Probe occasionaly fails to stop and crashes into part

[capnnick]

I've had machstdmill pro for close to a year now and I've watched my machine try to drive my wildhorse probe through my part and into the table several times now and I cant figure out why it keeps happening. My mill and control box is the first I've built so I was pretty sure the cause was something I had wired improperly or some kind of grounding problem but I'm at a loss as to how to correct whatever the cause is. Since it first occurred I've completely rebuilt and wired everything and hadn't had any trouble for awhile but yesterday I tried to do a corner probing op like I had several times that day and it down once at regular speed in the Z direction, backed off, began its second z probe at slow speed and decided not to stop going which put a nice dimple into the surface of my part and stalled my z axis before I could hit the emergency stop.

Cant figure out a reason hardware wise that Mach or MSM would think that the probe had failed to trip when it successfully triggers on the first touch off and then proceeds on the second touch off to push the tip straight up into the probe body until it bottoms out. By that point not only is one contact broken and mach should be seeing the signal, but they all are. The probe is simply wired with one contact going to the 5v on my breakout board and the other going to the pin I've assigned for the probe. Have you heard of this happening or have any idea what might cause it, hardware or software wise?

Things I've tried: Removing the LED and resistor in the probe, replacing the probe cable, and installing a smooth stepper to get away from my flaky XP machine and install a fresh copy of mach on my windows 7 64 bit pc.

[DaveCVI]

Hi,

The probe is simply wired with one contact going to the 5v on my breakout board and the other going to the pin I've assigned for the probe. Have you heard of this happening or have any idea what might cause it, hardware or software wise?

I suspect that this is the root cause of the problem. The probe does not go between +5 and the BoB input pin. The positive probe cable connection should go to the BoB input pin and the Gnd connection should go to ground. Then you need to add a pull up resister from the BoB input pin to +5v.
For detailed information please see this sticky post:
http://www.calypsoventures.com/forums/v ... p?f=6&t=38

With the probe wired as you described, when the probe is "triggered" (I.e. one or more switches are open) the input to the BoB is not being driven - it is left in a state called "floating". That can (and apparently has been) causing trigger events to be missed.

Dave

[capnnick]

Thanks so much. I rewired my probe as you suggested and have been using it everyday since without any surprises. On a side note I'm really liking the ability to use a touch pad and probe on different pins in your new beta release. It's been working great and since I didn't have any luck when I tried to build a circuit to combine the two signals, its nice to have an option that finally opened up automatic tool length measurements for me. I have had a couple of art codes thrown out that mach couldn't recover from, usually after some inactivity, but the probing and tool length measuring cycles have all been rock solid. Next time I get one I'll record the message and send it your way for bug squashing.

[DaveCVI]

Great! I'm happy the rewire fixed the problem. It's also good to hear that the two signal beta release is working for you. I now have several reports of it working fine and no bug reports for the new feature - I like that combination.
Dave