NPN NO issue on CoreXY

pugzor · 10 Dec 2017, 11:30

Hi there,

Upgraded my Hypercube Evolution from a MKS SBase to a Duet WiFi last week. So far I'm very, very happy with it, and I can't see myself using another board any time soon for future printers! I've got most things wired and working now. Case fans are working (love the additional outputs on the board, no need for an extra FET), all steppers are tuned, everything (bar one) is working as well as it can be prior to laying down plastic.

Currently using a V6 clone and inductive proximity sensor, but as soon as I get them working with the Duet I'll be printing out mounts for a genuine E3D Titan Aero and BLTouch combo. Just need to print well enough to do that! Unfortunately I was having issues with layer shifts which I ended up putting down to the SBase, so it was an early upgrade to the Duet.

The problem I'm currently facing is that my proximity sensor, a LJ12A3-4-Z/BX, seems to be powered up and detecting metal just fine, but I'm not able to get it to act as an endstop yet. I've followed the steps on the wiki (which is fantastic) but I've not had any joy. Figured I'd post in here to confirm I've done everything from wiring through to config.g modification. M119 doesn't see the endstop as being triggered when I place metal near it (or when there's no metal near it).

Here's a picture of my wiring, just to confirm it's correct. It's plugged into the E0 endstop, next to the the X endstop. If it is correct I'm very happy for it to be used on the wiki as an example. Please forgive my dodgy soldering skills, and of course, I have covered it with insulation tape.

https://goo.gl/uFtvuv (jpg hosted on Google Drive)

Here's a copy of my config.g file too (converted to txt to play nicely with Google Drive) which hopefully will shed some light. Maybe there's a dependency I'm missing?

https://goo.gl/Bi3Gmx (txt hosted on Google Drive)

Any help would be greatly appreciated!

Andy

whosrdaddy · 10 Dec 2017, 12:11

Had a look at you config, seems fine to me.
The only thing I would change is the line:

M558 P4 X0 Y0 Z1 F600 T6000 I1 H0 ; Set Z probe type to mode 4, Z axis only

to

M558 P4 X0 Y0 Z1 F600 T6000 I1 H5 ; Set Z probe type to mode 4, Z axis only, 5mm dive height

Normally when you put metal under your sensor, the Z-probe value should show 1000 in DWC main screen.

pugzor · 10 Dec 2017, 12:29

@whosrdaddy:

Had a look at you config, seems fine to me.
The only thing I would change is the line:

M558 P4 X0 Y0 Z1 F600 T6000 I1 H0 ; Set Z probe type to mode 4, Z axis only

to

M558 P4 X0 Y0 Z1 F600 T6000 I1 H5 ; Set Z probe type to mode 4, Z axis only, 5mm dive height

Normally when you put metal under your sensor, the Z-probe value should show 1000 in DWC main screen.

Thanks mate! Out of interest, what does the dive height do exactly?

Didn't fix the issue, unfortunately, but thanks for the tip.

whosrdaddy · 12 Oct 2017, 12:36

The dive height is the height it will probe at.
Do you get 1000 readout for the z-probe in the webinterface when you put metal under the sensor?

pugzor · 10 Dec 2017, 12:39

@whosrdaddy:

The dive height is the height it will probe at.
Do you get 1000 readout for the z-probe in the webinterface when you put metal under the sensor?

Unfortunately not. Having said this though, earlier I did notice a number (maybe in the 400s) there when there was no metal under it before I started playing with config.g quite a while ago. What would be the next port of call?

whosrdaddy · 10 Dec 2017, 12:43

the value needs to be 1000, if not then you have an electrical issue as your config seems to be correct.
What type of diode did you use?

dc42 · 10 Dec 2017, 14:47

The Z probe reading should be 0 when the probe is not triggered, and 1000 when it is (i.e. metal below it). Don't try to use the Z probe until that is working. With probe type 4 selected in your M558 command, you shouldn't get a reading of 400, except perhaps very briefly while the probe is switching between triggered and not triggered.

Some users have reported that using an ordinary silicon diode in that position doesn't work and they have to use a Schottky diode instead. I made a change in the firmware somewhere between 1.18 and 1.19.2 to alter the threshold, so a normal silicon diode is more likely to work with recent firmware than with older firmwares.

pugzor · 10 Dec 2017, 20:02

Thanks gents!

Used a BAT43 as per the instructions in the wiki. Image in my original post shows how it was wired - is that correct?

dc42 · 10 Dec 2017, 23:21

If the black, brown and blue wires from the sensor are ground, positive supply, and output, then the wiring looks correct to me. Is it definitely an NPN output sensor? What voltage can you measure between the blue wire from the sensor and ground when the probe is and is not triggered?

pugzor · 12 Dec 2017, 11:52

@dc42:

If the black, brown and blue wires from the sensor are ground, positive supply, and output, then the wiring looks correct to me. Is it definitely an NPN output sensor? What voltage can you measure between the blue wire from the sensor and ground when the probe is and is not triggered?

Appreciate your help with this. Ended up just going for a modification to put a mechanical endstop for the Z. Already had one and since I had to open up the back anyway, seemed just as easy (and frankly more reliable - not a fan of the inductive sensor). Sometimes it's easier to reformat a PC than try and remove a virus!