Advanced Search

Even then unknown I don't expect things to get better with the next administrations.

You can thank the UK gumbernment and my own pseudo socialist gumbernment. It's embarrassing to be an Aussie at the moment and we have 3 more years of the fools.

Unserstood, thank you

Hi Everyone,

I am in the process of wiring up my first CNC machine and just spent the day finishing up the wiring from my closed-loop stepper motors to the drive, and was wondering if I had made a big blunder. Since the encoder needs 6 wires for EB-, EB+, EA-, EA+, VCC and GND and the stepper power needs 4 more for A+, A-, B+ and B-, I used a single 10-core shielded cable from the drive to the stepper. Is this ok, or will it cause problems, and should I have used a 4-core and a 6-core cables instead? I don't have much of an electrical background, so I am really unsure about this. Thanks for your help

Gautham

Hi Everyone

I am in the process of wiring up my first CNC 5-axis machine and I wanted to know if I can use this Rotary Encoder  to home the A-axis and if so how would I wire it up. From the research I have done, my understanding is that it can be done but it involves some editing of VHD files that I didn't really understand since I am new to LinuxCNC. Thanks for the help

Gautham

Thinking about this a little more, where is the documentation link to writing custom conversation macros? Basic ones,I can learn on or test.

I have a few conversation mill events in python, but just can't figure out how to try and load them. Looks like I need a gfx screen and then links to the macros. Which then produces some gcode?

Any help or videos would be awesome. I want to add a nice pocket and frame event.

AG

richcolvin wrote:

What did you do to fix it?
 

I updated to GTK+ v3.x, and asked ChatGPT to help me update my lzpanel.ui file to the latest version. It was amazingly helpful in this regard, although I think there were a couple hiccups along the way.

It's almost time to make V2....!!! (like for real)
Big enhancements in the work holding area. quick change 5 side vise indexing, and some other tricks. 430mm x 270mm/310mm travel And a 10 tool robodrill style tool changer.


 

 

 

For axis or joint jogging, there are HAL IN pins for jog-counts. For example, for axis jogging there is axis.L.jog-counts IN S32.

When movement is permitted (enable pin high, in manual mode, not at a soft limit etc) a change to the jog-counts pin will cause movement.
Since the jog is incremental, normally the value of the pin is increased incrementally by the control method (such as a jog wheel), not set to a specified value.

This means that there is potential for a backlog of incremental jog movements to be accumulated, if the rate at which the incremental logs are executed in one direction exceeds the rate at which the jog counts pin has changed.
This manifests as a continuous jog which ends once the backlog has been cleared.

My question is, is there currently any way of accessing the number of counts which make up this 'backlog'?
Or accessing an analogous value such as the DTG for the joint/axis if the backlog of jog counts is summed?
This data should exist in some form in the motion planner, so it should be possible to get it into a HAL OUT pin.

GStat does seem to have a get_jog_increments function, but since the other arguments reference GStat's internal jog parameters, so I think this just returns what was set by the set_jog_increments function.

In summary (as far as I know), there is an IN pin for jog counts, but no OUT pin which actually indicates how many counts are 'queued' in the motion planner.
Is it possible to make this into a HAL pin in a current installation?
Or what might be changed in the current motion planner to make this value a Stat/GStat attribute or HAL pin?

Thanks!

again, cheers once again for pointing me in the right direction.

unfortunately i'm still far to green to get a working hal snippet for pwm control, but chatGPT isn't... after a bit of back and forward this was the result:

setp lcec.0.0.PWM-OUT-0 98304000 #sets frequency to 1.5khz

loadrt scale count=1
loadrt conv_float_u32 count=1

addf scale.0 servo-thread
addf conv-float-u32.0 servo-thread

setp scale.0.gain 3.5

net gui-rpm spindle.0.speed-out => scale.0.in
net scaled-rpm scale.0.out => conv-float-u32.0.in
net rpm-u32 conv-float-u32.0.out => lcec.0.0.DUTY-OUT-0


i now have a working PWM out!

i did try outputting to analog out but there seems to be some problem with it, for example 20 000 rpm would output 3v, 10 000rpm would give 5v, 7500 rpm would output 2v. It's not the end of the world, but hopefully i can get it sussed at the very least for others that try this board.

once again,
cheers for the help to get a working PWM out!

Yes that may prove very valuable. The worst part I see is that this isn't a single country problem it's spreading
everywhere. It seem the powers that should never be don't like it when people don't follow the pushed narrative
or the easy in which things can be fact checked independently.  
I guess it's just another day in mouse utopia. 

It all started with websites like Quora and User Benchmarks back in the days and the subsequent invasion of blogs.
Now this is just a fallout form that as when so called AI hit and started multiplying like viruses, there is no going back to Wikipedia actually being very useful ever again.
-
While thinking about this, does anyone else think it is a good idea to make a topic about useful websites and video channels here on the forum? It might help a bit to slow down the extreme misinformation going on.

So I just came across this . All I can say is The internet is loosing it's usefulness.

It's in the image you downloaded. Look for the components folder I think it was called. Open that, you should see the Eth files. Open the terminal(command line) from here and run the halcomplie command that's in step 8 of the instructions. 

What did you do to fix it?