Advanced Search

That’s a solid approach, especially starting from Clough42’s ELS project. Mounting an encoder and motor on the dividing head is a clean way to get a lot of capability without going full CNC. The DR-DRO setup makes sense too, it’s way more flexible for oddball indexing and manual machine tricks. Doing something similar on an MH800E with LinuxCNC sounds like a natural next step, especially since you can blend manual control with some pretty advanced automation when you need it.

First thing is that my data path isnt /usr/share anymore, instead some documentations told me to get the file seperated copyed into the machine folder, for me i got the base ui and handler there : /home/.../linuxcnc/configs/.../qtvcp/screens/qtdragon_hd/. The second thing is your talking about taking your file name it "qtdragon_hd_handler.py", if i go like that i would override the current base handler, that doesnt make sence to me. What i exactly did know is

-Rename ur .py data to vacuum_popup.py (using it as a second handler)
-Implement your 3 lines to .ini like that:
EMBED_TAB_NAME=vacuum_popup
EMBED_TAB_COMMAND=qtvcp vacuum_popup
EMBED_TAB_LOCATION=DIALOG
-/home/.../linuxcnc/configs/.../qtvcp/panels/vacuum_popup/ is the path where the ui would be

If i start then it seems like it load but it cant cant work with the class: Like you will see in the error code i attached as .txt

I feel kind of sorry but i really want to learn the things, even if the start is currently kind of hard. But your help is getting me confident i can make it.
Thanks for your help

 

I took some time to modify some BIOS settings, performed several restarts, and ran tests. This is the result I was able to achieve.

sorry for being late in my answer, but job and family demand their time too.....This is what i get : joccnc@JocCnc:~$ sudo chrt 99 ping -i.001 -c 60000 -q 10.10.10.10
[sudo] password for joccnc:
PING 10.10.10.10 (10.10.10.10): 56 data bytes
--- 10.10.10.10 ping statistics ---
60000 packets transmitted, 60000 packets received, 0% packet loss
round-trip min/avg/max/stddev = 0.246/0.395/0.733/0.032 ms

In the meantime, I was testing around with the HOME_ABSOLUTE_ENCODER = 2 setting in axis SIM. It turns out that there might be a positional accuracy issue when homing. Ive created a github Issue for it already. It seems like its reading the joint position and not the encoder position when setting the new home so on re-home, it shifts the axis location in software by your home-offset value while hardware remains the same.  

juliankoenig87 wrote:

Der Titel dieses Threads suggeriert aber, dass du eine 7i96 nutzt. Wenn dem gar nicht so ist, bin ich bei euch.

Was soll der titel sein?

will do

Selber Bauen macht meisten mehr Spaß

Der Titel dieses Threads suggeriert aber, dass du eine 7i96 nutzt. Wenn dem gar nicht so ist, bin ich bei euch.

timo wrote:

a) How do they probably do it on my "store bought" milling machine? No need to home it. The beast knows where it is after it is powered on.

Absolute encoders.

timo wrote:

b) Can the home position be defined in the center of the machine travel? That would make the request easy. The axis can be homed at the position that was reached and the G54 can be "touched off".

Yes, even with normal encoder or without them, must have home switches.
It will home normally at axis ends, and move to wherever you would like it to be home.

timo wrote:

c) Can the machine be homed normally, then after disabling the motors moving it around and keep track of the position? Provided some feedback is available e.g. rotary encoder on the hand wheel, or linear encoders?

(that would allow to use the Linux similar to a normal DRO, when the motors are disabled.)

This kind of tracking without motors might be also interesting to teach a machine.

Yes, all that can be done.
Teaching should also be possible without encoders, jogging the machine to required positions and saving coordinates.

Just questions that come to mind (i did not do research on it)

a) How do they probably do it on my "store bought" milling machine? No need to home it. The beast knows where it is after it is powered on.

b) Can the home position be defined in the center of the machine travel? That would make the request easy. The axis can be homed at the position that was reached and the G54 can be "touched off".
c) Can the machine be homed normally, then after disabling the motors moving it around and keep track of the position? Provided some feedback is available e.g. rotary encoder on the hand wheel, or linear encoders?

(that would allow to use the Linux similar to a normal DRO, when the motors are disabled.)

This kind of tracking without motors might be also interesting to teach a machine.

  The really sad part is that other countries will look at it as a template for how to do things. I have noticed
how little value most governments seem to place on their citizens. Oh they will say one thing, but what
they do is what truly matters. Just another example of George Orwell's AnimalFarm 
"All animals are equal, but some animals are more equal than others"

But what do I know.  

Danke für eure Nachrichten und Vorschläge,
hab soeben Nachricht von meinem Kumpel bekommen:
„selber bauen mach mehr Freude “
Ok ich wollte nur mit meinen begrenzen Kenntnissen helfen,
bin da wo etwas zu schnell und zu weit vorgeprescht!

Bis hin wurde Schritt- und Richtungs-Signal vom Parallel Port, via selbstgebauter Karte, an die Motortreiberkarten geleitet.
Wir werden diese Schaltung übernehmen, da der Parallelport mindestens 5V ausgegeben hat.
BILD

Unschwer zu erkennen unten der Parallel Port, Rechts Analog 0-10V, per Relais Spindel ein und darüber die Referenzschalter.
Links die Anschlüsse für Takt und Richtung.

Ich werde unsere Lösung sobald sie verbaut und getestet ist hochladen. (Wenn ich darf!?)

Danke nochmals an alle Meldungen zu diesem Thema!

So it has been slightly more than 2 weeks, more like five or six because of reasons. But I have managed to make some progress over the holidays. The first step was obviously to clean up the entire machine, it might not have seen any serious use in a long time, if ever, but it was still in need of a serious cleaning.I started by removing all the cables and electrical parts from the machine since basically all of those will be replaced anyway.It turned into a neat little pile



After that I started cleaning the machine and even though I hadn’t really planned it I also pulled apart like 90% of the mechanics and cleaned all the parts individually.This was really necessary as I hadn’t been able to get all the plastic chips and old grease of any other way.



This included pulling apart, cleaning and re-greasing both the needle bearings for the ballscrews and the linear carriages.  As far as I can tell all this hardware is still available, the bearings  and what not, so it shouldn’t be an issue to service it if any parts give out in the future. 





Just look at that before and after. Some of the balls didn’t circulate at all since the traces were so full of hardened grease, depending on the loading direction. I got all of them running again after first wiping them down with break-clean and then flushing them with penetrating oil to loosen and rinse out all the grease.I didn’t remove any of the supported rails so hopefully I haven’t caused any alignment issue. We’ll see when everything is back together, I will do a proper alignment check then. 



It cleaned up real nice so I take that as a confirmation that this is a fairly low-hours machineIt’s mostly back together again and I think the next step is to work on the motor mounts. To that end I took a bunch of measurements when I had it apart to be able to manufacture some adapter plates, I will start to just 3D print them and when I have a design I like i’ll cut them on my old CNC router. It hasn't been used in a while so that will be nice. 



I also have the main layout of the electrical system finished, I just have some details left as well as doing the schematics, I am planning on doing a full set of schematics for this machine, I have been wanting to try QElectro tech and I think this is a nice project for that.Cheers   
 

The arc interpolations need some work, G1 seems pretty good, will work on arcs next, maybe not ready for testing yet...

ps i am reading tommylight post for the conf in hal file, it look promising!