Black printbite and IR Sensor - very inconsistent heights?



  • I have added black Printbite to my cartesian printer. It is now producing very variable heights when doing a bed mesh check - with points 10mm apart showing as over 0.6mm different in height with big peaks and valleys.

    The bed is 4mm aluminium, mounted on the two front corners and mid back.

    A sheet of scrap 3mm aluminium from the garage simply laid over the bed is apparently smoother than the PrintBite so something doesn't seem right.

    Has anyone else had similar issues with black PrintBite and the IR Sensor?

    Richard


  • administrators

    Are the optical components on your IR sensor through-hole or SMD parts?



  • It has SMD components, David. If it helps identify it, I bought it only 6 - 8 weeks ago.

    Richard



  • Did you get the gloss or matte?



  • Matte, I thought it would work better. I've even tried roughing up the surface with a scouring pad. No change.

    Richard



  • Do you use black Printbite, cbusillo, and if so does it work OK with the IR probe?

    Richard


  • administrators

    Please confirm that you have the P parameter in the G31 command set to 500.

    What trigger height are you getting, measured from the bottom edge of the sensor board to the bed?



  • The applicable line in config.g is:
    [c]G31 P500 X9 Y21 Z2.26 ; Set Z probe trigger value, offset and trigger height[/c]

    The 2.26mm was set using the method on this page https://miscsolutions.wordpress.com/mini-height-sensor-board/

    Richard


  • administrators

    How much higher is the edge of the senzor board than the nozzle? What I am after is the height of the sensor board above the bed when it triggers, which will be somewhat greater than 2.26mm.



  • Oh, I see, sorry. The bottom edge of the board is about 3.25 - 3.3mm above the PrintBite (it is difficult to measure so used a taper and measured that afterwards).

    What should it be?

    Richard



  • The height measured with G30 S-1 is repeatable when measured in the same spot. These are readings from one location:
    2.382
    2.380
    2.382
    2.385
    2.385
    2.385
    2.380
    2.382
    2.382
    2.382
    2.382
    2.382
    2.380
    2.378
    2.380
    2.380

    Moving the nozzle by X=10mm and I get the same level of repeatability but around 2.200mm. Another 10mm across and it is back to 2.380

    I turned the bed compensation off to test this.

    Richard



  • Thanks, swoozle, but that seems to be related to the clear / white PrintBite as it talks about grid lines. The black that I have does not have those lines. I'm thinking of changing to an inductive sensor if I can't get this right.

    I'd use the piezo but I don't know how I'd connect it to my x-carriage or bed - neither was designed for that.

    Richard



  • can you lower your IR Sensor by about 1 mm in relation to the nozzle?

    I think the recommended distance is between 1 and 2 mm but it is a while since I used one



  • Yes, I could try that but I don't have much height to play with. The dimension I quoted to David is the height of the bottom of the sensor board above the PrintBite when the sensor triggers - 3.3mm approximately. At that point the nozzle is 2.26mm above the PrintBite so the height difference is close to the recommended 1.5mm - actually a bit lower than recommended but it never fouled a print on the previous print surface (PEI on a black PCB heatbed).

    Now that I've made the change to aluminium plate and added the PrintBite I'm beginning to wonder if that inductive proximity sensor isn't my best bet. Either that or a BL Touch, perhaps.

    Richard



  • I've just used Blu-Tack to mount an inductive sensor under my carriage and tested that manually by monitoring the LED on it as I lowered the nozzle. Even like that it is repeatable to less than 0.01mm and the peaks and valleys seem to have gone.

    I just need to figure out how to wire it up (I'm already using an NPN inductive sensor on Z-max as my Z-home and only want this one for Z-probe, not homing.

    Richard


  • administrators

    I'll ask Jason again for a sample of matt black PrintBite to test the IR sensor with.



  • Thanks, David.

    It does seem to be something to do with the interaction between the PrintBite and the IR sensor. I'm busy printing a mount for the inductive sensor and will try that later.

    Is it possible to connect the NPN inductive sensor yo the Z-probe input of the Duet WiFi through a Schottky diode? Is it the Z-Probe_In pin that I connect to the diode and the sensor and then just set it as a switch input?

    Thanks.

    Richard


  • administrators

    @RGN01:

    Is it possible to connect the NPN inductive sensor yo the Z-probe input of the Duet WiFi through a Schottky diode? Is it the Z-Probe_In pin that I connect to the diode and the sensor and then just set it as a switch input?

    Yes, see the wiki. If you use the Z probe input then an ordinary small signal silicon diode will probably work too if you don't have a Schottky one.



  • Hi Richard, Generally it has been positive feedback with the IR sensor on the black (matte) PB. Dave will have some to test early in the week. Possibly a silly question but any surface imperfections or scratches on the PB that might be distorting the readings?
    J



  • Thanks, I found the new one (the format is growing on me) and got it sorted. The applicable lines in config.g are as follows:

    [c]; IR Z-probe
    M558 P1 H4 F250 T2000 ; Set Z probe type to unmodulated and the dive height + speeds
    G31 P500 X9 Y21 Z2.26 ; Set Z probe trigger value, offset and trigger height

    ; Inductive Sensor
    M558 I1 P5 H4 F250 T2000 ; Set Z probe type to digital and the dive height + speeds (I1 = invert signal)
    G31 X9 Y21 Z2.26 ; Set Z probe trigger value, offset and trigger height
    [/c]

    I'm constantly amazed at the ease of changing things with the Duet!

    Richard

    (edited to fix typos)


Locked
 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.