Author Topic: Homing switches  (Read 420 times)

ggallant571

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 2913
    • View Profile
Homing switches
« on: October 24, 2018, 08:15:29 AM »
I have been using homing switches for a couple of years now and have an idea to improve the precision.

  1. Wire a CPU control line to the micro stepping mode pins on the controller
  2. Jog the machine using mode 16 till the switch closes.
  3. Back off till switch opens.
  4. Switch to stepping mode 1 and back off 2 complete steps.
  5. Resume stepping mode 16.

My thought is let the magnets in the motor force the alignment to increments of 1.8 degrees.

Not being a stepper motor or controller guru this is just an idea. Perhaps resetting the controller would do the same. Comments welcome!!!
SARCASM - Just one more service we offer here.

Zax

  • T2Laser
  • Global Moderator
  • Hero Member
  • *****
  • Posts: 6805
    • View Profile
    • T2Laser
Re: Homing switches
« Reply #1 on: October 24, 2018, 09:06:19 AM »
It may not be true for the stepper driver you use but most lose precision when changing the micro-stepping selection.

In any case I don't see the need for improved precision on these low cost lasers, there's more error in the rest of the mechanism than the steppers.

ggallant571

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 2913
    • View Profile
Re: Homing switches
« Reply #2 on: October 24, 2018, 10:12:00 AM »
My hypothesis  is the motor will "snap" to the natural 1.8 angle by the second step command and the micro stepping will reset from there. No data to back up thta assumption. My controller has jumpers on the micro-step inputs so it will be easy to wire up a test platform. Proving that it is "better" will be quite different. I was thinking about:

  1. Generating a 40mm square/triangle/circle test pattern.
  2. Etch it into an anchored test board at medium power - Get crisp lines, no char.
  3. Move the laser mechanism to some random location and remove power.
  4. Repower and perform a homing cycle.
  5. Redo the 40mm square with low power. Visual examination.

I also have some machinist indicator dials. Might be more precise than etching but need a mechanism to attach to fixture.
SARCASM - Just one more service we offer here.

Lob0426

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 1627
  • Let the Magic Blue Smoke out!
    • View Profile
    • Richards RasPi Web Server
Re: Homing switches
« Reply #3 on: October 25, 2018, 01:45:18 AM »
I ran a series of repeat tests with Bencutlaser. I tested with two different A5 sized machines. I would cut a shape then move the axis random distances. Then home and burn again. These were done at engraving power levels and slow speeds I saw excellent repeatability.

I never powered off the steppers or controller and then ran the test again. A dial indicator would be better than just a visual compare. I think that as you switch modes you might introduce "creep". If it stops between poles it will probably stay. If is at a peak between poles from micro-stepping I think it will have to fall forward or back. If it doesn't fall consistently it will creep. With a dial Indicator you will be able to see if it is staying in place or moving after power drops.

--------------------------------------------------------------------------------------
When I installed homing switches I took some time picking the switches. I bought a bag of 20. I then tested each switch for its "feel", travel, overtravel and it's sound. Some had a real crisp click to them and the click sounded like a glass breaking. Some were just plain mushy and traveled a bit they broke over, their audible click was not crisp and clear, closer to a crunch.

All of the switches that did not feel clean had a lot of travel, traveled more than half way before they broke over. They also had a lot of over travel. The crisp switches all broke earlier and had little to no overtravel at all.

A couple times I continued these burns until the shape fell free from the board. I noticed no stepping in the cut edges. Of course that could have been the edges were burned away. But I should have seen a coning effect or something other than the straight up and down sides I saw.

So the quality of the switches has a lot to do with it and then you pick the best. Another help is slowing the "homing feed" to about 350. Homing seek was about 5000.
Richard
A5 2.5W w/homing switches on L7
A5 2W diode self upgraded from stock 500mw
Printrbot Play. Ext bed, integrated tablet control.
Printrbot Simple Metal
USA