X/Y Zero Not Consistant [Archive]

nat_wheatley

01-15-2015, 05:58 AM

Occasionally, I'll have an issue mid sheet (typically a COM error), and need to run the X/Y zero routine, then re-run the cutting for that sheet. I'm using Ecabinets and the SB Link, so it's necessary to run the sheet again start to finish (no picking up where the issue occurred).

Virtually every time I do this the X/Y position is off by enough from the original that I need to scrap the sheet and run a fresh one.

I'm using the proximity switches to zero things. Also, there's no material movement as I'm cutting full sheets (with vacuum)

Is this typical, or is there something I could do to increase the re-zeroing accuracy of those 2 axis?

Thanks

Brady Watson

01-15-2015, 09:40 AM

Nat,
No this is not typical & there are 2 areas you should focus on to shore things up.

First, it is absolutely imperative that you have solid communication with the computer/control box. I recommend that customers install a PCI to USB 2.0 expansion card in their computer (assuming ATX tower not laptop) to get the SB on it's very own USB bus/channel. Then plug in any self-powered USB 2.0 hub (via Target, Radio Shack, Amazon etc) to the PCI card. Just plugging in the USB hub automatically creates what's called a 'Windows Enhanced USB Port' that is given priority over other USB ports with greater bandwidth and speed. Then plug the USB cable from the control box into the hub, making sure the cable is less than 12' long. If you have to go longer, there are solutions for that.

If using SB3 prior to v3.8.x, then close SB3 & locate SpeedTest.exe in C:/ProgramFiles(x86)/ShopBot/SB3/Diagnostics folder. Run it. You should have at least 70% COM efficiency. If running 3.8.x, then follow the instructions carefully under the UT and UD commands. Run the appropriate speed test BEFORE you make changes to your hardware so you can measure your results. (On a side note, Dells are notoriously lame in the USB hardware department...) You have to remember that the computer is part of the control system & you want to make it as rock solid as possible.

Second, you want to make sure that your prox switches are adjusted properly. They are inductive switches, which means they are looking for a metal target. The gap between the head of the target bolt and the face of the prox switch barrel needs to be adjusted so that it has a gap of about .03-.05". If you want to 'blueprint" your targets, remove them and buzz down the 3 or 6 'peaks' on the head of the bolt so that the entire head is nice & flat. No need to go crazy, just kiss the wheel of the grinder or belt to take down the high spots. A metal lathe works best. Then adjust gap.

When you are done, just chuck up a nice sharp v-bit and put it at your bottom 0,0 point. ZX, ZY or Z2 - then run SB Setup via the TS command. Choose the 'Make it easy on me' option and let it do the C3 calibration. Then you should be all set.

Also...it's cold this time of year & the air is very dry. Make sure 100% that you are running a BARE copper ground wire THROUGH your 4" dust hose from the dust foot joint (just fold over hose & clamp) to the DC itself (pull out & tie into any metal part that has continuity with electrical ground - anywhere on chassis is usually fine) - because static this time of the year is just brutal. This high voltage can cause all kinds of issues...0.025" copper-coated solid steel MIG wire works best. HD has it.

Imagine you have your ipod hooked up to some little speakers (your 5v USB signal and 24v prox switches). It's playing your music at a comfortable volume. Now your neighbor is having a party and the DJ has Peavey cabinet speakers like the kind you'd see at a concert. He cranks up the volume and now Hank Williams Jr. (http://youtu.be/k2GBQqSKl8w?t=55s) (50,000 volts) is STEPPING ALL OVER your your little ipod setup. THIS is what static electricity does to your COMs and sensors on the machine - so due diligence regarding grounding is imperative if you want a solid and reliable machine.

-B