Z-Zero losing position on new machine, anyone else having this issue?

[zeke]

A have a PRSstandard 48x96, purchased in 09. It was suggested by another member I post this here for more feedback.

I wonder if I have something wrong with my grounding or a different problem. I went back and checked all my grounding connections to ensure they were tight and I'm still losing position. Now it is really starting to get costly, 2 broken o-flutes and wrong cuts on type1 pvc because of losing position. I hope I can get beyond this soon. Each time I think I may have fixed an issue this same problem comes back. It is very challenging in that it appears the only way to determine if a problem is fixed is to use the machine until this happens again and in the mean time I’ve been thinking the problem is resolved. I’m very anxious to produce some products and realize the wonder of this machine in all it’s glory!!!

Tonight when I was cutting some type1 pvc the x lost it’s place by over 50 inches after about 10 minutes of cutting. I was running at 0.7 ips on both feeds, a comfortable pace, nothing difficult just straight lines before it lost position. I was only taking out 1/16th of material so I wasn’t taxing the machine at all. I then zeroed out again, turned off my dust collection to take that out of the equation and at the onset of starting the cut, the SB3 software quit and paused, I’m perplexed, probably need to call TS. I also looked at the Control box and no apparent problems persist, loose wires, etc..

It would be great if you could review the steps I took to ground my system and provide feedback, I’m not an electrician. I am going to create another post with the pictures to support the steps below that may help to paint the picture. I hope I’m not cancelling out the ground somewhere and over engineered this configuration.

1) A #6 ground wire run from the main service panel (I don’t have any sub panels) grounding lug (on the same lug as the main ground wire) to a junction box affixed to the bot frame.
2) Inside the junction box, there is another lug to distribute the #6 wire to #12.
3) Separate #12 wire from the junction box lug to xrail
4) Separate #12 wire from the junction box lug to other xrail
5) Separate #12 wire from the junction box lug threaded through middle of ‘x’ echain to ycar and attached there, continue from this connection through middle of ‘y’ echain and attach to zcar, then attach to z.
6) Separate #12 wire from the junction box lug to the control box ground, then attach from there onto frame, I scraped the paint off the frame at the connection point.
7) Separate #12 wire from junction box lug through x and y echains to the dust foot. The #12 wire is then connected to both the end of the vac hose internal wire and the external ground wire wrapped around the hose. I purchased the ground wire that is wrapped around the hose from Woodcraft several months ago in a grounding kit.
8) Separate #12 wire from the junction box lug to another lug on my vac stand bare steel upright that is attached to the same aluminum base as the dust collection vacuum and is used to support my vac hose boom arm.
9) Dust collection vacuum system grounded using ground wire (kit wire used) attached to same lug on vac stand as noted in previous step.
10) The dust collection vacuum hose that is attached at the dust collector has it’s internal hose grounding wire attached to the dust collection system grounding bolt using the same grounding kit wire.


Grounding pictures....

junction box lug

6646

control box ground

6647

frame ground

6648

xrail ground

6649

other xrail ground

6650

ycar ground

6651

zcar ground

6652

z ground

6653

dust collector lug ground

6654

dust foot ground

6655

dust collector ground

6656

vac hose ground

6657

display of outer grounding wire wrapped around hose

6658

[gc3]

What about the ground from the main panel to the actual grounding connection(rod)? Is your ground connected to your neutral bar in the main panel? If so have this corrected.




Gene Crain
www.plantasymaderas.com (http://www.plantasymaderas.com)

[beacon14]

If the dust hose is connected to ground at both ends - one end through the dust collector and the other end through the bot - you could be creating a ground loop. It's good that everything is grounded, but each item should only have one end attached to ground.

[tkovacs]

The above comments are both good.
It looks like you have done a exemplary job of grounding (and documenting it). Your feed rates are slow enough that you should not be overdriving the motors. My best guess is that you have a computer/SB controller communication issue. How long is your USB cable?

Terry

[pappybaynes]

The ground wire for the dust collection needs to go "through" the pipe, not around it.
Dick

[tkovacs]

Well - fwiw I run all my dust collection grounding by wrapping it around the pipe as Zeke did. My thinking is to keep the same ground potential at both ends.
Terry

[zeke]

Gene, the #6 ground is attached to the same lug as the main service ground wire, the neutrals are on the opposite side of the panel.
David, looking back at the "grounding your shopbot" instructions, I may have interpreted the instructions incorrectly on the hose. It says "If the dust collection hose leading away from the dust foot does not have a metal reinforcement wire in it, ADD one! Wrap a bare stranded wire around the hose, with each wrap about a hose diameter apart. Leave some extra at each end for connections." That said, I thought this meant to connect it at both ends. Now that you pointed out the loop, I'm going to take a closer look at the wiring and adjust as needed.
Terry, my usb cable is 10 feet long with a power boosted hub
Dick, I have heard that before and will give that a try also.

There are several other things I'm going to do to tweak the environment.

I read one post where the ground wire was removed from the Control box and the lost position went away. Any comments on this? This grounding was recommended by Shopbot.

Also, any thoughts about grounding to the frame, good idea or not?

I found a great site today blackviper.com on the web that gives you the configuration to scale back services on your workstation, I scaled the services way down today to try to eliminate my laptop as the problem. I'm going to try everything I can find to work through this and eliminate this losing zero problem. FYI, I just went and did a search and looks like Gary published this website last year. I've got several other items I need to look at as well. Thanks for all your feedback.

[beacon14]

the ground wire was removed from the Control box and the lost position went away

The control box is (or should be) grounded through the incoming electrical cable's ground. Adding a second ground might have created a loop. Removing the added ground to the control box would have eliminated the ground loop if in fact that was the problem.

[zeke]

- I installed Ferrite filters on every comm wire I could find and at both ends. I also have my power cables separated from my comm cables in the X and Y echains.
- I removed the ground wire wrapped around my dust collector hose and threaded it through the inside of the hose and is only grounded on one end.
- I eliminated the daisy chain ground wire to my control box and then onto the frame and ran the frame directly to the ground lug in my junction box.
- I increased my comm speed from 70.2 to 76.5, turned off all non-essential services. Operating like Apollo 13's return trip home now!

- Purchased a separate and powered usb hub.
- It is not totally clear to me about the ground looping scenario. I've seen instructions on grounding the Bot and included was the dust collector. If the dust collector is already grounded to the electrical plug that ties into the service panel, would adding another ground wire to the dust collector create a grounding loop similar to the Control box scenario?
- Still losing X and Y
- Today I ran a long cut and split the files up into multiple files and rebooted after each file. I'm concerned about the possibility of a memory leak corrupting my memory. Along the way I lost position a couple of times and then zeroed again. At the end of the cut, I did a clean cut, essentially not taking out any material at 1 ips and still lost x and y. There was some thought that maybe it was speed and feed related. I think this rules that out as well. I did not know but now assume that the wrong speed and feed not only affects the material and life of the bit but can cause a loss of position too, is that correct?
- Is losing position only prevalent on PRSstandards and not on Alpha's?
- I guess my next step is a more powerful computer. I currently have a Dell D600 laptop 1.6 mhz with 512 MB of RAM. Do you think this would work? An HP PC P4 3.0 GHZ WIN XP-2 PRO DESKTOP with 40GB Hard Drive, 1024MB of RAM. Anyone know what factors produce the highest comm speed? Is it all USB cards and CPU? Is there a way on a PC to run a comm test before purchasing it?
- Any other suggestions?

[signsbyjay]

First thing I would do is remove the powered hub, move the PC closer to the control box and run a direct USB cable.

[Gary Campbell]

Zeke...
Any chance that you did these changes one at a time so that you know which one, or two, fixed the problem?

Can you try another computer, or borrow a laptop?

On the dust collector question, you are partially correct. All of the metal components of the dust collector should be grounded thru the plug. This, however, does not apply to the dust collection hose, which in most cases is plastic and is a static generator. What you want to do is provide a path to ground for any static that is generated by air/dust moving thru the plastic hose right down to the dust foot. This is why the OEM foot has a gound wire attached. I prefer to route this back to the original bonding location. Hope this helps.
Gary

[beacon14]

- Is losing position only prevalent on PRSstandards and not on Alpha's?

No, but a Standard is much more likely to lose steps than an Alpha. Another thing I'd look for is binding somewhere. Drop the motors and see if the gantries roll smoothly on the rails, all the way from one end to the other. Make sure the motor is adjusted correctly but not too tight. And make sure the pinion gears on the motors are TIGHT - the wrench should bend when tightening them.

[kevin]

I had the same problem move usb cable to new port all is fine

[zeke]

Wanted to report back that it appears I now have reliable machine, whew! I have cut for approximately 20 hours without losing position. That is more than 4 times longer than I've ever gone without losing position, hopefully not a coincidence. I have made many enhancements to my machine to try to eliminate the problem. I finally ran out of items I could try on my end.

I contacted Technical Support and requested either a new main controller board or the whole controller. I received a new board, installed it and my losing position problem has not surfaced since. That is the only change I made for this attempt to solve the problem, I did not change the SB3 version, I'm still running v3.6.5. I guess I can't really claim victory perhaps until after 100 hours or so of cutting. However, I have never cut more than a couple of hours sometimes much less before losing position. What a journey this has been, good news / bad news. The bad news is this problem surfaced day one with my new machine and I've lost allot of quality time making products, broken bits, spoiled material losses! The good news is I now have an enhanced machine, allot more knowledge and it feels like I'm actually going to be able to produce some quality products after all.

OMG, to have a machine that is actually doing what it is supposed to do, I can't believe it!!! I hope I don't wake up from this great dream and I hope I don't eat these words!!! I wanted to communicate this out for others that may be having a similar losing position problem and have not been able to resolve it. I know there is a sea of issues that can cause this, however the controller board solved my issue, at least for now! Good luck to all of you trying to work through your machine-related quality issues, I know how painful, time consuming and costly it can be!

Happy Holidays all.....

[rb99]

Are the enhancements the grounding?

RIB

[curtiss]

Not sure of the full scope here but I drove a separate 4' copper clad rod into the ground and grounded everything to that. (in my case the ground is in the crawl space)

I understand that in some soils they say to use some rock salt around the copper rod to improve the ground.

[zeke]

Yes grounding was one of the enhancements, however that didn't solve my problem. Although that needed to be done anyway!