Random offsets and static electricity.

[gmm50]

Hi All:

Had the shark for 18 months now and moving to heavier productions runs. 10 hrs/day and 6 days per week.

We have problems where the controller/stepper combo seems to get off in one direction at a time. I'm suspecting static electricity.
We are cutting plastic (HDPE) and have a cabinet around the unit containing the debree. We only clean off the spoil board each run.

My question is where would you run a ground wire form/to?
The tip if the bit to earth ground is the answer but that darn bit keeps spinning and wraps the ground wire all around itself!!

We're using a Dewalt router (611) and the bit measures open as compared to the metal body.

Anybody have experience with this or even have any suggestions good or bad (I'll be the judge).

[gmm50]

We have the shark in an enclosure (wooden box) and have been removing waste plastic more often.
That and perhaps the monsoon season coming to an end (power input is better) has eliminated the issue for now.

[Rando]

gmm50:

When I first got my Shark, I was cutting acrylic and HDPE, and noticed similar issues. As you know, it's the separation of the material from contacting the other material (and the bit) that causes the charges to be "stolen" from one side of the cut to the other. The nice thing about this particular kind of static build-up is that it happens over time, one chip at a time. That makes it a lot easier to get rid of the charge, when compared to what's needed to protect against those static discharges we get when we get close to a charged machine. Am I right?

In any event, here's what I did. There are two halves of the circuit that have to be connected together. One end is the spindle. Not the casing, but the actual part that's spinning. Hopefully there's good electrical contact to SOME wire from the case. The other end is the machining bed and enclosure walls.

For the enclosure walls, if there is one, see if you can find some anti-static mat cleaner, like this: http://www.uline.com/Product/Detail/S-6 ... nti-Static. Spray that on, and lightly but completely wipe down your enclosure internal walls with this stuff, and let dry. It's not like you're gold-plating the inside...it's a pretty high resistance, but usually more than enough to bleed off that static charge. Next, make sure the metal frame of the enclosure is securely grounded. That ground should be on the same circuit as the controller box if possible, so that it can have a shorter path to the spindle motor, where the "other half" of the charge is building up.

Next up, the machining bed. On the shark, it's the aluminum extrusion channels, connected cross-wise by aluminum angle or some sort. Hopefully on your machine there is good conductivity between the several extrusions. Remember, aluminum oxide is an insulator, so just because two anodized surfaces are in contact doesn't necessarily mean they have good electrical contact. If there's a aluminum bar or angle that's bolted to each (like on the HD and Anniversary models, like I have), then you're golden. The bed needs to be able to bleed off the charge to that ground connection (that then goes back to the motor spindle and bit). But, if you use the Z-0 sensing plate, a well-grounded metal bed, it won't read correctly....well, mine didn't anyway.

So...connecting the metal bed and the spindle (the spinning part, not necessarily the case), you need low enough resistance to bleed off the static but not enough to trip the Z-height sensor circuit. Turns out that value (after extensive use of multiplication ) is about 470K ohms. Such things are available in random places, online, and of course costs $10 to ship and $0.001 to produce. Some won't even sell you just one. You can use a 1/4W or 1/8W resistor, as there's no significant power dissipated. The 1/4W is easier to work with since it's bigger, but either are fine. Connect one end to that aluminum bar using some fastener-type-thing. I used a crimp lug under one of the bolt heads going into the channel extrusions, so good contact is there. Connect the other end of the resistor, using a (preferably) green wire, to the metal case of the controller, which is well-connected to the ground coming back from the spindle motor.

If you're using MDF under the HDPE, you might consider lightly (no soaking!) spraying the newly-faced MDF with the anti-static spray and letting it dry completely. That can help, but obviously isn't the complete solution on its own.

That resistor alone, for me, has completely eliminated the painful shocks I was getting ever time I approached the machine while it was cutting plastics. If it gives you problems with the z-height plate, then increase the resistor.

If getting one of these resistors is a problem for you, private-message me and I can drop one in an envelope for ya.

Hope that information helps.

Regards,

Thom

[gmm50]

Wow!!!

All things I should have known and none that I addressed. I am actually an EE (40+ years) but my degree is so old it says DC only on it now.

I think you right on on all your observations. The spray is a good idea. I would caution readers that the green wire, 3 rd prong and case ground might not be connected to anything like earth ground back at the power input panel. So check at the machine first since making them all ground only at the machine is not a good plan.

In electronic manufacturing plants the grounding straps everyone wears are connected to earth ground (Green wire) with 1 Meg Ohm resistors so your 470K ohm is a good choice.

We've taken the approach to keep the insides of the box the router is in (for dust purposes) cleaner. We don't vacuum it after every panel but certainly clean it out after every panel. That seems to chase the problem away but again we high humidity in SW New Mexico right now.

Thanks for your input and I believe your correct in all your observations and conclusions.

gm

[Rando]

Yes, tying the grounds together is for the metal-to-metal conductivity needed for the Z-axis sensing probe, and for bleeding the charge from the chip/bed back to the spindle/bit. Not about being "grounded" in the safety-return sense.

Regards,

Thom