Advanced Search

Linuxcnc will let you do anything but there is an expected paradigm. If you go against the paradigm, life becomes painful.
Think of the home position as being 0,0 on those Cartesian graphs you drew at school and the table represents one of those quadrants.
So if the current position is home, X & Y need to be  negative values to move away from home.

Another way of looking at the expected paradigm is that the front of the machine is where you stand at the machine. Home should be on your left and closest to you with the gantry going across your chest. So to satisfy that rule, you need to stand at the back of the machine behind the gantry.

So you could
1. stand behind the machine
2, Turn the machine around so the back as shown is the front of the machine
3. Live with negative X & Y coordinates
4. Move your home switches to the front left of the machine
5. In your homing configuration, set the home position to max-X and max-Y coordinates so it knows the home switches are at the back of the machine.

I might add if your origin/home respects the cartesian origin and you accept the quadrant you are in, it all becomes moot when you place your workpiece on the table and set the part origin with G54 offsets. Once you touch off, everything is referenced to the part origin.
 

For basic information on your configuration you need to post your .hal and .ini files without that we are just poking around in the dark.

Es gibt zig Dinge die schief gehen können.
Besonders mit (billiger) Elektronik.
Das Durchmessen vom LPT Kabel ist da wohl eher der Anfang, als das Ende.

Wenn die Motoren funktionieren zeigt das ja erstmal nur, dass es grundsätzlich geht. Und Linux startet.
Was war denn jetzt mit der Möglichkeit, dass das Relais "klebt"?

Ich würde mich jetzt von einer Seite durchwühlen.
Entweder aus Richtung Relais, oder aus Richung Computer.

Was ist denn schon genau ausprobiert. Ich würde mir eine Liste machen, was ich alles schon probiert habe. Die kannst du dann auch hier veröffentlichen dann geht es evtl. weg vom bloßen Raten.

a) M3 mit Drehzahl (siehe Tommylight)
b) hal show, nachsehen, ob Linux CNC den Pin am Parallelport versucht zu schalten
c) Led mit Vorwiderstand oder mit Spannungsmesser direkt am LPT messen, ob das Signal hin- und herschaltet
d) LPT Kabel wieder anschlißen, selbe Messung wiederholen
e) Computer abstöpseln und alle Verbraucher von der Karte abstölpseln und probieren, ob das Relais mit Spannung am Steuerpin schaltet.

Danach müsste man doch wissen, ob es auf der PC Seite hakt oder mit der Steuerkarte was nicht stimmt.

Gruß Timo

after the great work of master Rod, now we have the clear and direct tool to create a tangential knife configuration like that one imp,emented in Mach3 , to detect the cretical angle and rise then align with the new angle then dowen to cut depth , continue the g-code 

Yes, sorry, , Thanks for your aanwer. I m still strugeling to make Vismach going. ....I
is there any description for Vismach 5 Axis for a Linux newbie? I think the AI made mistakes....

Right now I found out my toolpath from Powermill is alredy with RTCP
How can I turm off RTCP in Linixcnc?

I wonder if they ended up using the linuxcnc board or not. i was pushing a remora thing because they were using klipper, but i think they eventually just went linuxcnc and made a rpi breakout for it because they dont need as fast stepgen as a cutting cnc. I havent checked in for a while thought

I'm definitely wrong because I was a machinery repairman. I started in the '80s with Olivetti CNZ machines, all transistors with delay lines. The last 20 years of work have been spent welding and cutting. I built a small CNC and a plasma cutter for fun. In the photo, you can see the machine in the home position; the crosshead to the left is negative, and the carriage coming forward is positive. It doesn't look right because with the handwheel clockwise, one axis is positive and the other negative, it's easy to make a mistake. But if I invert the crosshead to make them go the same way, the plot is mirrored. But I don't even waste too much time on it because I have Sanmotion AC motors, so I'll do everything properly.

The "error finishing read" has nothing to do with jitter or ping time.
In short, this means network conntroller works well, but breaks connection after some time.
Cause may be the corrupt hardware, bad network driver (try DKMS), or bad kernel version (try other and beware upgrades).

Co mam podać ?

Hallo,

ich habe über die stepconfig eine neue Maschine angelegt,
die Endschalter entfernt. Alles gespeichert. Referenzpunkte manuell
angefahren. Dann in die Mitte der Aufspannplatte gefahren und jede
Achse genullt. Programm eingelesen und gestartet. Die Maschine und Steuerung haben alles prima ausgeführt
Meine Maschinenkonfiguration probetap kann ich in stepconfig nicht
einlesen, da alle Enstellungen dann gelöscht werden.

Viele Grüße karlhe

I'm sorry but you do not provide ANYWHERE near enough information to give further assistance.

More info?

Hello all

I recently briefly mentioned my EL5021 modifications in my Maho MH400E retrofit discussion. Due to three additional terminals requiring the same modification, I decided to write a guide on how to modify the Beckhoff EL5021 encoder input terminal for both 1 V_pp as well as 11 µA_pp signals (and in theory TTL signals too - untested).

For anybody who's interested, the detailed guide can be found over on GitHub:
github.com/PedPEx/EL5021_multi-signal-mod

A video of a modified EL5021 reading 11 µA_pp signals from a Heidenhain LS403 linear glass scale is also available on YouTube:


If you have any further questions or suggestions, feel free to open an issue, reply here, or write me an email.

Tak jak w temacie mam uciety widok AXIS , brak napisu narzędzia jak to naprawić w linuxcnc 2.9..8.  
 

Now it is running, i made ini and hal with Ki.
When running a toolpath in Vismach simulation it shows perfect but the machine makes strange things.
In fact I have not reaaly a plan what setting i need in Linux for 5 Axis...
Scale is correct