Advanced Search

Cakeslob is spot on.

The latest firmware that supports the NVEM / EC300 and EC500 is this repo.

github.com/scottalford75/Remora-RT1052-cpp

The version you have used is very old original firmware and it didn't have the configuration upload capability.

Latest firmware is github.com/scottalford75/Remora-RT1052-c...ora-rt1052-3.1.3.bin

I've took it from here :
github.com/scottalford75/Remora-RT1052/tree/main/firmware

There is a firmware mentioned by me and it looks like it's been made for nvem board. Scott could you please explain which one is correct for nvem boards?

Lpkkk wrote:

I've flashed remora-rt1052-nvem-0.2.1.bin, is that the right bin? 

probably not, but I dont think you are working out of the correct repo

github.com/scottalford75/Remora-RT1052-cpp/tree/main/Firmware

Aah of course, H is the once-per-rev signal. Somehow didn't occur to me, but now it makes sense, thank you!

They are connected directly to the ballscrews, the spindle encoder is connected via belt. They use "Heldt & Rossi SM 807 DC" servo drives for X and Z and a "Baumüller BKF 12/70/400" for the spindle. Controller is a "Sinumerik 3T".

 
 
 

For now, m5 solves the problem in my tool measurement procedure. I do not use the built-in qtvcp procedures. Let someone smarter figure out how to solve the problem, I probably can't help.

Returning to the topic of the axis buttons function. Could you add the "go to zero" function?.
I only added the titles to the axis_tool_button.py file to the screenshot. I do not have the functions working yet.

They look like 'normal' differential encoder outputs.  A/A-, B/B-, and H/H-... with H being the once-per-rev index signal.

Should be a standard Mesa connection.

Are those encoders connected directly to the ballscrews, or to the motors?  What's driving the motors?

 

After years of using a very old (1930) and worn out lathe, I decided to get myself something new.
Found a Weiler 160 CNC in very good condition (400 spindle hours) that has been used lightly in a school. The buffer battery is dead and the screen does not turn on anymore, so a perfect project for a retrofit.
The documentation is one of the best I've ever seen, everything from bearing preloading up to motor change of the toolchanger is described. Tons of drawings from every part. The whole machine just screams german quality.
 
 

Only thing I'm not really sure about is the measuring system. It uses incremental encoders with 6 signal pins instead of glass scales. Does anyone know how I connect them to the mesa cards? Can I use the normal encoder inputs for that or do I need something else?
 

Yes, basically LinuxCNCs motion component already does this with the MPG count.

sudo apt update
sudo apt upgrade
sudo apt install linuxcnc-uspace

Hmmmm why?
LinuxCNC can use any encoder as MPG, so no need to convert anything, even on parallel port it works.

Please see attached 

I am building an MPG control for a Chinese 920 lathe I am converting to CNC. I want the controls to have the same intuitive feel as the original manual lathe, with the option of overriding to use full CNC, so will be fitting an MPG where the standard controls would be. I have the prototype setup working, using _encoder_ to decode the quadrature encoding (it is strange the the decoder component is called "encoder"). This produces a count, but I'd also like quadrature decoded to pulse + direction for further processing.

Is there any way of doing that with standard HAL components?

 

When I loaded latest debian, I thought linuxcnc will be included.

Can any one point me to the procedure where I can install linuxcnc to my existing debian 12.6

I already constructed a DIY machine running on linuxcnc using another pc, but I like the full version of debian 12.6.

Many thanks

Philip

yeah I figured I would have some fun with it. LOL

Already dealt with, thank you for "never gona give you up" and "spam with spam on top of spam" !