BLTouch-3 Open Drain - compatible with Duet's Z-Probe input?



  • It seems that version 3 of BLTouch changed the output type to open drain (search for 'open drain' here https://www.antclabs.com/bltouch-2 ) and the timing diagram in same page suggests that it's normally closed.

    This resolves the uncertainty that was expressed here regarding the output characteristics BLTouch (good) and eliminates the need to cut a trace to convert it to 3.3V output (also good)

    Is the DWF compatible with open drain connected to the Z Probe input? I couldn't find a pullup resistor in the schematic but it can be implemented in the MCU.

    The relevant code in my config.g looks like this:

    M574 Z1 S2              ; Set endstop controlled by probe
    M307 H3 A-1 C-1 D-1     ; Disable heater on PWM channel for BLTouch
    

    Edit: also, considering that the new BLTouch does not have active sourcing, will this introduce any significant delay and error in the detection of the Z endstop signal?


  • administrators

    It works according to this post https://forum.duet3d.com/topic/10023/bltouch-v3-0-information-setup-and-o-scope-pictures. If you are worried about the response speed then you can add an extra pullup resistor between the IN and +3.3V pins on the Zprobe connector. Or, if you are using a Duet Maestro or a revision 1.04 Duet WiFi (both of which can tolerate 5V on the IN pin), you can program the BLTouch to do active pullup.



  • Thanks @dc42. The no pullup configuration give ~0.5ms 90% rise time so I decided to go with a 3.3K pullup tied to 3.3V. This way I don't need to rely on M280 for the 5V active mode.

    External rise time with no external pullup:
    0_1557942567392_SCR02.PNG

    External rise time with external 3.3K pullup: (it's interesting that it's pulled up fast only up to ~2.6V and then rises slowly)
    0_1557942540452_SCR03.PNG



  • I ended up adding a 3.3V pull up resistor this way:

    alt text


  • administrators

    @zapta said in BLTouch-3 Open Drain - compatible with Duet's Z-Probe input?:

    External rise time with external 3.3K pullup: (it's interesting that it's pulled up fast only up to ~2.6V and then rises slowly)

    That may just mean that you are using a x10 probe and the capacitor in it isn't adjusted correctly. Most oscilloscopes have a 1kHz square wave test signal output to help you adjust it.



  • @dc42 said in BLTouch-3 Open Drain - compatible with Duet's Z-Probe input?:

    Most oscilloscopes have a 1kHz square wave test signal output to help you adjust it.

    I tested the compensation and it seems to be OK. Also, if I disconnect the input from the board I see a sharp rise up to 3.3V. Not sure what's going on there but the printer works well.



  • @zapta said in BLTouch-3 Open Drain - compatible with Duet's Z-Probe input?:

    3.3K pullup

    I'm having issues with inconsistent probing (closer to 0.1), and i'll rather give a pullup resistor than 5V with a 240ohm resistor. What's the spec's on the pullup your using? And did using other trigger values in G31 help you anything with consistancy?



  • @exerqtor said in BLTouch-3 Open Drain - compatible with Duet's Z-Probe input?:

    What's the spec's on the pullup your using? And did using other trigger values in G31 help you anything with consistancy?

    It's a 3.3k resistor on the z probe connector, between the Z Probe In and +5V pins. I did't do anything special in my config file but also didn't measure consistency.

    How did you measure consistency?



  • @zapta
    Ok, 1/4 or 1/8W?

    Well i haven't done a "scientific test", but when i run a highres probing on my bed (244 points on a 328x328 bed) with 4mm glass it turns up with uptwards of 0.5mm deviation between points around the middle of the bed and other random spots around the surface. And can't understand that even being posible. I've got max 5 probes with 0.05 allowed deviation between the points.



  • @Exerqtor

    1. It dissipates less than 10mw so any 3.3k resistor should do, even 1/8W. If you have handy a resistor of similar value (e.g. 2.7k or 4.7k) they should also do.

    2. To test the consistency of the probe you need to test at same point. What you described may be due to bed surface or gantry rails curvature or misalignment.



  • 1: Ah ok, haven't got any so i have to order anyways 😛

    2: Yeah i know, but bu the 0.5mm difference that comes up between probing points that have 20mm between each other ain't true. As you say it might be something structural with the printer (haven't upgraded to MGN's yet). Still i have a feeling that the BLTouch reads somewhat inconsistent at sometimes to, it has to probe one spot 3-4 times before it's happy.


 

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