[fraser80]

Hey,

I recently finished building a CNC router I am spending too much time aligning and touching off my workpiece. I would like a 3D touch probe but I haven't found what I am looking for.

From my research, the really good probes are expensive and have accuracy which is much higher than needed for CNC routers (0.0001" range). There are also a lot of cheap Chinese or DIY probes but they suffer from dirty contacts and inconsistent readings.

I have decided to design and build my own probe purpose built for use on CNC routers but I would like some feedback before I get too far into this.

Here are the specs I have come up with:

• Accuracy of 0.001"
  
  • Most CNC routers are not that accurate, so better sensing accuracy isn't necessary
  
  
• Robust, Low Maintenance
  
  • Should not need to take apart to clean contacts
  • Only maintenance should be adjusting runout
  
  
• Simple runout adjustment
  
  • I'm thinking about eccentric ring. Rotate rings relative to each other to adjust amount of offset. Rotate together to select direction of offset.
  
  
• Operate on 5V to 12V input voltage
  
  • Used to power internal circuit, LEDs, switches
  • Isolated from touch input/output wires
  
  
• Touch triggers a switch
  
  • Probe signals a touch by completing a circuit between probe input and probe output signals.
  • Normally open circuit closes when probe signals touch.
  • Works the same way as a conductivity touch probe.

Undecided Specs:

• Ruby Probe or Stainless Steel Probe?
  
  • I don't think a ruby probe tip is necessary since the desired accuracy is only 0.001"
  • CNC routers cut softer materials like wood, plastic, aluminum so wear shouldn't be a concern
  • Stainless steel would be cheaper, easily customized.
  
  
• Cost
  
  • How much should a good CNC router 3D touch probe cost?

I am a mechanical engineer and I have been discussing this with an electrical engineer. We think we have a solution for a simple touch probe. We have identified the issues with popular switch based probes and believe we have a solution.
It will require a custom electronics board to make it work so I would like to know if you guys think we are heading down the right path.

What do you guys think?

[elfrench]

You'll need resolution which is ten times the desired accuracy in order to actually attain that accuracy.

[A_Camera]

Creating specs is the easy part... making a prototype is the next step. Make one and see what is good and not good in the idea.

Anyway, what is the aim? To sell it or just a DIY project for yourself?

I agree that ruby tips should not be necessary for probing on soft material, but it is a once per life time cost, so it doesn't really matter, unless you can make high quality stainless tips but not ruby tips. Also, your tuning mechanism seems to be more complicated than the traditional three screws principle, or am I missing something? I mean, with the traditional three screws you can use standard screw, but with your design you need to make custom rings.

[fraser80]

Yes, everything looks great on paper!

My goal is to make a basic prototype to see if my concept will work the way I plan. If the prototype works out I was thinking about doing a kickstarter or something so I can make a batch order. Not really to make money but a batch would allow my to source better components to improve the quality.

The idea with the eccentric rings is to make the adjustment somewhat toolless. Adjusting would involve rotating the rings which make up the housing by hand to dial in the adjustment. It would have very fine adjustment and I think this is more intuitive vs loosening and tightening three screws (but maybe that's just me, haha). Please correct me if I am misunderstanding, but it looks like the three screw method involves lifting or lowering the PCB on the bottom of the probe. I would prefer to have the PCB firmly mounted.

I am open to suggestions or advice because I would rather make a change on paper before I invest a lot of time building and sourcing components.

Thanks,
Fraser