Advanced Search

Ok myself and my partner are totally new to LinuxCNC having worked with Mach3/4 in the past. The problem we are having is that when we try to run a job we get error messages stating "cant issue MDI command when not homed.We first load our GCode and home the machine (front left of machine as your looking at it), we then move to the spoiler board just in front of the stock lower the Z axis to the spoiler, hit the touch off button put in -25.0 as our stock wood is 25mm thick, we then raise the Z axis and place it on the stock with the Z resting on the wood we then hit touch off for the Y and X axis leaving it at 0.0 and we use P1 G54 on all three axis. When doing the Touch off screens each time it sends up this error message, when we try and run the programme we then get the message "Program exceeds machine maximum on X, Y and Z. I'm hoping someone can enlighten us as to why this is happening is it something in the stepconf or INI file we have overlooked.
Many thanks in advance.

tommylight wrote:

Origin should be at bottom left, or near left when looking at the machine.
Not a rule, but most are set that way.

With Fusion 360, it does not matter. I always touch off at top left even though the machine origin is front left.
Thats how I did it on a manual mill and its more repeatable against the fixed vice jaw.

You need an F truck

Chris,

I just added the ability to create a spin box and it's a HAL float pin. The preset is loaded into the HAL pin at startup. Then in your G code you and read the HAL values with #<_hal[name.of.pin]>

Hit me up on the IRC if you want more info on this.

JT

Origin should be at bottom left, or near left when looking at the machine.
Not a rule, but most are set that way.

I do not have any torch with shield, so i can not test although i do have 2 of THCAD for now.
I have to go pick up a Hypertherm 3070 as i do have it's torch and cables in my shop, but the transformer was to big to fit in a Skoda SuperB!

Glad it worked out and thank you for reporting back.

What gui? sounds like you have enabled a manual tool change for premeasured tools.

Todd wrote:

I don't think that is the correct link to the Linuxcnc Buildbot.
The one here.
buildbot.linuxcnc.org/dists/
shows a last build date of 9/15/2024

But it only builds to buster and not for later distros. You need this one
buildbot2.highlab.com/
I thought it was fixed today.
There is a PR that I have tested that builds debs the Debian way, Once it is pushed, you can build debs in a few commmands and Debian installs all of the dependencies for you. If I wanted master, this would be the way to go once pushed, It does not take long on an i7 with 20 cores!
github.com/LinuxCNC/linuxcnc/pull/2647

There has been a lot of improvements in Flex GUI the latest is HAL spin boxes. If you preset a value in the spin box when you start the configuration the HAL pin values are updated.

A couple of screen shots of probing





JT

Tommy, We use a 24k 5W resistor as suggested by PCW a long time ago.
You need to calibrate the adjustable voltage on the power supply to be 24 volts.
And you need to monitor the ohmic voltage when configuring it and ensure the ohmic threshold is above the off voltage when the torch is wet.
Initially, Steve had this set too low even though it worked for me.
I was going to ask Steve to try a threshold of 24.9 volts to see how reliable it is as an experiment. I have always seen a full 24 volts on contact.
There are so many advantages of using a THCAD for Ohmic sensing, it would be good if a few like minded people could experiment further.
 

Thank you, Tommy, happy to say that removing the base period and adjusting the FERROR values to 0.1-1.0 has yielded results. The machine has been running my very long test program for quite a few hours now with no issues yet. I will continue testing and gradually increasing the acceleration and velocity.

Hopefully this will be the end of this saga, I will report again after a day or three of thorough testing.

for sure debian SO have as usual some limitations and problem. Think it's worth doing some tests.... I'll do them as soon as I have some time. For now I'm busy understanding the "local" file structure of Debian 12....it seems to be very different from the previous ones....and it's giving me problems....

i did find a dip switch setting that made a difference. SW8 is a pulse filter on is 10ms off is 1.5ms mine were off. After i turned that on the set point signal smoothed out a lot.

one other thing i found was the control loop sets based on the peak current rotary switch. and even if you want to change those the drive will default back when the power is cycled. i did have mine set at 6 but 9 is a better choice based on what the manual is saying. 9 should tighten up the position error signal.

i think the drive knows the power supply voltage and sets your current based on supply voltage. @ 36V i get 4,8A, the drive has a max of 7A but mine only go to 6A with the 36V supply.

no idea if it is up to date:
github.com/itain/Tango-PCB-Kicad