I am looking to upgrade my router to a Spindle. I read that the VFD Spindle is not compatible with GFCI outlets.
Will running the GFCI with an extension cord solve the issue?
I am looking to upgrade my router to a Spindle. I read that the VFD Spindle is not compatible with GFCI outlets.
Will running the GFCI with an extension cord solve the issue?
Nope:
Basically you need a VFD designed to work with GFCI outlets.
You would want to bypass the GFCI with the extension cord and plug the extension cord into a normal outlet.
To elaborate on this. As I understand it, a GFCI outlet is basically checking to make sure that the amount of current going out the live wire matches the amount of current coming back in through the neutral wire within a certain amount of phase offset. The issue is the GFCI outlets are designed around 50-60Hz because that is what the frequency the power company delivers the power at. A Variable Frequency Drive (VFD) varies the frequency of the current to drive the motor. This current can leak through to the power going back to the wall from the VFD. As such the power coming in to the neutral on the GFCI is not the same frequency as the power coming in from the live wire. An extension cord does not change the frequency.
Keep in mind I am not an electrical engineer and have not exhaustively studied GFCI outlets or VFDs so I could be way off on that. ![]()
US electrical code requires GFCI on almost every outlet. I have a Jet 12-21 VS Lathe and when plugged into a GFCI it pops the button on the GFCI outlet. I wired up a single outlet without GFCI and it works perfectly. GFCI is required anywhere there may be water present like a garage. In my house most of the outlets are GFCI/Arc Fault combination breakers because interior outlets require that now. In the Kitchen my garbage disposal is a GCFI because there is water present.
What a GFCI circuit does is monitor all 3 legs, ground, neutral and hot for the voltages between them. When it senses a variance in voltage between the 3 legs it pops the outlet off to protect the user from shock. An example would be a washing machine. It sits on rubber feet and if something shorted to the metal case it would shock you when you touched the case. By sensing the differential between ground, neutral and hot it would deactivate the outlet and you would not get shocked by touching the metal case. So the GCFI has a purpose but occasionally you have to not use GFCI because of a problem like the VFD. The VFD feeds back when you change the voltage and/or frequency to drive the spindle and that causes GFCI to sense that variance and pop.
Do a google search of âHow GFCI worksâ and here is the first thing that popped up for me.
How Do They Work? A GFCI constantly monitors current flowing through a circuit. If the current flowing into the circuit differs by a very small amount (as little as 0.006 amperes) from the returning current, the GFCI interrupts power faster than a blink of an eye to prevent a lethal dose of electricity.
@gdon_2003 seems to know more than I about it ![]()
Or improve the one you have.
The paper you linked to does explain part of what is going on.
Rather simplified, GFCI (RCDs in the UK) work by subtracting the currents in each line ( neutral and hot(live) ) and anything other than zero indicates current is flowing some other way, to earth (i.e. possibly through a human body).
Most modern electronic equipment has a âswitchingâ power supply (used to convert to another voltage than the line voltage more efficiently than heavy old wire transformers). The switching in these devices causes spikes on the line which can radiate off as radio waves (EMI) which is unwanted. FCC and CE regulations place limits on this so electronic components are added to âtrapâ the spikes to prevent them getting down the line. The cheapest method is to send them to ground (earth) using capacitors, precicely what the GFCI is designed to spot. Capacitors become better and conducting spikes at higher frequencies and VFDs are good at generating lots of high frequency and high power spikes.
An alternate method for stopping spikes from going along the line is something in the line itself which provides an easy path to 50/60Hz but a poor path for higher frequencies. This is a choke which does not provide a path the ground at all. With the correct choke, it will encourage the currents in the two lines to be equal and opposite. Unfortunately, good quality chokes are heavy and expensive (compared to capacitors) so are often not fitted.
You can get external mains filters with 2 stages of choke and very low leakage currents which should get rid of the noise without triggering the trip. Maybe not this precise model , but something like it. You can also get them without sockets for wiring in-line. You can see in the schematic 2x L1 and 2x L2 which are the coupled chokes.
It has come to my attention that many super-cheap VFDs actually ship without any EMI suppression capacitors, probably because the cheap method was causing lots of domestic consumer complaints about tripping and the easy solution is to make it even cheaper! Worse, they often donât have any grounding (earthing) either. If youâre dealing with a budget VFD then youâd do well to check for both of these points.
Iâve been running mine on a gfci but with an emi filter for the last few weeks and havenât tripped yet. Even with a good voltage spike of having the zed plow where it shouldnât. Itâs just one observation biased moment but Iâm keeping an eye on it and hopeful itâll continue.
Share the name/brand of EMI filter?
What brand of GFCI outlet do you have? (or is it whole house filtering)
Iâve used these to get rid of ground loops on studio gear Amazon.com and thought it might work well to poor man condition the line a hair. Reading that you all do your best to be a stable vfd made me hopeful it wouldnât need a lot of correction.
The gfciâs I have are socket level for my workshop. Iâll see if I can suss out what was installed a few years ago but the wall itâs on is a bit buried at this time.
As explained well above, thereâs no inherent reason you canât run a VFD or any other rectifying input power supply on a GFCI, the issue is that many of the cheap VFDs are lacking a decent noise filter on their input.
Iâve been running a cheap HuanYang 2.2kW VFD on RCD (UK name for GFCI) in two separate houses without any issues for a few years. When I first installed it I made sure I used braid & foil shielded cable for the spindle power and to put a 2 stage LC filter on the input of the VFD.
The 2 stage filters (lots of them around, the units Jason linked from RS are likely good quality, Digikey and others will probably have decent units too but seem to call them 3 stage) are basically the part that the chinese VFD manufacturer saved money by leaving out, so what youâre doing is putting the missing parts back.
Hereâs a couple of quick examples, Iâm not sure what size VFD or supply voltage youâre running, I am on 230V over here and put a 20A rated single phase filter on my VFD, which was oversized but not a significant increase in cost.
https://www.digikey.com/en/products/detail/delta-electronics/10DRCG5/1718582
You wire them up in line with the VFD, donât put anything else downstream of the filter other than the VFD, if thereâs other devices you want to filter, get them their own filter, put the filter as close to the VFD input as you can, installed in a properly grounded metal box.
EDIT - At the risk of being a bore⌠Remember the GFCI canât help you once the VFD is in the power circuit, the VFD output is not protected by the GFCI and you need to ground things properly, test that the spindle shell is properly grounded etc.
When selecting an EMI filter the primary consideration is how many amps it can carry. Make sure it is rated for your application. A 2.2kW spindle is going to need ~20A if youâre on 110/120v or ~10A if youâre on 220/230/240v.
The next consideration is the attenuation, higher is better. If not published, the value of L should be a good indicator, bigger values of L should give more attenuation.
These two interplay, for a given physical size/cost of filter, a higher current filter will typicall have lower L and poorer attenuation. Therefore:
As well as the original socketed filtered inlet I linked to and the tin boxed unit Liam linked to you can also get open units like: https://www.amazon.co.uk/dp/B0BX9VJTCY (I have not checked the safety/quality of anything I have linked to, that is your responsibility). Oviously, an open unit like this must be enclosed away from where it can be touched. It is important to keep the connection between the VFD and the filter extremely short (just a couple of inches, max).
In all cases, when assembling any system (turnkey or homebrew) it is essential to check with a multimeter for a good connection between the ground pin on the supply plug and all exposed metal items (EMI filter case, VFD ground terminal, VFD heatsink, Spindle body). Anyone who suggests that any earth/ground should be disconnected to make things work better is wrong⌠whilst this approach may sometimes fix some symptoms it is a dangerous cure for the underlying problem.
How did you wire it? Can you post pictures?
I know itâs just a single datapoint, but Iâve been running my whole machine spindle and all on a SquareD Q0 GFCI/AF combo breaker for over a year and have not had a single trip. I had no clue I wasnât âsupposedâ to run on GFCI until I read this thread, which surprised me because as @gdon_2003 said, itâs code to have all garage 110V outlets on GFCI due to it being an âoutdoor environmentâ.
I think there is a lot of variance in the sensitivity of different GFCI brands. If the GFCI you have works then great. If it does not work for you then you can change to a non GFCI outlet. A lot of variable speed devices feedback on the ac line when you change speed. That is what is tripping the GFCI off. In my case a variable speed lathe controller trips the GFCI circuit in my shop. Luckily I had a single outlet circuit in the shop I could remove the GFCI outlet and replace it with a regular outlet.
I installed my own wiring and used GFCI outlets on the first outlet on branch circuits because most GFCI breakers take two positions in breaker box and that limits the number of branches available in my breaker box. That gave me the flexibility to make branch circuits available. My breaker box has 42 slots and 220 takes 2 positions per breaker as well as GFCI/arc fault breakers. So for 120 vac circuits that eats up a lot of positions that are not usable. That is whyy first outlet in 120 branches are GFCI outlets and not GFCI breakers.
This topic was automatically closed after 30 days. New replies are no longer allowed.