-
Configuring LinuxCNC
-
Basic Configuration
-
Problem with Gantry Homing
Problem with Gantry Homing
Less
More
-
Posts: 24
-
Thank you received: 5
-
-
23 Oct 2018 19:33 #119276
by sevaz
Hello everyone! I have a problem with gantry machine homing procedure. Two motors on one axis doesn't do the last synchronized final move in the same time. That means, when the first one touches the limit switch, it starts backing off, without waiting for the second one.
If i'm not mistaken, it should work like in this video:
This is the tutorial i used to set up my gantry machine:
forum.linuxcnc.org/49-basic-configuratio...-linuxcnc-2-8-master
Here are my configuration files:
Where did i go wrong? Maybe i misunderstood something?
Please Log in or Create an account to join the conversation.
Less
More
-
Posts: 24224
-
Thank you received: 4798
-
-
23 Oct 2018 23:47 #119301
by andypugh
The video looks right, so I assume that isn't your video?
You are apparently using the latest master / 2.8 so that's not the problem, and joints 1 and 2 both have home-sequence -2 so that seems OK.
I think that a HOME = 0 and a HOME_OFFSET = 0 is a possible problem, I think that the machine will park on the limit switches after homing.
Can you describe exactly the behaviour you see? Is the machine assembled? Are the limit switches definitely not crossed-over?
The following user(s) said Thank You:
sevaz Please Log in or Create an account to join the conversation.
Less
More
-
Posts: 10204
-
Thank you received: 3279
-
-
24 Oct 2018 09:11 #119321
by rodw
linuxcnc.org/docs/devel/html/config/ini-homing.html
The initial HOME_SEQUENCE number may be 0, 1 (or -1). The absolute value of sequence numbers must increment by one — skipping sequence numbers is not supported. If a sequence number is omitted, "Home All" homing will stop upon conmpletion of the last valid sequence number.
I think -2 is not valid.
If that does not fix it, I would also check your limit and home switch wiring is correct. When I've had issues like that it is because I've had something incorrect.
Also using very slow homing velocities helps isolate problems as you have time to see things happen.
The following user(s) said Thank You:
sevaz Please Log in or Create an account to join the conversation.
Less
More
-
Posts: 24
-
Thank you received: 5
-
-
24 Oct 2018 09:17 - 24 Oct 2018 10:02 #119322
by sevaz
Thank you for your replies, but it seems the problem is gone by itself. I just misunderstood the way homing process functions.A joint goes to the limit switch, then back until it opens again, waits for the other motor does the same and only then they both make the final move. I only realized it when i made a proper test installation with actually working two stepmotors (before i used hal scope and dro). So yes, now it's working as it meant to be, nevertheless the documentation for homing sequences isn't too friendly for beginners
Last edit: 24 Oct 2018 10:02 by
sevaz.
Please Log in or Create an account to join the conversation.
Less
More
-
Posts: 24224
-
Thank you received: 4798
-
-
24 Oct 2018 09:26 #119323
by andypugh
I think -2 is not valid.
-2 is fine as long as there is also a -1 or a 1.
I think it used to be the case that the homing sequence had to start at zero, but when negative numbers were introduced to hint that motors were in multiples it became difficult to configure the first in the sequence to home together. ( -0)
-0 does actually have a unique floating-point representation, but doesn't work as an integer.
The following user(s) said Thank You:
sevaz Please Log in or Create an account to join the conversation.
Less
More
-
Posts: 567
-
Thank you received: 323
-
-
-
Configuring LinuxCNC
-
Basic Configuration
-
Problem with Gantry Homing
Time to create page: 0.239 seconds
