Unable to home z axis

Kayjay

@Phaedrux
Well, there is a small improvement, the bed now rises in the steps I have set in homeall, I have to keep homing but by hte time it reaches the top it crashes or rather would if I don't do an emergency stop. I've also tried holding my finger under the probe whilst it is homing , the LED lights but it doesn't stop, neither does it stop as it lights just before I hit emergency stop.

; homeall.g
; called to home all axes
;
; generated by RepRapFirmware Configuration Tool v3.1.4 on Sat Aug 15 2020 09:44:17 GMT+0100 (British Summer Time)
G91                     ; relative positioning
G1 H2 Z-25 F6000          ; lift Z relative to current position
G1 H1 X-305 Y-205 F1800 ; move quickly to X and Y axis endstops and stop there (first pass)
G1 H2 X5 Y5 F6000       ; go back a few mm
G1 H1 X-305 Y-205 F360  ; move slowly to X and Y axis endstops once more (second pass)
G90                     ; absolute positioning
G1 X150 Y100 F6000       ; go to first bed probe point and home Z
G30                     ; home Z by probing the bed

; Uncomment the following lines to lift Z after probing
;G91                    ; relative positioning
;G1 Z5 F100             ; lift Z relative to current position
;G90                          ; absolute positioning

The Z-25 setting is temporary to enable a quicker bed rise so I don't have to hit homeall a hundred and umpteen times. Obviously I stop it crashing before it reaches it's limit

droftarts

@Kayjay does the probe still give sensible numbers in DWC, ie 0 when not triggered, 500+ when triggered?

Can you post some pictures of the Duet and wiring?

Ian

Kayjay

@droftarts Yes, the probe numbers are still correct, 0 and 537. I've also managed to upload the video I made yesterday, I suppose it's out of date now as it shows that the bed doesn't rise after homeall but I thought I'd post it for completion. Here's the link, you may have to wait until 4pm for it to show, it's the earliest it would allow me to show it.

https://youtu.be/Xtr5PvMXFso

I'll also take some stills of the Duet wiring. Is it possible to post pics on here or should I put them somewhere else?

Kayjay

Kayjay

@Kayjay said in Unable to home z axis:

Kayjay

Kayjay

Kayjay

@droftarts
Are the pictures above ok or would you like some more specific areas, particular clear shots of wires and plugs, not a prob if you require them.

Keith

Phaedrux

@Kayjay said in Unable to home z axis:

Well, there is a small improvement, the bed now rises in the steps I have set in homeall, I have to keep homing but by hte time it reaches the top it crashes or rather would if I don't do an emergency stop.

Do you mean that G30 seems to be working now, or is it the G1 H2 Z-25 that is moving it up? That command will move it until it crashes, the probe won't stop it.

Can you get a better photo of the probe wiring and the ir probe itself?

Kayjay

@Phaedrux It's a bit of a mixture, if I comment out G30 the bed will rise whenever Homeall is activated, the head centres to the middle of the bed and the bed will rise the amount I have set in Homeall but if G30 is active ie not commented out then things return to as they were, all stops at the point G30 should home. I have also included two stills of the probe plug on the Duet and the plug on the probe itself.

Phaedrux

What happens if you move the print head to the center and then issue G30 in the console?

Also, can you send G31 and report the result?

Kayjay

Just an intersting item I found in my bits and bobs box. I bought it years ago, it was poorly printed as you can see but the circuit design and construction was good. It's a filament diameter sensor to detect variances in filament thickness and as such the flow through the hotend could be adjusted. I know Marlin was adjusted to enable this to be used and I used it on my old Prusa Mendel to good use but it meant me changing from Repetier (which was far superior at the time and may still be) to Marlin. If anyone wants to make one it can be found on Thingverse.

http://www.thingiverse.com/thing:454584

Keith

Phaedrux

While we still don't know what is wrong with your probe, I had an idea to try switching to manual probing so that you can maybe at least get up and running.

In config.g change your M558 P1 to P0 for manual probing. That will mean that when it gets to the G30 part in your homeall, it should pop up a dialog box asking you to jog the Z axis to touch the nozzle. Let's see if that works.

dc42

In your config.g file you have this:

M558 P1 H5 F120 T6000                   ; set Z probe type to unmodulated and the dive height + speeds
G31 P500 X0 Y25 Z2.5                    ; set Z probe trigger value, offset and trigger height

That looks correct to me, however we can't rule out the possibility that there is a non-printing character in one of those lines that is messing things up. So please send the following commands from the console, with no extra parameters:

M558
G31 T1

and check that they report the expected values.

droftarts

@dc42 said in Unable to home z axis:

So please send the following commands from the console, with no extra parameters:
M558
G31 T1
and check that they report the expected values.

Did this today with Keith, response was correct. One odd thing was probe reading was 0 to 1000, rather than 0 to 530. I suspect a damaged probe.

Ian

Kayjay

Thanks for the responses there Ian, you saved me answering the individual posts. I have the Duet2, both daughterboards and probe/cables all packed up ready to go to the P.O. in the morning.

Kayjay

@dc42

Ian is going to check the board over for me to see if it's faulty or the probe is etc, etc.

If per chance I need a new board are you shipping yet or should I go to a dealer, say best for me woukd be E3D-Online.

Thanks
Keith

droftarts

TL/DR: Weird G30 probing behaviour? Check if you have a sys/deployprobe.g file and a probe that doesn't need deploying!

Just for posterity, I'm posting to say that this issue has finally been resolved, though the Duet does need to be reinstalled in the printer, as Keith ended up sending me his Duet, probe and probe wiring to work out what was wrong!

The main problem was that a Mini Kossel example sys folder had been used as a starting point. The Mini Kossel featured a probe that was deployed by moving the effector around in X, Y and Z in order to 'twang' one of the belts and deploy the probe! The Kossel setup had then been overwritten with a Cartesian setup featuring an IR probe. But this left some files behind, including deployprobe.g, retractprobe.g and homedelta.g.

When G30 is sent, the firmware will run deployprobe.g if it exists. On Keith's machine, this caused a bunch of XYZ moves to be called (making it look like it was trying to home X and Y) even with a plain G30, usually followed by an 'axis not homed' error message and no Z homing. I only tracked this down because I put the SD card from my machine (which is a broadly similar Cartesian machine) into his Duet, and it worked perfectly on a bench setup with 24V PSU and motors on X, Y and Z. So I knew it was something on the SD card, and eventually noticed the deployprobe.g file!

There were a couple of other minor things that I initially thought were causing issues. First, the crimps in the Molex housing at the probe end of the cable where a small gauge, so weren’t held firmly in the housing and the signal pin particularly was prone to loose connection, though the IR probe showed 1000 when triggered, and I’m still not sure why. Though it worked when I tested it, I recrimped the connector with the correct size crimps. Second, the Duet was mounted on an acrylic frame, which was covered in thick paper, but without standoffs (5mm is recommended) and screwed down so that some pins pierced the paper. Acrylic is generally not electrically conductive (but can have surface treatments that are) but can hold a static charge. However, though some pins were touching (I don't know exactly which ones), I don't think this made any difference.

Edit: config.json added for backup: config.json

Ian

Phaedrux

Well done!

Kayjay

@droftarts Slight clarification though it doesn't really matter now. I did use stand off's for the acrylic board they were underneath it, they were screwed into the existing Rumba holes and then the new holes in the adaptor were spaced for the Duet, there just wasn't room for two lots of stand offs under the base board in the bottom of the Bigbox, there's only 70mm space and 15mm is taken by the electronics cooling fan plus the wiring to the board. It was a bad idea by it's creator and as we discussed I'm mounting the Duet to new holes direct to the bottom of the Bigbox electronics compartment.
Keith