Probe troubleshooting tips

[Jeff_Birt]

About a month ago Ted Grinnel posted the following on the awwcw.com forum as he was having trouble posting here. I asked for his permission to copy it over here but forgot until I had probe problems tonight. First I'll post Teds tip and them my own...

Ted Grinnel:
I put my machine back together today after a Z-motor problem and started scanning a rather large project (12 hour). The machine stopped after 15 minutes and posted "Check Probe". Each time the machine would continue scanning after I pushed the enter button.
This reaccured 4 times before I decided it was not just a glitch. I checked the ribbon cable and plug connector, no help - another 3 occurances. At this rate it was going to take me 2 days to complete this scan.
I decided to find out what this probe was made of. I took it apart and found that the contacts were dirty. Imagine that 700 hours of scan and the contacts are already dirty - go figure.
I cleaned them and recoated them with di-electric compound. The project is still running 8 hours later.
By the way di-electric compound is a silicone grease that is used on electric connection to keep water out. You can pick some up at your auto parts house, used to be used on points in distributors, not used on all electric connectors in you car.

I was getting a Z-stall error a few minutes into the scan. Looking into the machine the probe looked like it was stuck at and angle. Testing with the sensor test menu revealed it seemed to be working though. Fllowing Ted's lead I removed the three screws holding the probe together. What I found is that one of the copper contact bars had popped loose from the plastic where it had been heat staked down. I plugged in the soldering iron and staked it back down. Cleaned off the old dialectric grease and applied a little new stuff. It's been scanning happily for two hours now

[pastime106]

Jeff, Thank you for posting this, my probe has been doing pretty good, but every now and then I get a little glitch with the z-stall. I will be sure to take a look under the hood.

Thanks!
Tom

[Gman_Ind]

My brand new probe was dirty inside from day one, the spring was a bit off as well. There was loose plastic flash inside that might have eventually lodged in between one of the contacts disturbing a scan,
There are three contacts wired in series inside the probe. A continuity tester will allow us to bench test this to make sure it is working properly. The probe should have continuity in the normal state and will break continuity if it is deflected.

[benluz]

After scanning several items with no problems,all of a sudden the "Check probe" message kept showing up.Took the probe apart ,cleaned ,checked it out with the ohm meter and all checked out fine,but the error still kept occurring The only thing that was different is that I had a new sled for the larger object I was scanning.I reach down and moved the side adjustable stop ,think it might be binding.Well it finished the scan with out another problem a total of two hours,where as before it might have went 5 minutes at most.I've noticed with masking tape getting caught on one of the rails and several other little things that the on board computer is very sensitive to increased resistance in any of the moving parts,which is good, it keeps the machine from damaging itself,the product and a whole host of other things.I just wish the computer would translate these into a more specific error messages for the end user to trouble shoot.I think the second generation carvewright will probably be one of the most popular cnc type machines ever made.
Ben

[Jeff_Birt]

This weekend while setting up a scan I had to stop the machine just after it had moved the probe down and started scanning; I aborted the scan with the probe down. When the machine went to home again after I tried to restart it it did not raise the probe, the Y-axis just came towards the front of the machine and ran the probe into the side of my scanning sled! So, just make sure you have the Z-axis all the way up before lettign it home with the probe.