Static Electricity

[jgro]

Hi all,

Just thought I'd give everyone a warning. Don't touch anything unless you are grounded. My machine was running a program and I accidently touched one of the motor housings. I felt a static shock and boom everthing stopped. I fried my xylotex board. I'm not a happy camper right now. Just thought I'd warn everyone.

jgro

[Al_The_Man]

More importantly, is your system grounded? All motor frames, the metal structure of the machine and any electrical cabinets/boxes etc should all be bonded together and connected to a good earth ground. On a well grounded system all ground wires are taken back to a common ground plate together with the main ground conductor. Also any power supplies that have grounded commons should also be connected at this point, as well as any outer shields of signal wiring etc.
Al

[keithorr]

Check the documentation that comes with your particular system.

From my AHHA Power System Manual.

"4.6.4 Motor Cable Grounding

DO NOT connect the cable shielding to the motor or frame of your machine. The cable shielding should be trimmed and heat shrink applied. The reason you do not want to connect the ground cable to your motors, is a ground loop noise problem could occur. Just make sure your machine is grounded properly to earth ground."

[Al_The_Man]

I think that needs a bit of clarification, I believe the motor grounding you are refering to is the motor signal wires which should be shielded and yes, should only be connected at one end to avoid ground loop current etc. the shield should be connected to the common ground I mentioned in my previous post, but the motor frame should be grounded to the common ground point, This is an industry standard and is outlined in the wiring code for machine tools. Otherwise you will get the rusult that happend in this instance.
Al

[samualt]

Gee, does that mean we can't run our Tesla Coil next to our CNC Router? LOL.

http://hot-streamer.com/pool/i4a.jpg
Bert's Tesla Page

Seriously, I'm grounding everything. If nothing else just string a 10 gauge wire from your machine to a pipe that is at least 2 feet, or more, in the ground (or plumbing pipe). All the metal casings on the machine and the frame needs to be grounded somehow. Sorry to hear about your loss jgro!

[keithorr]

My disagreement is with your statement "as well as any outer shields of signal wiring etc."


Consult your vendor's literature and follow their instructions. I'm not phsycic and wouldn't assume to know how another person should set up their machine. I wouldn't trust my investment to the assumptions offered on an internet message board. Too many variables.

I'm using steppers, and the wire leads were supplied by the vendor, and NEITHER end of the cable shield is "grounded", or attached to a ground, or provides bonding to either the controller or machine. As per the manufacturer's recommendation and construction.

As far as how I "ground" my machine or controller cabinet, I followed my vendor's instructions on that also.

___________________________________________

Just looked at a spare motor cable, and the end that connects via a 5 pin plug to the controller has the shield going to "pin 3-ground". The original cables, installed on the machine by the vendor, do not have that connection made. I'll have to ask someone from AHHA what gives.

[Al_The_Man]

I guess the point I am trying to make is that over the last 30 yrs or more since the advent of CNC machining and the inclusion of sophisticated electronic controls, a certain standard has evolved (usually the hard way) as regards the methods of wiring etc, I would suggest that any manufacture that does not heed these lessons is asking for problems.
I know in my experience of retro-fitting machines, I have occasionly taken a short cut, to my regret.
It is very unusuall to supply a shielded cable and not at least ground one end as this usually defeats the purpose and might even cause problems.
One only has to look at most major servo motor manufacturers and amplifier suppliers to see what they recommend as to their equipment and they all appear to conform to the industry standard way of wiring.
Some manufacturers of machine tools even go to the extreme of recommending a ground rod be driven in next to all their machines to avoid earthing problems.
One of the definitive publications on this subject is a manual put out by the National Fire Protection Assoc. entitled Electrical Standard for Industrial Machinery. This basically is an enhanced version of the Electrical code to cover machine tools.
Al

[wakeboard]

Actually doubt you fried your xylotex controller...you just burned a fuse!...I did this on 2 power supplies by accident....both of them the fuses blew!

[HuFlungDung]

I've been taught to do it the way that Al recommends. If the shield of a shielded cable is not grounded at one end, then it has nowhere to drain the noise signal to. If it is grounded at both ends then it just becomes another conductor in the system, and if the other grounds are not up to snuff, then it has the potential to carry some juice on its own.

[keithorr]

Well, I'm glad to have read all this. My cable shield is unattached at both ends, even though the documentation states that one end should be grounded (at the controller end). A back up cable has the ground lead intact, but it was a stock lead from the manufacturer. I started to doubt the setup until I read through the installation notes.

I'm using an open loop stepper system, so I don't have to worry about an encoder feedback suffering from noise.

The thinking at the time by the installer was that noise wasn't going to be as big a risk as a line voltage short.

What was a factor is that I dry mill graphite. The thinking was to keep the machine isolated from the driver box as much as possible, since live (not noise) current might find it's way from the spindle, to the machine, to the cable shield, and back to the controller.

The machine is enclosed, and the enclosure is rife with graphite dust when things are going, to the point sometimes you can't see anything trough the window except a black cloud. There is good exhaust ventilation, but it doesn't always keep up with the amount of conductive airborne particles that end up coating every possible surface and void.

If the cable shield was grounded to the driver box, and the graphite dust made a connection between the shield and the machine, I would have the straight line connection that HuFu mentions in his previous post.

I was told at the time to make sure the machine was grounded well, and keep the driver box as isolated as possible.

I apprecitate that there are standards, but one size may not fit all. If I didn't refer to my vendor notes, I might be tempted to reconnect the shield ground.

[arvidb]

Originally posted by jgro Hi all, Just thought I'd give everyone a warning. Don't touch anything unless you are grounded. My machine was running a program and I accidently touched one of the motor housings. I felt a static shock and boom everthing stopped. I fried my xylotex board. I'm not a happy camper right now. Just thought I'd warn everyone. jgro

You should never ever ground yourself anywhere near electricity! If you touch anything live while grounded, you are fried! I'm not kidding, this might very well be lethal!

If you need to ground yourself to protect your equipment, use an approved ESD wriststrap with a built in safety resistor.

Arvid

[cnczane]

Found this on the web and saved it for special moments just like this...
http://www.cae.wisc.edu/~dnelson/electrocution.gif

[It's an animated GIF that'll play outside of a browser... ]

Note the chef is grounded, and his toaster is not (because of a frayed cord, the "hot" is touching his case, which is not earth-grounded, which would then have tripped the breaker.)

IF the guy reaches for the toaster while touching the sink, he will present a low-resistance path to ground (he's mostly water and electrolytes you know).

"Ground-fault (circuit) interrupt" equipment (GFI or GFCI) senses the amount of current "going out" and "coming back" and if it differs by as little as 15 mA it will trip and open the circuit, figuring that some of the current must be "escaping" to ground.

We were always told in school that "it only takes 16 mA to fibrillate your heart" (i.e. screw up the rhythm so it won't beat well enough to keep you alive).