Octoprint/Klipper diagnostics

When you install the kiauh way:

klippy.log is in ~/printer_data/logs 

read log via 

~/klipper/scripts/logextract.py ~/printer_data/logs/klippy.log

When you install the Octopi way..

Klippy.log is in tmp.

read Klippy.log using  ~/klipper/scripts/logextract.py  i.e

cp ~/tmp/klippy.log . 

~/klipper/scripts/logextract.py  klippy.log 

Octoprint logs are in:

~/.octoprint/logs

its not uncommon to have klipper fail to connect if you use mcu_pi.

to restart klipper, type following in linux login

sudo service klipper restart

systemctl stop octoprint

service klipper restart

systemctl start octoprint

to add debug into .cfg

{ action_respond_info("_PRINT_START") }

alias cssh='ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null'

sudo service klipper stop;rm klippy.log ;sudo service klipper start

swap between octoprint and 

sudo systemctl disable octoprint.service

sudo systemctl enable moonraker.service

or

sudo systemctl enable moonraker.service

sudo systemctl enable octoprint.service

Voron Resonance testing

 Based on https://www.klipper3d.org/Measuring_Resonances.html  (I used the pins pictured, not in the table)

ADXL345 pin	RPi pin	RPi pin name
3V3 (or VCC)	01/17	3.3v DC power
GND	06/25	Ground
CS	24	GPIO08 (SPI0_CE0_N)
SDO	21	GPIO09 (SPI0_MISO)
SDA	19	GPIO10 (SPI0_MOSI)
SCL	23	GPIO11 (SPI0_SCLK)

The accelerometer, while cheap, was difficult to wire.  The connection to the pi involves somewhat difficult to clamp wire.

If I did it again, I would use old USB3 cable.

The instructions are again a bit minimalist, I forgot to follow RPi Microcontroller document.

The key is to zero the klippy.log document, the reload.  Klippy.log usually has the errors.

cd /tmp

rm klippy.log

touch klippy.log

attempt to reload firmware, then 

cat klippy.log

to create the plots

find . -name 'calib*' -exec ~/klipper/scripts/calibrate_shaper.py {} -o {}.png \;

Geared Extruder on Prusa i3 mk3s

Geared Extruder on Prusa i3 mk3s

mk3s has overheating problems on extruder motor. this totally fixed it (also fixed with Mk3s+). don't need new firmware, just marlin codes 

M502 

 needed to add: 

 M92 E518 

 M350 E16 

 save config to eprom with M500

 show with M503

LDO NEMA 17 Geared Extruder Motor LDO-42STH25-1004ACPG3

Torture Toaster testing

 The Prusa Mk3s+ printed it prefect first time.  I suspect because it had a Gear Motor (thus had the extrusion tuned for the motor)

https://www.printedsolid.com/products/ldo-nema-17-geared-extruder-motor-ldo-42sth25-1004acpg3

The Bear Mk3s+ failed on .1, I assume because of slight over extrusion.  Changed the extrusion multiple by 2% and it passed (on PETG..)

M92 E276

M500

The Voron failed even after extrusion correction.

top is Prusa, bottom is voron.

The following is after shaping and small chamber fan. The voron passed except for the "handle" looks ugly (its a long 80 degree overhang).

Voron passed with cheap ABS+ after using Ellis's settings

    https://github.com/AndrewEllis93/Ellis-SuperSlicer-Profiles
    and
    https://ellis3dp.com/Print-Tuning-Guide/
    he has profiles.  You have to pull them into prusa slicer, then you can compare.  (basically you pull in the .ini, it messes everything up, but you kinda rebuild it and I think its good).
    I had to discard all but filament profile.

Voron Trident Formbot kit Build

 

The Voron Trident build took about 40 hours + 40 hours of tuning...I had Covid at the time so I would have been able to build it faster normally, but was able to concentrate on it almost 100%.

I printed the parts with a (untuned) Prusa MK3s and a MK3s Bear Clone.  The Bear clone was placed in the Voron Shipping Box with plastic in front and a space heater...worked very well.

Long parts were difficult (most had a failure or two) because the chamber temp was ~100F (because it was winter and 40F in garage.

The manual was minimalist and I don't think it would work as first printer. 

I had a lot of difficulty with the Z Stop (you have to assemble the z stop prox switch by hand, its a bit goofy.)  

I replaced the hammer head nuts for the left clear panel with t-nuts.  It seems like you have to remove this panel a lot.

Problems:

 There needs to be a wire run on the rear left AB drive frame part.

  If the z stop moves, it caused the head to press against the plate.  I will reprint with PC if it moves again, but it should be very tight.
    the second one is the reason for the ABS w/r XY joint parts.  The system seems like it is built to twist on too much pressure between the bed and the nozzle.
This is pretty awesome.  If your live adjust Z is too low you get a very ugly perimeter, and should stop the print.  Even extremely bad values are unlikely to seriously damage your build plate or nozzle.  In addition, it seems the extruder can still push filament out while live adjust Z is too low.

There is very little room for the camera.  I don't think anything but the pi-cam could fit, and wiring is very tough.

Long Exposure Day Time Photography

While planning for a Long Exposure Daylight Photography workshop I found remote triggering to be artificially difficult.

Using "WiFi remote" app on your phone, completely takes over the phone, preventing the use of a timer, and disabled ISO64 on my D850.

As a result, I decided to use Exposure stacking.

The setup to take the photo is super easy:

Set camera on a tripod, in aperture mode with appropriate depth of focus.  Use ISO for fine adjustment of Exposure time.

Camera must have long exposure NR turned off

Long exposure target exposure time between 1 and 8 seconds.  

If your target is 5 min, that's 300 photos at 1 second.

Target between 4 and 8 seconds per photo.  

Its best to use a N1000 (i.e. 3 stop ND filter) alone.  (multiple filters cause vignetting).

8 seconds Exposure time should be:

1/125 for ND1000

1/1000 for ND1000+ND8

Camera should be on tripod, AF on, VS off

Take the photo and check histogram.

Put the ND filter(s) on, turn off AF

Take another shot and check histogram and vignetting.

Start interval intervalometer.

Nikon 850d has minimum time for internal intervalometer delay of .5 seconds. (750d is 1 sec).

With a remote shutter you can just leave it down, and it will take photo after photo.

For Processing:

download imagemagick

https://imagemagick.org/index.php

put each series in a directory, pay attention to only include 8sec shots.

run imagemagick on the command line

"C:\Program Files\ImageMagick-7.1.0-Q16-HDRI/magick.exe" *.NEF -evaluate-sequence mean OUT_mean.TIF

"C:\Program Files\ImageMagick-7.1.0-Q16-HDRI/magick.exe" *.NEF -evaluate-sequence median OUT_median.TIF

"C:\Program Files\ImageMagick-7.1.0-Q16-HDRI/magick.exe" -resize 25% -delay 20 -loop 0 *.JPG OUT_gif.gif