Unable to complete initial camera calibration

120 views
Skip to first unread message

Matt Callaghan

unread,
Sep 9, 2022, 10:58:10 AMSep 9
to OpenPnP
Hello,

I have been working through the guide to get my Liteplacer and smoothieboard set up to work with OpenPnP.

I am currently stuck on the "Primary calibration fiducial position and intiial camera calibration" in issues and solutions.

A green crosshair and circle should appear around the fiducial, this is not the case for my install.
In addition to the graphics not appearing when i hit the auto-adjust button the program waits until the command timeout has elapsed and says "Timed waiting for move to complete".
If i manually adjust the feature size box the program hangs for several seconds after i press the up or down key, the program recovers and becomes responsive after a few seconds but nothing changes.

See attached a screenshot of my issues and solutions page on the initial camera calibration as well as a TRACE level log dump.

I have tried running openpnp on a linux (ubuntu) machine and my windows laptop, both exhibit the same issues.

If anyone had some suggestions as to what to try I would be very appreciative.

Thanks,
Matt
PnP_problems.PNG
openpnp_log_errors.txt

mark maker

unread,
Sep 9, 2022, 11:18:32 AMSep 9
to ope...@googlegroups.com

Hi Matt

you got some unexpected sequences in the log.

Please send the machine.xml

_Mark

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/34e3b1e6-5a89-412d-8e9d-0944748f678dn%40googlegroups.com.

Matt Callaghan

unread,
Sep 11, 2022, 9:18:04 PMSep 11
to OpenPnP
Hi Mark,

Thanks for your reply.

See attached for the machine.xml

Matt
machine.xml

mark maker

unread,
Sep 12, 2022, 4:08:57 AMSep 12
to ope...@googlegroups.com

Hi Matt

you got a Smoothie firmware that is not in Grbl mode. This is the historical default for OpenPnP, i.e. OK.

OpenPnP actually supports both modes, which it detects by the X-GRBL_MODE variable in the firmware detection, yours is at 0 =off:

But you seem tho have manually set a HOME_COMMAND to $H which is the Grbl mode dialect. Maybe that was inherited from when you had a different firmware.

OpenPnP Issues & Solutions will only propose a HOME_COMMAND if it is empty, because those tend to be highly machine dependent. Set it to empty, press Apply:

Then use Issues & Solutions to propose the proper HOME_COMMAND. Accept it:

I'm not entirely sure this will already solve the reported problem. But let's take it one step at a time.

_Mark

Matt Callaghan

unread,
Sep 12, 2022, 4:45:15 AMSep 12
to OpenPnP
Hi Mark,

Thanks for taking the time to help me.

I have applied the changes to the home command as you suggested. Homing works as intended.

Unfortunately it has not yet solved my original issue in the initial camera calibration page.
See attached for a log where the last step i did was press the "auto-adjust" button on the initial camera calibration issues and solution page.

I also tried changing the naming of the rotation axis from C to A however this didn't do anything (rotation worked before and after the axis name change).

Matt
machine.xml
openpn log after home commandf reset.txt

mark maker

unread,
Sep 12, 2022, 5:18:23 AMSep 12
to ope...@googlegroups.com

Hi Matt

the only thing that remains strange are these:

You've set your driver to Full 3rd Order Control. Unfortunately, Smoothie does not provide that. you probably meant Simulated3rdOrderControl:

But I cannot see how this would explain the error. Still: please set to ModeratedConstantAcceleration for now.

Also strange is this sequence:

2022-09-12 16:20:18.768 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << Smoothie
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.769 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.770 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.770 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:18.770 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:20:21.769 GcodeDriver DEBUG: [serial://ttyACM0] >> G21 ; Set millimeters mode, 10000

A ton of "ok"s`? But why? It sort of points to a serial comm problem.

However, the sequence in the log does not point to a comm problem, the M400 is confirmed.

2022-09-12 16:31:05.599 Scripting TRACE: Scripting.on Camera.BeforeSettle
2022-09-12 16:31:05.599 GcodeDriver DEBUG: [serial://ttyACM0] >> M400 ; Wait for moves to complete before returning, 10000
2022-09-12 16:31:05.600 GcodeDriver$ReaderThread TRACE: [serial://ttyACM0] << ok
2022-09-12 16:31:05.600 GcodeDriver TRACE: [serial://ttyACM0] confirmed M400
2022-09-12 16:31:15.600 Scripting TRACE: Scripting.on Camera.AfterSettle
2022-09-12 16:31:15.601 MessageBoxes DEBUG: Error: java.lang.Exception: Timed out waiting for move to complete.

The fixing of the homing command was all done in one session. I'm not sure how, but maybe there was still some "strangeness" lingering from using $H in the same session.

Please exit OpenPnP, reset/unpower Smoothie, and retry.

_Mark

Matt Callaghan

unread,
Sep 12, 2022, 8:05:56 AMSep 12
to OpenPnP
Hi Mark,

I changed to the simulated3rdordercontrol type however my rotation axis stopped working.
I worked through the issues and solutions from the beginning again and it prompted me to change it to >Toolpathfeedrate, this returned the rotation axis working, on progressing further the issues and solutions page suggested moving to >Moderatedconstantacceleration, which i did so and the rotation axis still worked fine.

However I have now found that if i start openpnp fresh and power the machine on but don't home the machine the "Primary calibration fiducial position and initial camera calibration" page starts working.
It will go through the auto-adjust process and finds that 52 gives the best result, the graphics overlay on the camera works.
On pressing accept at the bottom of page (of the initial camera calibration) the machine drives to home and stops, eventually it throws an error "Timed out waiting for move to complete".

If I home the machine before attempting the initial camera clibration the graphics overlay on the camera does not work and the auto-adjust does not work.

I have included two text files, one of them is from the terminal window where i started openpnp which seems to have a few extra java errors compared to the log from within openpnp.
"openpnp log after changing motion type.txt" contains the log from within openpnp after i had done >start openpnp > turn on machine > auto-adjust > accept > "Timed out waiting for move to complete" > homed machine > auto-adjust (failed with command timeout)

It is getting late where I am so I will have to pick this up in the morning, I will try the latest machine.xml I have generated so far on my windows install and see if anything is different.

Thanks again for your help Mark.

Matt
openpnp log after changing motion control type.txt
openpnp log after failed image calibration then home.txt

Matt Callaghan

unread,
Sep 12, 2022, 8:08:22 AMSep 12
to OpenPnP
I forgot to attach the latest machine.xml to the previous post.
machine.xml

Matt Callaghan

unread,
Sep 16, 2022, 1:13:39 AMSep 16
to OpenPnP
After hitting my head against the wall for a while I decided to start with a fresh OpenPNP configuration, new SD card on the smoothieboard (same config.txt and firmware) and a more powerful computer with xubuntu as the OS.
I am finally able to pass the initial camera calibration, still going with the rest of the setup but I am a very happy person now I can progress.

Thanks again for your help Mark, I'm sorry to have wasted your time chasing gremlins.
Reply all
Reply to author
Forward
0 new messages