Strange issue with Z Zero touch plates.

[jimskeet]

Hi
I have a really strange issue with my touch plates for z zero.
I have a touch plate that I use to zero my Z. I decided to add a 2nd touch plate but fixed at the table to use with big tex tool setter. My touch plate is connected to one pin and the ground is through the spindle ground. So when the touch plate touches the tool, probe input goes active and mach3 gets the signal. I connected the 2nd touch plate on the same pin. So any one of the touch plates touching the tool gives a probe signal to mach3. Tested them by touching them on the tool and they work as expected. Here is the issue....

When I have both the moveable touch plate and the fixed touch plate (which is always on the table since it is fixed) on the table, the probing routine stops before it touches the touch plate and acts as it got a signal. Stops and retracts according to the routine. If I remove the moveable touch plate from the table and place at a little distance from it (it still stays connected on the input) the routine runs fine. I did some experiments and found out that the routine acts like it got a signal even if the 2nd moveable touch plate doesn't touch the table but is held over it at a close distance.
I uploaded some videos for anyone to check.
When this happens I do not see the led of the probe to turn on even instantaneously so if for some reason mach3 receives a signal it is so fast that doesn't even blink the led.
I have tested it many many times. There is no short anywhere. All the cables are fine.

Any help because I spent like 4 hours today trying to find a logical explanation with no success.

Videos of the issue
https://www.dropbox.com/s/pruqnkkkef...Table.mp4?dl=0
https://www.dropbox.com/s/uergok6om0...Table.mp4?dl=0
https://www.dropbox.com/s/94zzryndtf...Table.mp4?dl=0

Similar Threads:

• strange emco 325 issue
• Strange code issue!
• Strange Issue? Maybe?
• Probes and Touch Plates
• Strange issue with a C10 BOB

[ger21]

Ground loop, or interference.

[jimskeet]

Thanks for the answer. Do you have any suggestions on how to troubleshoot this?
Since it happens only when I bring the 2nd plate close to the table, does this indicate that it is more possible to be an interference than a ground loop? Wouldn't a ground loop produce the issue no matter where the moveable plate was?

And if so how would I prevent interference to get to my touch plates?

I have routed the cables of the touch plates away from other cables as much as I could.

[mactec54]

Thanks for the answer. Do you have any suggestions on how to troubleshoot this?
Since it happens only when I bring the 2nd plate close to the table, does this indicate that it is more possible to be an interference than a ground loop? Wouldn't a ground loop produce the issue no matter where the moveable plate was?

And if so how would I prevent interference to get to my touch plates?

I have routed the cables of the touch plates away from other cables as much as I could.

It seams that you are getting noise/ interference from the stepper motor, that would be a good place to start, with the wiring, some more shielded cable and better connections Ground as well

You should not connect your touch plate ground wire to the main power supply Ground, your Breakout Board should have a isolated Ground connection for your touch plate

[jimskeet]

Mactec thanks for the answer.
My touch plate is connected on one of the inputs of the bob. But only one cable going to the Pin15. The ground is acquired through the spindle ground connection. Should I make a new ground connection from the Gnd pin of the board to the machine? Wouldn't this create a ground loop since the machine would have 2 ground connection?

I thought to do this test.
Wrap the touch plate cables with aluminum foil and connect it to ground. So this will act as a shield, and see if I still get the issue. This will show me if it is interference or not, right?

[jimskeet]

After further testing I am even more frustrated.

I disconnected the 2nd plate and used only the moveable one. I found out that even with the one plate connected the issue occurs, but a lot harder. Before installing the second plate and try to use the big tex tool setter, I was using my plate like this.
I would place it on my material, move the spindle like an inch or less from the touch plate and hit auto tool zero.
It worked fine every time. It happened to stop before touching the plate like 2-3 times during the last year so I just thought it as a bug and didn't bother with it.
Now with the big tex tool setter, the spindle starts the probing move from the top of Z travel so it takes a lot more time to touch the plate. This revealed that even with one touch plate connected the issue still happens. But a lot harder. With both plates connected it stops like 1-2 seconds after it started. With One plate only connected it some times doesn't stop before actually touching the plate and some times stops like mid way. But noticeably a lot harder than when both touch plates connected.

I even managed to make it stop once by having both touch plates disconnected from the control box. I was touching the frame of the machine and moving around it and close to the control box. This happened only once though. So the issue may is cause by something in the control box???

I attach 2 photos of the control Box interior and the area that the cables exit the control box to go to the machine.

The vfd is not the cause of the issue because the problem happens even with the vfd powered down.

Any help is more than welcome cause this is driving me crazy...

[ger21]

Many people add a capacitor across the input, which filters out the noise. While not the ideal solution, it works for many people.

[jimskeet]

I will try that. Do you know what F value capacitor should I get? Also will the capacitor work on the dc voltage input?

[mactec54]

Mactec thanks for the answer.
My touch plate is connected on one of the inputs of the bob. But only one cable going to the Pin15. The ground is acquired through the spindle ground connection. Should I make a new ground connection from the Gnd pin of the board to the machine? Wouldn't this create a ground loop since the machine would have 2 ground connection?

I thought to do this test.
Wrap the touch plate cables with aluminum foil and connect it to ground. So this will act as a shield, and see if I still get the issue. This will show me if it is interference or not, right?

Try the foil first, a capacitor is not a fix, and will slow down the activation time of the touch plate

The main power that supplies your machine, do you only have ( 1 ) main supply cable that powers everything

[jimskeet]

I tried the foil with no success at all.
I have 3 phase power supply at my home.
The control box takes power from one phase. The PC that runs mach3 takes power from the same phase.
The VFD takes power from a different phase. But vfd doesn't cause my issues since the issue is there when the power on vfd is disconnected.
In the control box everything is powered from the single phase. 2 power supplies for the closed loop stepper drives, 5v for the board and 12v for the homing switches

[jimskeet]

I tried also to use a debounce value in general config. I went up to debounce interval at 4000 and the issue still happened. If it was noise from EMI shouldn't it not trigger the input at such a high debounce value?

[ger21]

Debounce has no effect on the probe input, because it adds a delay.

[jimskeet]

Debounce has no effect on the probe input, because it adds a delay.

Oh
Didn't know that. Thanks for the info. One question. When you said a capacitor before you meant to place the capacitor so one leg is in the pin input and the other leg in the ground input? Or connect the capacitor in series with the touch plate?

[ger21]

Not sure, electronics aren't my specialty. But a lot of my 2010 screenset users have used them when they had similar issues.

I think a .01uF or .1uF will work.

As Mactec said, they do slow response time slightly.

[Kenny Duval]

I pulled a ferrite bead off of a VGA cable that I no longer needed and looped my fixed touch plate cable through it and fixed my problem when I had it. It would do the same thing randomly while measuring tools and occasionally while probing.

[jimskeet]

I pulled a ferrite bead off of a VGA cable that I no longer needed and looped my fixed touch plate cable through it and fixed my problem when I had it. It would do the same thing randomly while measuring tools and occasionally while probing.

Thanks for the answer man.
Will this do the same thing with the capacitor???
Also when you say you looped the touch plate through it how exactly do you mean? Can you draw a wiring diagram so I get this clearly?

[coherent]

Mactec thanks for the answer.
The ground is acquired through the spindle ground connection. Should I make a new ground connection from the Gnd pin of the board to the machine? Wouldn't this create a ground loop since the machine would have 2 ground connection?
t?

I would imagine you are best off grounding to the same board you connected to pin 15. I've always ran a 2 wire cable for height probes.. one to the input pin on my breakout board and the other to the breakout board ground and never had an issue. I would think the way you have it grounded through your spindle ground would be more susceptible to interference or noise.

[jimskeet]

The thing is that the ground through spindle is not there for the touch plate. Spindle ground goes to the vfd. So there is no way to remove the ground connection through the spindle. When the machine was built it was intended to have a cable go from ground to the endmill on the spindle that is to be probed. But we realized that it wasn't needed since the spindle was already grounded.

[ger21]

That's because your breakout board is not isolated.

[jimskeet]

Yeah I got that. So the best thing would be to have an opto isolated board for this to work without issues? If the capacitor or ferrite bead trick won't work what else is there to try?