Finally decided to get round to setting up probing on my TinyG V9.

Controllers g2core and TinyG
1

Finally decided to get round to setting up probing on my TinyG V9. Followed various advice from various people and still haven’t got any further.

My current setup is as follows:

Touch plate connected to DI5 (zmin).
470ohm resistor from DI5 to 3.3v
100nf capacity between DI5 and Ground.
Other side of touchplate connected to Ground.

Here’s the setup of my $di5 input:
$di5mo=0 (Active Low)
$di5fn=1 (Limit)
$di5ac=1 (Stop)

Build and firmware version:
[fb] firmware build 83.09
[fbs] firmware build " 083.09-116-gf418"
[fv] firmware version 0.98

I hook up the clip from the touchplate to the bit, and put the touchplate below the bit and start the probe command.

Pretty reliably now the head will move down towards the plate, and will touch it, then stop, then proceed to continue moving into the plate. Not good.

So I’ve setup a little test with some foam underneath the touchplate to let me avoid destroying my setup. I’ve discovered that when the bit hits the probe the first time it pauses for a very short moment, then proceeds in the same direction. If I move the plate away from the bit (opening the switch) after the first contact and close the circuit again by bringing the plate back into contact with the bit then the end of the probing is properly detected.

Now I know @sszafran you were going through similar issues before, what was the eventual solution? @Riley_Porter_ril3y any suggestions?

2

Looks like the di5 input state ($in5) is occasionally getting stuck high, not quite sure why. When I manually close the switch again (touch the plate to the bit) the state reverts to what it should be.

I tried testing the voltage between the bit and the plate when it was still being registered as ‘on’ (i.e. $in5=1) and it was measuring 3.3v which should indicate the the switch is open (since we’ve set active_low on the $di5 config).

Guess i need to get a scope.

3

Hi guys. I have recently seen what you are seeing. I’m working on a fix. I believe it has to do with how we’re doing debouncing. I’ll report in the group (and on the github issues at https://github.com/synthetos/g2/issues/126) once I push a fix.

4

Yeah it looks like a debounce issue, I’ve done some more testing this morning and can reliably reproduce it without moving the machine and just touching the probe and clip together. It’ll happen about 30% of the time or so and stick on, then release on the next touch. Let me know if there’s anything I can do to help, happy to test out builds.