Need to find X and Y positions of two nozzles

I am looking for a way to find the X and Y positions of two nozzles on a printer. The two nozzles are switched rapidly to approximately the same position. A video can be seen at [www.youtube.com] I need to determine the relative position to a reasonably good accuracy - the target is within 25 microns.

Relative Z positions of the nozzles has been solved and is within 5 microns.

Any thoughts on what methods I can explore to solve the XY problem?

Mike

How about printing a cube, starting with 100% infill. Do a few layers with one nozzle, then switch to the other nozzle and do another layer. Measure any offset, adjust and repeat. Then when it's as close as you can get, change to vase mode, (single wall), and repeat. Do quite a few layers like that which should be perfectly aligned. It may be that a finger tip is more sensitive to any differences than an eyeball. (Neat mechanism BTW).

---

[somei3deas.wordpress.com]
[www.youtube.com]

Thanks deckingman but I am trying to find an automatic way of lining things up. So far the only thing I have come up with is a capacitative method with targets plated onto the glass of the print bed. While this looks promising it does mean that it will not have any independence of bed surface. I have seen a method which prints reference lines and uses a camera and some built-in smarts to calculate the error. I think that kind of programming is waaaaay above my pay grade.

At the moment I am using a jig with a couple of dental mirrors which is able to look at the nozzle from two sides 90° apart at the same time.



Jig and view of side and edge of a bamboo stick ca 4.5mm x 1.5mm (laughingly called a teaspoon in a local cafe) Left-hand mirror shows the thin edge and right-hand mirror shows the broad surface 90° around.

A few years ago we experimented with calibrating the nozzle offsets of an IDEX printer by printing sets of lines with each nozzle and then using an IR sensor to locate the offsets of each set of lines. It worked, but we didn't take it any further. You could try something similar. The code to store the positions of the transitions from the IR sensor is still in the RRF source code I think, but commented out.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Thanks dc42, that sounds like the most exciting project that I have come across in a long time. It is also appropriate as my new printer now sports a single underbed piezo to get the reference Z data for the two nozzles while the remaining bed level and flatness data is obtained with one of your IR sensors.

There should be enough work here to keep me out of mischief for a good while. I will need to set up experiments to answer questions like:-

• Is the sensitivity cutting across a printed line in the X and Y direction as good as the sensitivity in the Z direction? If not, how can it be improved?
• In view of the geometrical position of the IR emitters and receiver, does the sensitivity vary with the direction of travel?
• What is the best way to extract the line centre from the IR response? peak pick or average distance from rising edge to falling edge?
• Will the sensitivity vary with line width, thickness and material?
• .... and many others

I will report back with the results of my experiments as they go.

Mike