Advanced Search

That kind of math isn't something i have worked with in years and years and years lol. My broader point was that, maybe it is worth looking to build a NN that's not super big , but it might be faster and more efficient. Particularly if it can be run in using a GPU compute model , but it needs to be a very small and tight NN, it can't be a huge model. Just a narrow implementation that can handle the calculations from the weights.

Yeah I don't know what they are up to.

Does it matter if I connect the limits switches to the main board 7i96s of to the I/o 7i84?

Is there any delay in communication between the 7i96s since it’s a serial connection?

Or am I overthinking this?

Well my machine not that big. Lol
But good to know.

If you just want to turn something you can delete lines N32 and N33.
It's the Fusion360 preprocessor that adds those two lines, don't know if there is a setting for the two extra axis U and W.

50 to 100M should be fine

We originally tested at 1000 ft (~305M)
 

Is there a max limit for the serial com cable on a 7i84 can be from say a 7i96s?
 

Hey, nevermind. I just found tommylight's explanation of installing on Mint Debian. This sounds much easier and more within my (lack of) Linux skills.

The machine was unattended (briefly) as I moved around the workshop.

I admit that, if the machine was in the Pause state, I may not have noticed it.
Better error/status reporting next time.

I use standard USB mouse and keyboard.
This fault could indicate some EMI effect on one of those connections.

I found in the manual:
- gmoccapy.program.current-line (s32 OUT) - Indicates the current working line of the program.
Need then to generate a display, or memory location, with a one event memory.

Thanks for the comments,
John.

Hi

Thanks

The 9V battery is for the laser pointer, while waiting for it to be connected to an output of the USC controller board (the most tedious part is running the cable through the cable track).

1. Nice build and nice cuts
2. why is there a hanging 9V battery on the torch?

beauxnez wrote:

MOTION: num_joints is 33, must be between 1 and 16

33 ?
Should probably be 3 for normal machines.

Weis ich nicht, hab nie EtherCAT nutzt.

Thanks for digging a bit more.

I've been racking my brain and reading as many examples of functioning orient examples as possible and had a thought:

My INI file has ORIENT_OFFSET commented out.  Figured it would default to '0' if left out, since I haven't decided where I want the spindle zero position to be (based on chuck key position or taper features).

Now I'm wondering if that's causing orient to misbehave.  Soon as I get back to the lathe this evening I'll uncomment it and see what happens.

Fingers crossed.

EDIT
No joy, same behavior with ORIENT_OFFSET set to 0.

Now I wonder if orient isn't working right as I have two instances of it at the same time (main and sub-spindle).  Have to test that theory next, if it's not really plausible.

Hello everyone,

I built a plasma cutting table (2500x1250mm) three years ago for my business needs, and so far, it has given me great satisfaction.For the past few weeks, I have been upgrading it because, until now, it was running in a limited way due to Mach3.
Having had some experience with LinuxCNC (EMC) several years ago, I decided to convert it using QtPlasmaC.

QtPlasmaC has allowed me, among other things, to:

• Have a real Z probing (precise and configurable)
• Implement proper plasma cutting management
• Use THC
• ...

Hardware used for this table:

• Industrial PC "Hystou" (  link  )
• 17" industrial touchscreen monitor (4:3)
• USC Board from PICO Systems (  link  )
• 4x "STEPPERONLINE" DM860T_V3.0 drivers (  link  )
• 4x stepper motors with encoders
• X and Z axis driven by ball screws (5mm pitch)
• Y axis driven by belt
• Plasma source: Hypertherm Powermax 105 SYNC

I am reaching out to share the THC system I have developed and tested over the past few days (not without some difficulties).
I am using an analog input of an Arduino UNO to measure the arc voltage, which is then sent to the PC via "USB".
A Python script retrieves the value in "real time" and creates a HAL pin that can be connected to QtPlasmaC.

I am using:

• An Arduino UNO
• An isolator/converter (  link  )

The Powermax voltage divider is set to 20:1 and connected to the input of the converter (configured for 0-20V). The converter output (set to 0-5V) is connected to the Arduino's analog input.
The Arduino sends a smoothed value every 100ms and is particularly stable.
This allows me to configure QtPlasmaC in Mode 1 (using Arc OK from the Powermax) and let QtPlasmaC precisely control the torch height.

Here are the first tests of the THC, with a sheet placed on spacers:
 

Position of the sheet during cut:


First produced part:


Another improvement I made: a cross-shaped laser pointer:


Now, I still need to fine-tune the settings, especially the axes, which are causing me a lot of trouble and failures due to the "following error.", connect the laser pointer to the controller, then, I need to restore the electrical cabinet.I will send you photos of the whole setup once it's finished.

Edit: The scripts are now as an attachment