Advanced Search

Still looking for a freelancer to help get a tool sensor setup in software

Thanks guys. I am collecting up the wisdom of the crowd, and will try to work through everything systematically.

Doesnt work in sim either

That bug should be resolved in recent builds of the framework.  Its not a monokrom bug. It came out due to an errant print statement.

"What's the first rule?  Do NOT use print statements in the code.  Use LOG.debug"

I'm talking about the PWM value. So this value is a voltage value? Like 17V? I know how to limit this value. But I never measure this value on the Mesa. It's always only 0.35V as soon as I enable it.

The output voltage should be PWMvalue/PWMscale

How do you setup a tool sensor exactly? I cant seem get it right after a week of trial and error of setups, parameters changes. This is not a wiring issue or a debounce issue but a setup issue.

Current Setup:
LinuxCNC 2.9.4
2" tool sensor on the bed
Attached is my setup files

From this document  linuxcnc.org/docs/stable/html/gui/gmocca...uto-tool-measurement  I cant tell if I am doing something wrong or where the discrepancy is.

This example setup would cause the Z axis to come down to get to the -20 setpoint. This results in a crash into the probe and a "Probe tripped during a non-probe move" would appear. The solution to this was to set the Z to 0 so the probing search would start immediately

For this next issue is touch off x and y (not z because the document says to press "Block Height") I immediatly get a error when pressing x or y touch off. 

"G-code error in change.ngc near line 76, file ended with no percent sign(5) or program end (m2)"

Now if I just ignore these errors I get the same error when doing a tool change after it probes it. 
 

@rodw Thanks for the guide, it really helped me! And some Claude assistance

joshualee777,

Not sure if my Quick Reply was successful, so sorry for the duplicate post.

Hi! I've just successfully completed my initial test setup with a single JMC IHSV57. See github.com/rokoter/LinuxCNC/tree/main/ethercat/ihsv-homing. Hope it helps!

Launch LinuxCNC with cia402-ihsv-home-variant.ini
The Z startup error can safely be ignored

Yes, the value comes from the PID. My power amplifier is a -10/+10V. Is the value a voltage value? If the value is, for example, 17, I measure at the Mesa only 0.35V.

Hi! I've just successfully completed my initial test setup with a single JMC IHSV57. See github.com/rokoter/LinuxCNC/tree/main/ethercat/ihsv-homing. Hope it helps!

Launch LinuxCNC with cia402-ihsv-home-variant.ini
The Z startup error can safely be ignored

Yes, in the hal file, the analog outputs are all controlled by PWM values.

The actual PWM value shoudl be set by the output of a PID component that
in turn reacts to the difference in the encoder position to LinuxCNCs
commanded position.

It wont let you change the mode, its gets locked out. I gave up on getting probe basic to work

I've now loaded the standard 7i97t ini and hal files, and after a few minor adjustments, LinuxCNC is working (name, IP). In the hal file, I've only commented out the encoders and end positions for now, as they aren't connected yet. Is it even correct that the analog outputs are displayed as PWM? The only difference is that the value is set to higher values ​​(e.g., 17) when using the LinuxCNC interface.

@suremetall Yeah, that's what my thoughts are exactly. I could keep messing around with the original hardware, waiting for one thing after another to fail, but once this retrofit is done I should be able to run it for years to come.

The biggest challenge for me at this moment is the motor adapter for the z-axis (ram), which doesn't have a lot of space. 

I already created a design, but I'm having some issues with the tolerances and manufacturability I can't really get my head around.