Advanced Search

No reaction. Gmoccapy assumes that the spindle is running but nothing...

I have an onboard realtek and an intel pcie ethernet adapter. Which is best for stability with the Mesa ethernet control?

Onboard is realtek "rtl18168h-2"
PCIE is intel         "82573"

zz912 wrote:

If you want to make a control panel with many buttons, I recommend a combination:
7i96s + cable DB25M-IDC26 + 7i74 + up to 9x 7i84

That or 7i96S + DB25-IDC26 + 7i85 + up to 5X 7i84 (4x on 7i85 and 1 on 7i96S)
Of course, instead of 7i84, there are 7i70, 7i71, 7i72, 7i73 that can be used.

As MythBusters would say: well. there's your problem!
83 C is to much, although some go as high as 109C without issues, but some will start throttling at 63C, all this is from experience and i still have those PC's around.
So, this for sure warrants opening the PC, yanking out the CPU cooler, clean, repaste with thermal paste, mount the cooler back. Also a bit of dusting with high pressure air or using a brush might prove worthy.

No, not yet, i am just to damn busy, sorry for that.
I still think there is something not right with the wiring and all the jumps through the cabling, so if you are willing to investigate further, use a DVM set to measure resistance and check:
-if the G250 has the same gnd for step/dir and power
-if the signal gnd is the same with Mesa signal gnd
pretty sure the last one will not be 0 ohm, but do check that and report back, powered off when measuring.
If it does end up not being 0 ohm, use a 10-100 ohm resistor between the G250 gnd and Mesa signal gnd, then power on and test.

You wrote that you were considering a tool changer. Calculate how many inputs and outputs you will need. If you miss them, consider a combination:
7i96s + 7i84

If you want to make a control panel with many buttons, I recommend a combination:
7i96s + cable DB25M-IDC26 + 7i74 + up to 9x 7i84

Of course, you can start only with the 7i96s itself and buy other cards afterwards.

There is also the question of whether the closed loop motors you are using have an encoder output and whether you want to introduce it into the LCNC.

[qtpyvcp.app.launcher][ERROR] Could not load /home/mavijet/argecnc, make sure that the name or file path is correct. (launcher.py:134)

Seems that you are trying to load something that does not exist.

You are great - Thanks a lot.
I don't get the error any more.
The calculated offset is shown in the GUI (QtDragon_hd), depending on the position of the spindel.
But there mus be a nother mistake in my HAL file perhaps, because the offset has no affect to my Z-axis yet.
There is a indicator "EOFFSET LTD" should it be enabled?
I made two Screenshots, perhaps you have an idea whats wrong in my config.


     
 

About 30 seconds after LinuxCNC starts, the temps rise and the fan starts running hard. Since the temperatures go well over 60°C, I wonder whether the processor has protective functions at these temperatures and does not allow OpenGL to function at all? The question is because yesterday while researching in a video about the current ASUS PN50 I saw that the construction of the heat sink from the processor to the cooling unit is supposed to be a defect? The video shows that heat transfer cannot take place because the surfaces do not have optimal contact. What does this theory look like?

I guess I have lost everyone's interest or at least exhausted their list of suggestions.  I want to thank Tommylight and PCW for their help!  While the issue remains unresolved, they both helped me move forward in my understanding of LinuxCNC.

I have not given up on the Gecko G250X, but I must get this controller working ASAP.  As such, I have ordered DM556T stepper drives, which others seem to have working with LinuxCNC and Mesa boards.  Unfortunately, it will take a few days to get them.

I hope the DM556T will be more plug-and-play than the G250Xs seem to be.  Of course, this means I have to rearrange and rewire half of the control box.  Maybe I will get it right this time!  I still need to figure out outputs as they seem to be NC when I thought they would be NO. 

The config is more to do with allocating the FPGA resources.

There’s actually 2 drivers involved, the hostmot2 driver which handles the “heavy lifting” and the hm2_xxxx driver which handles the physical communication.

With regards to the SPI speed just pick a speed that is reliable. There’s no need to get hung up on it.

It’s amazing what one learns when the docs are read. It’s also useful to read the manual for the card you are using as well.

Are you really using full steps (Microsteps=1)?

That would typically be quite rough and
susceptible to resonance.
 

Interesting, was unaware that drivers could be "shimmed" on-the-fly this way.

Ok so it looks like, to experiment with LinuxCNC SPI speed on Rpi4, something like this could be done to the .hal file:

loadrt hm2_rpspi config="num_encoders=0 num_pwmgens=0 num_stepgens=3" spiclk_rate=31250 spiclk_rate_rd=12500

Where the speeds are out of:
- 62500 - 62.500 MHz
- 41667 - 41.667 MHz
- 31250 - 31.250 MHz
- 25000 - 25.000 MHz
- 20834 - 20.833 MHz
- 17858 - 17.857 MHz
- 15625 - 15.625 MHz
- 13889 - 13.889 MHz
- 12500 - 12.500 MHz
- 11364 - 11.364 MHz
- 10417 - 10.417 MHz
- 9616 - 9.615 MHz

Yes, if I can avoid it, I don't want to have to buy a new PC straight away. Of course I'm happy, although I can still hope for the great support ;o) Now the temperatures follow. To do this, I opened and closed the LinuxCNC several times, 4 times slowly in a row. The error messages always remain the same

cnc@cnc:~$ sensors
k10temp-pci-00c3
Adapter: PCI adapter
Tctl:         +56.8°C  

nvme-pci-0400
Adapter: PCI adapter
Composite:    +46.9°C  (low  =  -5.2°C, high = +79.8°C)
                       (crit = +84.8°C)

iwlwifi_1-virtual-0
Adapter: Virtual device
temp1:        +53.0°C  

amdgpu-pci-0500
Adapter: PCI adapter
vddgfx:        1.33 V  
vddnb:       681.00 mV
edge:         +53.0°C  
PPT:           8.00 W  

cnc@cnc:~$ sensors
k10temp-pci-00c3
Adapter: PCI adapter
Tctl:         +83.2°C  

nvme-pci-0400
Adapter: PCI adapter
Composite:    +47.9°C  (low  =  -5.2°C, high = +79.8°C)
                       (crit = +84.8°C)

iwlwifi_1-virtual-0
Adapter: Virtual device
temp1:        +56.0°C  

amdgpu-pci-0500
Adapter: PCI adapter
vddgfx:        1.26 V  
vddnb:       774.00 mV
edge:         +74.0°C  
PPT:          16.00 W  

cnc@cnc:~$ sensors
k10temp-pci-00c3
Adapter: PCI adapter
Tctl:         +78.2°C  

nvme-pci-0400
Adapter: PCI adapter
Composite:    +47.9°C  (low  =  -5.2°C, high = +79.8°C)
                       (crit = +84.8°C)

iwlwifi_1-virtual-0
Adapter: Virtual device
temp1:        +56.0°C  

amdgpu-pci-0500
Adapter: PCI adapter
vddgfx:        1.22 V  
vddnb:       774.00 mV
edge:         +70.0°C  
PPT:          15.00 W  

cnc@cnc:~$