I don’t home that often but I do it regularly enough that I would have noticed any issues. Never had any until today when I got an Alarm 8 for Z. After some troubleshooting, I noticed that sometimes it was successful, other times not. But what I noticed is that for the successful homing, the seek rate was slow. It failed when it was high. So I reduced it from the 4300mm/min to 1000mm/min and that seems to work fine. I wonder why this started happening suddenly. All the connection look fine.
All I can think of is to make sure the sensor and the part of the Z that it senses are clean. Maybe test the sensor with a wrench to see if it acts differently than the X or Y.
The AltMill is crazy by the way. 4000mm/min is the default maximum speed of my LongMill for X and Y! I can go a little faster with the SLB but homing is still at 1500mm/min.
@_Michael Good idea to verify the cleanliness of the two parts. That’s the only thing I didn’t check. Other than the two Y sensors that were loose, the X and Z were still tight. And other than my upgrade to 1.5.7, nothing else changed. I downgraded to 1.5.6 but still had the same issue. It’s only when I reduced the seek speed that it worked consistently.
Now, it’s possible I had the problem before and never noticed. There are two seek speeds used for homing. One set to 4300mm/min when far from home. One at 150 mm/min when it gets closer to home. When homing failed, it was seeking at the high speed. When it worked, it was seeking at a low speed. What I’m unsure about is if that low speed was 150mm/min. It definitely seemed faster than that…probably more like the 1000 mm/min that I ultimately settled on. And the spindle wasn’t even close to z=0, so it probably should have used a higher speed. So not sure what’s going on there.
But I’ll check cleanliness just the same. I’m due for my regular maintenance anyways.
If the AltMill works like my LongMill the fast speed is used until the sensor is tripped, then it pulls off and finally uses the slow speed to get a more accurate result.
Maybe you knew that but it wasn’t clear to me in your second paragraph.
EDIT: If it’s failing when it gets to the sensor increasing $27, the pull off distance, might help. My thinking is the faster the initial part is the longer it takes the mill to stop. If it stops close enough to the sensor that the pull off isn’t far enough for the switch to return to ‘off’ then grbl can’t do the slow accurate one.
@_Michael No, didn’t know that. I figured the sensor detected the approaching metal and slowed down at some point. Makes senses that it would trip (without generating an error) and then fine tune the homing.
Check my edit above if you didn’t notice it.
1 Like
@_Michael I’ll give it a shot
1 Like
@_Michael Bingo. Changing it from 1.5 to 2 seems to have fixed it. Back to 4300mm/min seek rate. Thanks muchly.
1 Like