an older device, such as a drill , might function might help. A drill with a three prong plug would represent a machine where the hot and neutral connected to the motor windings which were not in contact with a metal drills housing. The ground line would be connected to the metal housing and is suppose to shield the operator from the motor windings ever touching the housing. If an operator comes into contact with the hot line current might flow through the body to another ground point. This would especially happen if the ground line were not really good. Perhaps the screw inside the metal housing holding the ground connection had rusted a bit or was just loose so that the resistance in the ground wire path was greater than the resistance via the operators body to ground. A GFI that ensures that the current in neutral and hot wires are balanced would trip.
A surge of current is still a just a current and in concept a GFI should function the same as for normal operation. However, the sensor for checking for balance is not perfect but does try to measure an absolute difference in balance not a percentage of the load current. So any error in the sensors balance measurement will increase with current amplitude, such as a surge. Perhaps the reason the shut off now trips the breaker GFI but the power up does not is simply because the shut off is more abrupt and so generates a stronger surge. After all shutting off a motor, especially when there is a brake can develop a very large back voltage and current. (This is the reason in DC circuits you will commonly see diodes strapped across a coil in one direction. It is to send this current back through the coil until it can die out rather than to generate a large voltage at the controlling switch and electronics). Why at the breaker and not at the wall GFI. Perhaps because at the breaker box the ground and neutral are connected just on the up side of the GFI, but at the wall GFI they are not. So there is less resistance in the Neutral and ground wires simply because the path is shorter before they are shorted? (At the main breaker box the Ground and Neutral are tied together.) My guess would be that if you moved the wall GFI to a point right at the breaker box and ran a longer wire from the lathe to the GFI it might not trip on power up but still wound on power down.
So where am I headed. It would seem to me that something has changed in either inside or outside the VFD. This change could be in either the Neutral or Hot path but with respect to the ground, which is not just the ground wire, but the metal parts of the equipment.
I suspect that in all of the trials that you have done that you have disconnected all the wires at most points and reconnected them, but if not I would try this. Remove and clean all of the ground, neutral, and power connections and then tighten them back down. I would also do this for the GFI wiring both at the GFI and at the connects to and at the breaker box. Since a difference in the impedance between the neutral and the hot in combination with the inductive pick up from a surge of currents around the ground wire might just induce current in the ground wire. Clean connections are especially needed around the VFD and around the motor if it is grounded to the mount... and the mount to the cabinet. Inspect all of the places where you would think the power and ground and neutral should be well insulated, frayed wires, wire feed through clamps that pinch the insulation, etc. Having eliminated these things then the focus moves to the VFD. It is quite possible that the internal parts have become dirty and there are leakage paths that have developed. Or that there are insulators, or separation distances that have changed. Looking for these is more difficult since if it is not at the outside connectors one has to look inside and we have little perspective as to what we are looking at once the cover is removed. But is most likely in the power components which will be connected to the large capacitors.
Aside stories.
1) Small amounts of dust and dirt around electronics can cause strange effects. At work long ago a group of us had access to an expensive, early HP desktop computer, HP45? Quite expensive. Before the PC etc. It ran on HP Basic. It worked great for our calculations and several of us used in an office environment. However, it started acting up and just not working correctly. After a while we finally paid an HP tech to come and look at it. He took it apart, lifted a large circuit board out, held it vertically and sprayed cleaner on the board, letting the excess condensed fluid run off .... and then put it back together and left. Prior to this it did not really even look dusty. However, it was fine after this.
2) I do not have GFIs on my machines either at the wall or the breaker box. They are each on a separate circuit and breaker at the box and are essentially hard wired (dedicated outlet in one case). My walk-in basement is pretty dry unless it is just raining very hard outside.
3) I was nearly electrocuted as a young boy. I was in the basement setting up a train set and so was plugging the train cord (no ground, 2 prongs) into an extension cord that had no ground. I had no shoes on and was standing on a damp concrete floor. As the two metal prongs were entering the extension cord holes I managed to get my thumb across the two prongs (I think or something?) as I was pushing them into the extension cord holes. I am not for sure what happened, but I could see the electrical arc that developed between the extension cord "holes" and my hand/thumb. The train cord dropped completely away. This current flow continued until the uncontrolled flagging of my body cause the extension cord to come unplugged at the wall. All in all I was very lucky that the extension cord was not longer! I had a burn on my thumb which became a small scar, but which eventually disappeared. My hand muscles and mind would not respond to release the cord. Anyway, I understand what it means to be shocked to the point that you have absolutely on control of your body or mind!
When I went to college I studied Electrical Engineering ... and Physics, but I cannot say that this accident was the motivating factor! As a boy there were other interactions with electricity which went awry, but I after that event I was always more cautious.
Dave L.
