Advanced Search

I just upgraded my system to the new linux/CommandCNC and am currently trying to configure/debug it. My A axis motor, which runs the far side Y axis stepper won't run, It is selected in the configurator and it is checked for reverse direction. None of the polarity boxes are checked. I pulled the plug out and tested the resistance as in the manual - I get the result shown, about 2 ohms on either side-by-side test of the outer two plugs and infinite resistance on alternate ones (1st and 3rd or 2nd and 4th). I swapped the two plugs that go to the Y axis so the one that moved was plugged into the socket on the bladerunner box for the one that didn't move and vice versa. I still get no movement on the far side of the table. The last thing I tried was checking the stepper windings with an led, Curiously, the motors that work don't light the led, the motor that isn't working does (both windings light when tested). Seems like it must be the motor but not sure what is wrong with it - anybody have an idea?

A silly diversion while I wait for things.
I saw how an arm atc mechanism works, and thought "how hard can it be?"

Turns out... quite hard. haha. I found it basically impossible to designs a cam path in fusion. It might be doable somehow,. but it could not think of a way. So, instead I dumped it into my animation software (3dsmax) and animated the motion as I would like it to be. I then basically drove the rollers into the drum wheel using a voxel based volume tool and it ate out the paths (not unlike a cam simulator). Neat trick, but its not super precision because the voxels are only 0.1mm. I think it might still be worth doing some 1/2 scale resin test prints to see just how much play is tolerable in the system. The end goal would be to make most of the mechanism in SLS nylon.

Sorry. The most I got was "at some point". It is not any sort of priority to bundle it up. I expect there is at this point a real mix of things that "must be" returned and things they very much do not want to give out and it actually takes work to separate it. This is after all the first PP/linuxcnc variant that could be considered properly industrial worthy.

You will be able to get the installer for $30 as normal, but in line with what's said before, it is less and less something that can be made to work on a non tormach 1500 machine. (I asked if it made sense to tryin it on my machine and the answer was basically not unless I buy all the same electrical parts they use).

"Anyone have a contact at Tormach - even just to ask about the trajectory planner and some other key improvements"

They are probably reading this.

So finally on my holiday break and looking to try and connect my 40w Diode laser to my machine

want to get some guidance before i flash my 7i96s and cause my whole CNC to stop working.

would like to be able to use the spindle control, whether that is scalled or as a percentage to adjust the 0-5v for the intensity of the laser

so i understand that i have to flash the board with 7i96s_2pwmd.zip, which will add PWMGen 1 to STEP4+/STEP4- 

Is how i have the diagram, would this be the correct way i need to wire to the board?

also is this the correct coding i need to add into my hal/ini file?

(HAL)

# Enable PWM generator on channel 0
etp hm2_7i96s.0.pwmgen.00.enable true

# Set the PWM frequency (adjust based on laser specs)
setp hm2_7i96s.0.pwmgen.00.scale 5000 # Example: Scale to 5000 RPM for 0-5V lasers

# Link PWM output to motion spindle speed command
net laser-power motion.spindle-speed-out => hm2_7i96s.0.pwmgen.00.value

# Link PWM enable to motion spindle enable
net laser-enable motion.spindle-on => hm2_7i96s.0.pwmgen.00.enable

# Link motion spindle-on signal to a GPIO pin
net laser-on motion.spindle-on => hm2_7i96s.0.gpio.000.out

(INI)

[SPINDLE]
SPINDLE_TYPE = PWM
MAX_SPINDLE_SPEED = 5000
# Match the scale in HAL MIN_SPINDLE_SPEED = 0

Hi, is it maybe possible that you have save old boards from original control unit and if you are willing to sell them? Thanks.

github.com/ozzyrob/Linuxcnc-FPGA

ihavenofish wrote:

This is just 4th hand information though, and I know nothing about coding so take that for what it's worth
 

Any idea, even 4th hand, what it would take to get hold of the PP source code?

I assume it's still 'open source' as it's a derivative work, but I suspect it'd be counter productive - much less a bit rude - to demand anything.


Anyone have a contact at Tormach - even just to ask about the trajectory planner and some other key improvements?

You can link to this project.

After using a USB-blaster and a ft2232 board, Ive found the ft2232 board to be the more reliable.
After messing around with the colorlight boards I found they needed quite a bit of rework for use.
So I went with a Spartan 6 dev board and some custom boards and I was able to use Mesa 7c81 firmware built to suit.

I just ordered this board and am gathering information. Has anyone tried flashing it with the USB-Altera-Blaster STM-32?

No gigabit is not required, the mesa boards do not support this speed.
As no based thread is required jitter is not a major issue, under 100000ns will work.

Stop overthinking.
With mesa hardware you don't need a base thread, only a servo thread is required. With that, latency isn't such a huge issue either. Trying to get an extra bit here and there does nothing.
The pre made image Rodw recently linked to is perfectly fine to run a machine, the only exception would be if you are using the RPi 5 spi interface to talk to a 7c81 or 7c80 as this requires a driver that isn't included in Rods image , this is not required for ethernet.
Invest more into learning the ins and outs of hal . This is more important that fussing over software as the images provide what is required, remember the only task these images are meant for is running a CNC machine.
If one reads the online docs one will fine some simple exercises that do not require a CNC machine.

Dear jjdege.
How can you solve the error when you install the ethercat package by command "sudo apt install -y linux-headers-(uname -r) ethercat-master linuxcnc-ethercat?

I think we would have to add the motion color into the plot point.
(qt_mismach.py line 287 ish)

Dear jjdege.
How can you solve the error when you install the ethercat package by command "sudo apt install -y linux-headers-(uname -r) ethercat-master linuxcnc-ethercat?