[MichaelHenry]

On occasion, perhaps once every 50-100 (manual) tool changes, the head on my Series 1/II PCNC 1100 will start to do a slow drop right after the tool change is called for. Pressing the ESC key stops the drop, but the Z-axis zero is messed up and can only be reset correctly by restarting Mach3. I haven't sorted out the details yet, but if Z is re-zeroed after the drop and without a restart, the tool change position is off quite a bit from what I normally use. Normal is Z-3 and it seems like the "new" tool change Z is much lower. That causes problems with long tools, like reamers in a large drill chuck.

The problem seems to be worse if I hit the PDB button before the spindle has stopped retracting to the Z tool change position, but that could be my imagination.

I'll call Tormach tomorrow, but has anyone else seen this sort of problem or, better yet, have a solution or a trouble shooting suggestion?

Mike

[john2532]

Do you think the Z-brake might be the problem? When I had a similar issue the guys at Tormach recommended placing a coin on top of the Z-brake lever and watch to see if it shook the coin off when the brake activated (theory is the lever moves so slightly that it is hard to see otherwise). My issue was that my lever was completely out of the operating detent, most likely from when I hoisted it on the stand and the lifting strap pushed the lever aside.

Most likely not the same issue but I thought I would throw it out there anyway.

John

[GLCarlson]

+1 on the Z brake lever. Check it anytime the head drops. Very easy to miss that it's out of position.

[MichaelHenry]

It's not the Z-axis brake lever as my Series I/II mill doesn't have one.

It happened again this afternoon and the drop seems to happen just after Z reaches the TC height following a Tool Change call. This only happens once in 50-100 tool changes so it will probably be hard to diagnose. My old monitor crapped out a few weeks ago and I initially had some USB issues (but none recently) so perhaps there is a clue there. I'm temporarily using a USB hub for the keyboard, wireless mouse transmitter, and the thumb drive used for program transfers and am having no problems with those. My Tormach Shuttle pendant is on a USB extension cable that is connected directly one of the PCs USB ports.

I also recently started using G-code produced by SprutCAM 8, though it's hard to see how that might cause an intermittent problem.

My PC is an old Dell 8200 Dimension and I am starting to wonder if it might be time to replace it.

Mike

[mike sr1]

I would check the lead connections to the stepper motor, for the head to drop I would think the whole motor would have to be de-energized and I thought at least one coil has power on it at all times even at idle. It would be worth checking anyway especially if its intermittent.

[tbaker2500]

When it is dropping, does it sound powered? The same as if you called for a decrease in Z? Or is it a freewheeling motor?

[TurboStep]

... I'm temporarily using a USB hub ...

I also used a USB hub temporarily and got some involuntary movements while jogging. I realized that they began at around the same time that I added the hub so I removed it, and haven't had any problems since. I now connect all devices (mouse, keyboard and Shuttle) directly to the PC and remove the memory stick as soon as I've copied the data.
For what it's worth, the brake (if you had one ) is held off when the mill is powered on and is applied only when power is removed.
Step

[Khalid]

Hub may be the problem. Just direct connect to the PC. The USB cable must be separaed from the power cable. Tell is this happen when the Spindle stop near the TC area?

[MichaelHenry]

When it is dropping, does it sound powered? The same as if you called for a decrease in Z? Or is it a freewheeling motor?

Hard to tell for sure - it's a fairly slow descent (nothing like a rapid), but my 1st thought is getting it stopped before the cutter runs into the workpiece so everything else is on my brain's back burner. It does sound a little like it is powered, but not the normal sort of axis movement noise.

Tormach suggested that I swap out my wireless trackball for a wired mouse and that I eliminate my USB hub. I'll try that along with another idea or or two over the next few days. At least it doesn't seem to be getting any more frequent.

Mike

[MichaelHenry]

Hub may be the problem. Just direct connect to the PC. The USB cable must be separaed from the power cable. Tell is this happen when the Spindle stop near the TC area?

Yes, it seems to happen right after Z retracts to the tool change location (Z-3 for me) but only once every 50-100 tool changes. I' started running the USB Sniffer utility that Tormach has on their web site and no problems there so far.

Mike

[WOTDesigns]

Sounds like interference to me. I had the almost the exact same issue when we first set up my 1100. ONLY happened after Z+ movement with a spindle power down at roughly the same time. I trouble shot it with tormach and wiped the pc, re installed everything... Still happened every now and again. Completely random and on known good programs from my other machine.

Ended up being the routing of the main power cable. It was too close to the control cable and the massive spike from Z+ movement and spindle brake/stop together would send enough (bad magic juju) to tell the Z to drop on the damn part it just finished cutting. Not powered, not rapid, but usually faster than I could react. First time it happened I destroyed my new touch screen diving for the (terribly located) e stop. That day the e stop relocation began ...

Brian
WOT Designs

[MichaelHenry]

Thanks Brian. I'll look into cable placement if none of the other approaches bears fruit. None of the cables has been re-routed or even touched in the past few months and the mill has been used pretty much daily in that time period with the problem only recently surfacing, so it seems unlikely to me but one never knows.

[WOTDesigns]

Definitely unlikely if it's a new issue, but it's the easiest to check =)

Brian
WOT Designs

[Khalid]

Can you replace the USB cable. Sometime it kinks and problematic at some specific position and this happen randomly.

[MichaelHenry]

Definitely unlikely if it's a new issue, but it's the easiest to check =)

Is that kind of like looking for your car keys under the street lamp because the light is so much better there?

Seriously though, I am a big fan of checking the really easy stuff first, even if it doesn't usually bear fruit.

- - - Updated - - -

Can you replace the USB cable. Sometime it kinks and problematic at some specific position and this happen randomly.

That's a possibility - I did re-route one of the USB cables recently.

[WOTDesigns]

LOL. it's exactly like looking for your keys under the street light first. I would feel pretty dumb spending 9 hours looking around in the dark and THEN finding them right in the damn light I didn't spend 2 seconds to check good analogy

Brian
WOT Designs

[MichaelHenry]

Looks like my Z-axis problem is related to the USB hub or the cable from the PC to the hub. I ran the USB Sniffer program that Tormach has available on their download page and it reports that both the keyboard and trackball ports are briefly disconnected and then reconnected a few seconds later each time the Z-axis goes into semi-freefall right at the end of the toolchange routine. It's very random and only occurs about once every 50-100 tool changes.

I have an original Series I, upgraded to Series II, with the keyboard/monitor arm on the left of the mill and have the computer in the base compartment on the right hand side. Consequently there is a pretty long run from the computer USB ports to the keyboard and trackball, long enough that I needed an extension USB cable to make the length. Rather than run 2 extended length cables, one each for the trackball and keyboard, I elected to install a wall wart-powered USB hub at the end of the extended USB cable. Since USB Sniffer reports both trackball and keyboard ports being disconnected and reconnected each time there is a Z-axis problem, I am assuming that there is either some noise near the system that is goofing up the USB hub or else there is a problem at the connection between the end of the USB cable from the PC and the extension cable connector. The PC is a Dell Dimension 8200 that is probably > 10 years old so I suppose there is also a chance that the PC USB port is starting to go flaky.

I'm just not sure why this should only happen at the very end of the Mach3 tool change macro, so if anyone has any ideas on that please pipe up.

In the near future I will be installing a new Tormach controller and figuring out a way to route everything so that the USB hub is not needed and try to minimize or eliminate the need for the USB extension cables. In the meantime it seemed a good idea to document the findings so far. This would have been much tougher to figure out without the USB Sniffer program.

Mike

[tbaker2500]

Nice work!
My experience with USB has shown that noise is the main reason for the connection to be reset. But the quality of the cable can make a HUGE difference. It's usually hidden, but when using a USB packet analyzer it's perfectly clear. A long run cable with poor shielding, and, poor quality control on the number of twists per inch will pick up the noise.
As to why it happens just then, it's could be that a specific pattern of noise is emitted at that operation which causes the fatal error.