Problems with touch system for x, y, z - Probe tripped during non-probe move

  • mitakata_net
  • mitakata_net's Avatar Topic Author
  • Offline
  • New Member
  • New Member
More
27 May 2022 09:29 #243973 by mitakata_net
Hello,
Some time ago, with your help, I made a very simple touch system for x, y, z, which works great. I had to change the computer that runs my CNC router and I installed the latest version of Linux - LinuxCNC 2.8.2 Debian 10 Buster PREEMPT-RT ISO. Everything is set up as before on the old computer, but now the touchoff system does not work. When starting at different stages of execution, it stops and shows an error - Probe is activated during movement without a probe
I will be happy if you tell me where to look for the problem or what other information is needed to find out where the problem may be.

Please Log in or Create an account to join the conversation.

More
30 May 2022 00:51 #244170 by andypugh
It sounds like there might be electrical noise on the probe input, or possibly a mechanical vibration is triggering the probe.

Do you have any sort of software filtering on the probe input?

Please Log in or Create an account to join the conversation.

  • mitakata_net
  • mitakata_net's Avatar Topic Author
  • Offline
  • New Member
  • New Member
More
30 May 2022 15:20 #244222 by mitakata_net
The system, which is quite simple,
It is a metal plate, which is connected to the ground, and the other end of the clip, which I connect to the tool and through a cable goes to 13 pins on the parallel port and in contact finds the values of x, y and z.
The cable I use is shielded, and until a month ago everything worked perfectly, more than 2 years.

Please Log in or Create an account to join the conversation.

More
31 May 2022 09:15 #244268 by andypugh
I think that this is most likely a hardware difference between the old and new parallel ports.
(Possibly the old port was a 5V port and the new one is 3.3V)

Assuming that the touch-plate is used to pull-down the pin, try pulling it up with a resistor (1k?) at the parallel port end. This should make it a lot more noise tolerant.

(Or you could use the debounce HAL function to filter it in software. The probe input is only sampled in the servo thread, so filtering in the base thread has no timing penalty)

Please Log in or Create an account to join the conversation.

Time to create page: 0.062 seconds
Powered by Kunena Forum