3D printer - Piezo Z probing

[tekkydave]

Yes, it connects to the z probe connector. Mine triggers all the time during xy movement but that is ignored. The 500mS pause before starting the probe 'dive' ensures all xy motion and vibration has stopped.

[tekkydave]

This my homez.g. If you are using an older version of Reprap Firmware change the H2 in the G1 Z5 command to S2.

; homez.g
; called to home the Z axis
;
; generated by RepRapFirmware Configuration Tool on Thu Dec 28 2017 14:34:20 GMT+0000 (GMT)
;

G91                 ; relative positioning
G1 Z5 F600 H2       ; lift Z relative to current position
G90                 ; absolute positioning
G1 X150 Y145 F15600 ; go to centre of bed
G30                 ; home Z by probing the bed

; Uncomment the following lines to lift Z after probing
G91                 ; relative positioning
G1 Z3 F600          ; lift Z relative to current position
G90                 ; absolute positioning

 

[Gina]

After a lot of testing and adjusting of the sensitivity, it's working! ?  Produced my first print since removing the inductive Z probe and relying on the Precision Piezo and the print is fine.  Now I shall be able to convert my other printers which will be especially helpful when changing hotends - yes, the bigger printers have interchangeable hotends.

[Gina]

Hmmm...  Thought I'd test repeatability.  FAIL!!

[Gina]

Well, I guess I can't expect anything to be simple!  Needs more poke from the nozzle and/or less from elsewhere.  I have the sensitivity pot turned up quite high so it seems the vibration from the X drive could be the problem.  That may improve when I rebuild in a nice rigid plywood box. 

Edited February 20, 2019 by Gina

[Gina]

Found the problem - wire off sensor - so have soldered it back on and the homing and mesh grid compensation worked fine manually.  Now trying a print again.

[tekkydave]

I replaced the tiny wires to the sensor with a pair of wires from a ribbon cable. Much more robust and easier to crimp. Be careful soldering to the piezo - too much heat and the piezo material just disintegrates. Also pays to have a few spares. You could use undrilled ones if you dont have the filament going through the centre.

[Gina]

Yes, I think I'll order some extra sensors when I order next.  I plan to upgrade both my Concorde and Giant printers with piezo sensors.  And yes, I was thinking I might as well get undrilled ones.  I managed with crimping the tiny wires alright as it happens.  I have a use for two of the now redundant inductive proximity sensors - endstops/limit sensors for automating my roll-off-roof.

[Gina]

They only had one amplifier PCB left ?  Oh well, the Giant printer can wait - lots of other things to do anyway.  Ordered the PCB and 3 undrilled sensors.

[Gina]

Been looking at the X carriage of my Concorde printer.  I seem to have made this so rigid that I can't find any flexibility in it.  When I push up hard on the hotend the whole X carriage seems to rotate very slightly and I 'm finding it very difficult to find where this minimal flexing is occurring.  I thought at first it was in the Y carriages but I couldn't detect any movement.  I think the flexing must be in the X rail extrusion.  This is pressing with a force of about a lb.  I'm not saying that using a piezo sensor won't work - they are so sensitive that an imperceptible movement will trigger them.  Once I get the second set of parts I can try some experiments.

[Gina]

I now have a second Precision Piezo sensor kit and will try it on my Concorde printer.  The sensor on the Mini is working fine.

[Dr_Ju_ju]

have you tried to do a bed height map yet ?

[Gina]

Do you mean on the Concorde or Mini?

Edited February 22, 2019 by Gina

[Gina]

I have the sensor set up on the Concorde between underside of extruder motor at the back and the top of the back X carriage plate.  This was triggering just the opposite of what's required - triggering as the print bed is raised towards the nozzle but not triggering on contact - so I reversed the sensor connections and it worked.  I guess the motor/extruder/hotend combo is being lifted when the nozzle contacts the bed. 

A slight adjustment is wanted as the X carriage is being lifted slightly though noticeably by the bed.

Edited February 22, 2019 by Gina

[Gina]

Here's a bed height map for the Mini using piezo probing and another taken some time later.

Edited February 22, 2019 by Gina

[Gina]

Here are photos of the Concorde X carriage.  The extruder motor is sitting on the rear plate.  This is where I put the piezo sensor when I was testing as seen in the 3rd photo.  It might work with an appropriate gap for it but currently it seems the pressure is too much - it doesn't work.  The rear fan is for part cooling.

Edited February 22, 2019 by Gina

[Gina]

I think if I were to make a small hollow if the edge of the rear plate to take the sensor and reduce the height so that it didn't touch the motor, it might work.  The principle would be similar to that in the Mini printer - the nozzle making contact with the bed would produce a slight bend in the front plate and the motor would then compress the sensor.

Edited February 22, 2019 by Gina

[Gina]

Yes, that works but I shall need to redesign the X carriage to make it easier to change hotends.  The main problem is the part cooling duct.

[Gina]

Well, it seemed to work on test with bench PSU and watching LEDs but not when connected to the Duet through the Z probe port.  The vibration from the Z motors always seems to be more than the tap from the nozzle touching the bed.

[Gina]

Would you believe it - the Mini printer piezo probing has stopped working too!!  I've printed several items on the Mini since changing to piezo probing so why should it go wrong now!!??  Must be "out" in sympathy with the Concorde printer!  Looks like I shall have to change the Concorde back to inductive probing as I need a 3D printer for my astro projects - at least the inductive probe and the copper foil sensing pads are still on it.

[tekkydave]

What is your M558 command in config.g look like?

[Gina]

M558 P8 I1 R0.5 H3 F420 T15600 A5 S0.03   ; P8 = Digital mode (unfiltered), I1 = Active Low, R0.5 = 500mS delay before probing	
                                          ; F420 = 420mm/min probe speed (7mm/sec)
                                          ; T15600 = Travel between probe points at 15600mm/min (260mm/sec)
                                          ; A5 = Max No of touches, S0.03 = Tolerance for multi-touch (mm)
                                          ; H3 = Dive Height 3mm

 

[tekkydave]

Ok, I see you used the one from my config.g

Does the piezo trigger when you tap the nozzle?

Have you also copied my G31 command?

[Gina]

Yes, yes and yes.

G31 X0 Y0 Z-0.2 P100                      ; X, Y, Z offsets, debounce interval

 

[Gina]

Thought I'd check something...  My maximum Z speed is 300mm/m = 5mm/s so not as fast as 7mm/s.  Guess that's the problem.

Edited February 27, 2019 by Gina