Using LJ12A3-4-Z/BX sensor as Zmin

699 views
Skip to first unread message

Tom Mertens

unread,
May 10, 2017, 3:47:06 AM5/10/17
to Smoothieware Support
Hi all,

So I have a LJ12A3-4-Z/BX sensor with a blue tip and I've wired it up to the 5v Zmin connection on the board. The LED light lights up when it's in contact with something metal but it still doesn't trigger when the firmware classes it as an endstop, probe or both. 

Any help appreciated. 

Cheers,

Tom

Arthur Wolf

unread,
May 14, 2017, 4:21:17 PM5/14/17
to Smoothieware Support
Hey.

First, please know these sensors' popularity is completely undeserved. They are unreliable, often aren't reproducible enough, and probe a surface instead of a point, which is very bad for probing work.
Some people like them, but I fear it's just because they don't realize their problems are coming from it ( of course some people will just be lucky too ).
A mechanical probe or something like a bltouch is infinitely better.

About your probe, can you check the output with a voltmeter ?

Cheers.

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--
Courage et bonne humeur.

Benoit Miller

unread,
May 17, 2017, 3:06:52 PM5/17/17
to Smoothieware Support
These sensors are not supposed to work at 5V, they are unreliable at best at this voltage. You should use 12V or 24V or whatever your "standard" voltage is.

Then, you need to use additional components to filter out the 12V, or you will blow your Smoothieboard. Since you have an NPN sensor, I recommend using a diode instead of a voltage divider, this is simpler and works regardless of controller or voltage.
  • Connect the brown wire of the sensor to your printer voltage (12V or 24V or...)
  • Connect the blue wire to GND
  • Connect the black wire to the cathode of the diode (I used an 1N4148...)
  • Connect the anode of the diode to the input of your board
  • Enable pull-ups on the input of your board (in your firmware config)

Now your board will see 3.3V when the sensor is idle (because of the pull-up), and 0.6V when the sensor is triggered.


BWN -------- 12V/24V
BLK ---|<--- Input
BLU -------- GND


Careful about the diode orientation. The diode is used to block the 12V from reaching the controller when the sensor is idle. If you wire it the wrong way you will damage your board. Use a multimeter to check before connecting the BLK wire to the endstop pin.


@Arthur, you might be interested in this video (https://www.youtube.com/watch?v=il9bNWn66BY) which debunks many claims about sensors. Spoiler: these inductive sensors are the most reliable ones, way ahead of BLTouch. :)


Cheers,

Ben.

Arthur Wolf

unread,
May 17, 2017, 3:30:40 PM5/17/17
to Smoothieware Support
I saw the video when it came out, it doesn't adress any of my concerns, in particular the fact that they probe a surface not a point, and that temperature, pressure and other factors can affect the result.
You'll note these are problems the datasheets for these sensors actually point out, last time I checked ...
They are essentially potentially good as endstops, but bad as probes ...

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Benoit Miller

unread,
May 18, 2017, 8:21:34 AM5/18/17
to Smoothieware Support
The fact that they probe a surface and not a point is not an issue if your surface is flat and you probe sufficiently away from the edge. If the surface is not flat, why are you using it as a print bed? :)

I agree that blindingly using these sensors without understanding their limitations is foolish, but it is equally fooling to dismiss them outright. Some very successful commercial printers rely on these sensors and their users have no issues (PrintrBot, Prusa i3 MK2...). This has been my experience as well, the temperature variation has never been an issue for me. Note that I am specifically talking about inductive sensors. Capacitive sensors are near useless because they are affected by everything around them (air pressure, humidity, temperature, etc), as you point out.

Now I no longer use these sensors, because they do not sense the actual printing surface, you still have to add some offset to allow for your PEI, glass, or whatever. This makes them useless for mesh/grid leveling (maybe this is what you meant?). However they work quite well for bed tilt compensation (3 point leveling).

Cheers,
Ben.
Reply all
Reply to author
Forward
0 new messages