VFD spindle controler noise

[bballish]

Hi
Im having a problem that I hope you can help me with. I have wired up my Huangyang vfd spindle controlor and when I turn the spindle on mach4 seems to just lock up. The vfd is not connected to the motor control box it is mounted on a wall about 3 foot from the motor control box and the motor supply cable runs directly to the spindle thru the wire carrier tracks. The spindle wire is shielded cable and the drain wire is connected to earth ground. So I was hoping you could give me some suggestions for getting rid of the electrical noise as that looks like the problem.

Thanks

[Bob at PMDX]

Let me make sure I understand your configuration:

The vfd is not connected to the motor control box

I presume the "motor control box" is where you have the SmartBOB and (if I remember) the C-10 break-out board.  So there are no wires going from the the control box tot he VFD (i.e. 0-10V control voltage, or signals for forward/reverse), correct?

it [the VFD] is mounted on a wall about 3 foot from the motor control box and the motor supply cable runs directly to the spindle thru the wire carrier tracks.

"motor supply cables"?  Is the the AC power to the VFD?  Or is this the wiring from the VFD to the spindle motor?  Or both?

The spindle wire is shielded cable and the drain wire is connected to earth ground. So I was hoping you could give me some suggestions for getting rid of the electrical noise as that looks like the problem.

"spindle wire" - is this the wiring from the VFD to the spindle motor?

So, if I read this correctly, with no control wired from the SmartBOB/break-out board to the VFD, you manually turn on the spindle (from the VFD controls) and this causes Mach4 to stop responding.  Is that correct?

The usual cause of this kind of problem is electrical noise from the VFD/Spindle motor corrupting the USB communications.  This could be conducted noise through AC power lines or signal control lines, or radiated noise.  And these issues can be challenging to track down.  Our usual suggestions include

- making sure all VFD wiring is as far as possible from the PC, SmartBOB, break-out board, and limit/home switch wiring (or any other signal that feeds into the break-out board)

- making sure that the AC power for the PC and any break-out boards is as isolated as possible from the VFD (and stepper/servo motor) power wiring.  This can mean powering them from two different AC lines from your breaker panel, or if they have to run off the same AC power line from the breaker panel, they should "meet" at the wall outlet, not at a common outlet strip on the machine.  This isn't always possible, but we may ask you to try this for testing purposes.

- use shielded wiring for home/limit switches, with the shield grounded to the break-out board's ground

It would be very helpful if you could draw a wiring diagram for your system and post that here.  Include all of the AC power wiring and grounding connections that you have in your system.  Pictures of you machine, specially the PC and control box) would be helpful as well.

Bob

[bballish]

Hi
The motor control box is where the motor drivers and c-10 board and pmdx-411 are. There are no wires running to the vfd from this box.

The motor supply cable to the spindle is the D.C. voltage from the vfd to the spindle.

The motor supply cable is the wire that powers the spindle. And yes I manualy turn on the spindle directly from the vfd.And when I turn on the spindle it locks up mach4.
The A.C. for the vfd is 220v and is a separate circuit from the 110v for the computer and the contol box.

The limit and home wires are not shielded so this might be the problem. I will fix that and any other things you mentioned and see if this helps

Thanks

[Bob at PMDX]

A quick test is to disconnect the limit/home switches from the breakout board and see if that alters the behavior.  How close is the control box to the spindle motor?

Bob

[bballish]

Hi Bob
I will disconnect the limit switch wires and see if that helps. Also the control box is about 4 feet from the spindle. Also when will we be able to home a slaved axis with a limit switch and also gantry squaring . I dont think I can work on this untill Friday. I have to fix 2 lawn mowers and a weed eater before my grass gets waist high.

Thanks

[bballish]

Hi Bob
I got back to working on my machine. I disconected the limit switch wires and unplugged the motor cables. When I turn on the spindle with everything unplugged Mach 4 runs good I can jog the axis's and the readout on Mach4 moves just like the axis where plugged in. But if I plug in any axis motor and turn on the spindle Mach4 locks up. So I have no idea how to fix this. I read on another forum that some Chinese spindles dont have the 4th pin on the motor jack attached to ground on the spindle case. So I took the cap off the spindle and sure enough the 4th pin didnt have any wire attached to it. So I soldered a wire to the pin and attached it to the spindle case for ground. But this didnt fix the problem. Also I was wondering how to check to see if my earth ground was good. But I dont know how to check it ?

Thanks

[Steve Stallings]

You stated that the motor supply cable was shielded and connected to earth ground.

This shield connection to earth should be made directly on the earth terminal inside
the VFD, not to some point on the cabinet. You can also try a filter on the output of
the VFD, but that gets expensive because you cannot use just any old filter. You
would need one designed to work in the high power environment with 400 Hz
currents present at the VFD output. An alternative that may help is to loop the
motor supply cable though a large ferrite donut a couple of times. Here is a link
to a typical example:

http://www.digikey.com/product-detail/en/laird-signal-integrity-products/28B2400-000/240-2120-ND/571928

Having this cable inside the same carrier with the wires to a home or limit switch
for the Z axis is likely to couple VFD noise into the ground of the breakout board,
and back through the PMDX-411 and into the USB connection on the computer.
It would be better if you had a breakout board with opto-isolation for the switch
inputs so the noise would be mostly blocked by the breakout board. This will
possibly require an additional power supply depending on the breakout board.

If these suggestions do not help, you may be a candidate for a PMDX-424
which has isolation both on the switch inputs and on the USB connection to
host computer.

[Bob at PMDX]

Looking back though your previous posts, I see that you sent us a diagram back in Nov of last year.  I'll see if I still have that hanging around and if it has the detail that we need for this problem.  But at least a couple of pictures of your current configuration would be helpful - showing what is in your control box and how everything is connected, and one showing overall layout/position of the PC, control box and machine.

In addition to what Steve mentions...

Did you try leaving the motor cables unplugged and connecting the one or more limit switch wires and see what happened?

When you say you "unplugged the motor cables", are these the cables from the motor drivers to the motors themselves?  Or the cables from the C10 board to the stepper drivers?  I presume they are from the stepper drivers to the motors.

Are the stepper drivers inside the control box?  If so, try disconnecting the C10 from the stepper drivers and DO connect the motors to the motor drivers.  See if that still causes the issue.

Bob

[bballish]

Hi Bob and Steve

I will get back to working on this problem. Thanks for all your help ! I have a couple of questions that you might can answer for me.

(1)  My VFD is a Huang Yang  220volt single or 3 phase. I am running it on single phase 220volt. I have the 2 hot leads going to the r and s terminal inside the Vfd and the neutral lead going to the ground terminal of the vfd. I was reading on another forum last night and saw several threads that stated that the neutral lead should not be hooked up at all. They stated that the ground terminal should go to earth ground. If this is the case then it leads to another question.

(2) I have a older ground rod that is just outside my workshop. I am not sure but I think it was installed for cable T.V. years ago. I am not sure this is a good earth ground . How do I check this ground rod to make sure if it is good?

In the meantime I will try some of the other things you suggested. I will try and send some pictures also.

If I cant get it working do you think the pmdx-424 would solve this problem? I dont want to spend a lot more on this machine. Im on social security and the money only goes so far.


Thanks

[Steve Stallings]

In the USA, our safety ground (earth) and our Neutral are connected (bonded)
together at the breaker or fuse box panel. Single phase 230 volt circuits may
not even have a Neutral wire, but should have a safety ground. The safety
ground should never be used as a Neutral even though they are connected
together at the breaker or fuse box panel. This is because a loose connection
in the safety ground wire can result in the frame of the machine becoming
electrically HOT if that wire was also serving as a Neutral for anything that
draws power. This discussion is secondary to your noise issues, but I wanted
to make sure these basics were understood.

The important concept about shielding of the cable from the VFD to the spindle
motor is that the shield should connect directly to the Earth or safety ground
terminal of the VFD. The safety ground from your power circuit should also
connect to this terminal. If your power circuit has both a safety ground and
a Neutral, and you had used the Neutral as the connection for the safety
ground or Earth terminal on the VFD, and then used the the actual safety
ground from your power circuit as the connection for the shield of the cable
between the VFD and the spindle motor, then you would have a less effective
path for grounding out the noise in the cable. You do not need a Neutral
connection to the VFD, only a safety ground connection.

Your breaker box or fuse panel should have a bonding bar with the safety
ground and the Neutral connected together. This bonding bar should also
have a connection to an actual ground rod in the earth or a grounded utility
pipe that is buried in the earth. Additional ground rod connections are
sometimes used with plasma cutters but are rarely used with other types
of machinery. Rods driven into the earth as grounds vary significantly in
effectiveness. A purpose built rod with a copper or zinc coating is normally
used and it needs to be driven about four feet into the earth in a area that
makes effective contact. Rock or extremely dry soil do not work well.

Please verify that you have the spindle power cable shield connected as
recommended, and possibly try one of the donuts we recommended. If
you still have problems, we will work with you as a test case to see if
the PMDX-424 can fix the problem. We request that you contact us in a
direct email to discuss this option if needed.

[bballish]

Hi Steve and Bob
I ordered 2 of the ferrite donuts and I hope they will be here by Saturday. In the meantime I will check some of the other things you have suggested. When I get the ferrite donuts should I use 2 of them on the spindle supply cable 1 at both ends or just use 1 of them ? So Ill see what happens.

Thanks

[Steve Stallings]

Use one, at the end nearest the VFD. Try to loop the cable
through the donut 2 or 3 times depending on how much
room there is.

[bballish]

Hi Steve and Bob
I installed the ferrite donut and it fixed the problem. The machine is running great so far. So I guess it was electrical noise on the spindle cable.
But I wanted to thank you guys with all my problems I have had and I would recomend your products to anyone.

Thank you very much.