Author Topic: Limit Switch Ideas Requested for Quill/Spindle  (Read 5835 times)

Offline sparky961

  • Hero Member
  • *****
  • Posts: 844
  • Country: ca
Limit Switch Ideas Requested for Quill/Spindle
« on: July 17, 2016, 04:25:38 PM »
I thought it would be appropriate, since I was one of the first ones to agree with the idea of this new section (but not the idea's originator), that I be the first one to post here...

I had a thread going a few months back regarding the conversion of my mill/drill/lathe to CNC with LinuxCNC, MESA electronics, and servos.  I've come quite far but the project is temporarily on hold through the summer, as I do a lot of kayaking and backcountry camping during the good weather.  I get out in bad weather off season too, but not nearly as often.  Back to the point though, I have the X and Y axes connected, configured, and controlled with limit switches and servos.  The Z axis, however, is causing my brain to hurt a bit.  The issue at hand is a good way to mount the limit switches.  I have two ideas, both of which I can find fault with, and wondered if someone else could maybe contribute something better that I hadn't considered yet.

My criteria are as follows:
- Limit modifications to the underlying machine
- Reliable and repeatable
- Doesn't interfere with anything
- Somewhat out of the way to avoid damage to sensors, wires, my own hands or head.
- Just generally makes sense

Idea 1:
Similar to how you'd attach a cheap DRO to the quill of a machine.  There would be a vertical rod connected to the end of the spindle, captured by some sort of linear guides affixed to the machine body.  Attached to the rod would be one or two flags that would activate sensors mounted nearby.

Pros: Easy to execute, likely no machining of existing castings and such, some parts already made
Cons: More parts to fabricate, spindle has freedom to rotate a few degrees back and forth so would need to be rigid enough to prevent this

Idea 2:
Mill slots or drill holes into the spindle body (quill), offset radially and axially such that sensors could be mounted to a fixed area of the machine to sense the slot/hole and indicate limit of travel.

Pros: Could be simpler, slots would not require limiting the inherent rotation freedom of the spindle body
Cons: Requires removing spindle and machining holes/slots in fairly accurate locations, limited chances to change design along the way, debris may get in hole/slot, sensors may stick from the machine more and may interfere or be damaged


I can mock this up in 3D if my descriptions are unclear, but was hoping to avoid it for now.  Just think of your typical Bridgeport design and a way to know when the quill is full up or full down without adding a bunch of extra stuff.  The difference with this machine is that I don't have the typical depth stop mechanism you see on knee mills, just the non-rotating part of the quill that moves up and down.  I'll see about adding a picture with my phone after posting this (from my computer).

Thanks!

Offline chipenter

  • Hero Member
  • *****
  • Posts: 909
  • Country: gb
Re: Limit Switch Ideas Requested for Quill/Spindle
« Reply #1 on: July 17, 2016, 05:09:28 PM »
Are you planning to fit coolant pipe work iff so maybe combine the two ? I have been fitting standard micro switches on the X and Y not got to the Z axis yet .
Jeff

Offline sparky961

  • Hero Member
  • *****
  • Posts: 844
  • Country: ca
Re: Limit Switch Ideas Requested for Quill/Spindle
« Reply #2 on: July 17, 2016, 06:40:13 PM »
Are you planning to fit coolant pipe work iff so maybe combine the two?

No plans for coolant on this machine for the foreseeable future.  I have nothing to catch the mess and no need to run hot and fast.  A few squirts from a bottle would do when I really need it.

Offline RodW

  • Full Member
  • ***
  • Posts: 120
  • Country: au
    • Vehicle Modifications Network
Re: Limit Switch Ideas Requested for Quill/Spindle
« Reply #3 on: July 18, 2016, 05:51:58 AM »
Mount an upright to the spindle standoff and a couple of proximity sensors to the stepper mounting plate in such a way the limits are triggered when up and down as the standoff passes by. You might be able to get away with a single sensor that sees  metal for the travel and does not see it at each end.
RodW
Brisbane, Australia

Offline PK

  • Sr. Member
  • ****
  • Posts: 369
  • Country: au
Re: Limit Switch Ideas Requested for Quill/Spindle
« Reply #4 on: July 18, 2016, 06:00:09 PM »
Hey I know you mentioned not modifying the machine too much. But one, carefully drilled and tapped, hole in the side of the head would let you pick up the top of the quill, or the end of one of those slots with a simple proximity probe. It would be by far the neatest option.

Offline sparky961

  • Hero Member
  • *****
  • Posts: 844
  • Country: ca
Re: Limit Switch Ideas Requested for Quill/Spindle
« Reply #5 on: July 19, 2016, 12:39:30 AM »
Hey I know you mentioned not modifying the machine too much. But one, carefully drilled and tapped, hole in the side of the head would let you pick up the top of the quill, or the end of one of those slots with a simple proximity probe. It would be by far the neatest option.

This definitely has potential. I'll check out the feasibility tomorrow evening. Thanks, its an angle from which I hadn't considered things yet.

Offline sparky961

  • Hero Member
  • *****
  • Posts: 844
  • Country: ca
Re: Limit Switch Ideas Requested for Quill/Spindle
« Reply #6 on: August 28, 2016, 06:28:40 PM »
So, after some delay due to much time spent in the wild Canadian outdoors with some pretty amazing (frighteningly warm) weather this summer, I've returned to this part of my seemingly endless CNC saga.  One of the big differences this time is that I've been able to use the machine along the way to make or modify other parts.  Having a handwheel (MPG) on the machine means that even if other things aren't quite set up yet you can still drive the axes around as though it were a manual machine.  I wouldn't undertake another conversion without one.

I decided on PK's idea of drilled and tapped holes into the side of the head, picking up on the existing slot.  I've had a few miss-steps along the way. 

1. I quickly learned that the proxy sensors I have are triggered when deep in a hole surrounded by metal, not just metal at the end.  I think there are versions that only sense at the face but those aren't the kind I have.  This was solved by drilling and tapping the hole oversized and adding an aluminum bushing tapped inside and out. 

2. Unfortunately while working away at it a bit too late into the evening and my judgement clouded by sleep, I was using one of the sensors to turn thread an early prototype into the hole.  The threads were too tight and I ended up snapping one of the sensors.  Foolish, yes.  But it happens.  I went to bed directly thereafter.  A new one (plus a spare) is coming from China - maybe.  Until then, I more carefully made two more bushings and they are installed and waiting.  I tested where the switch is tripped by installing the remaining "good" one in either hole and I'm happy with the results.

3. After getting all of this sorted out, I went to re-install the driven timing pulley.  Argh.... I knew it would be close but didn't think it would interfere with the sensor.  Yup, it does.  This is where the project remained while I went away for a week on vacation.  Probably a good thing lest I get frustrated and break something in my impatience to get it done.  The time away did some good because when I returned I decided it would be a fairly simple matter to turn down the timing pulley to the next suitable size.  Since I got the rotary axis working first and I still have the profile cutter I used to make it in the first place, it should go pretty smoothly.  And it did.  Instead of having an 80 tooth timing pulley, I now have a 64 tooth pulley.  The new diameter clears the sensors and the belt path will run between them.

Video of the timing pulley being cut below.  No audio, and no apologies for shaky video and poor lighting. :P

Oh, and if you watch through to about 2:15 all that clattering around is the broom I knocked over while bumping into the shelf behind me..... yes, small space indeed.



This was the simple hand-written program I used.  I ran a few iterations of the program taking about 0.010" cuts and continually checking and increasing the depth to a maximum of 0.055" deep.
Code: [Select]
G90 G00 Y2.0700 X0.0000

#<TEETH>=64.0
#<CURRENT_TOOTH>=0

O101 WHILE [#<CURRENT_TOOTH> LT #<TEETH>]
G90 G00 A[360 / #<TEETH> * #<CURRENT_TOOTH>]

G00 Y1.9550

G01 F10 X-1.2500
G00 Y2.0700
G00 X0.0000

#<CURRENT_TOOTH> = [#<CURRENT_TOOTH> + 1]
O101 ENDWHILE

G90 G00 A0

M30

4. These things are never ending when you're working by the seat of your pants instead of a completely thought-out design.  Because of the new belt path requirements and to use the existing belt, the motor mounting plate needs to be redesigned.  The motor will be raised and the center distance increased.  I plan to have a new plate laser cut but for now I just sawed the thing in half and drilled two new holes.  It works better than I'd have thought, but it's on the list of things to replace as things progress.
« Last Edit: August 28, 2016, 08:22:19 PM by sparky961 »