Welcome! Log In Create A New Profile

Advanced

New experimental firmware: all kinematics in host

Posted by KevinOConnor 
Re: New experimental firmware: all kinematics in host
June 22, 2018 08:16AM
Thank you sir (takes a bow).
Re: New experimental firmware: all kinematics in host
June 30, 2018 02:29PM
Kevin - I got a new extruder so needed to update my extrusion steps. I took the opportunity to update to the latest Klipper. Seemed to be working fine after the update. I did an 20mm XYZ cube and noticed my X and Y were a little long so I went back and recalculated the steps for them. I changed the X and Y steps and restarted and suddenly my square was way off. If you consider front/left as origin, the cube was drawn from left to right. That front face went back a bit so it was say 1mm back by the time it hit the corner. Same thing for the right side - from where it started to where it ended, it moved 'left' about 1mm. Reverse was true for back and right sides. So I ended up with a parallelogram. I played with the x and y steps some more and it was worse. Printed something bigger and the effect was even more pronounced. I went back to my original steps and it was 'almost' right - probably like its always been honestly and I just never noticed. Trouble is that I'm .2 to .4mm out in each direction on a 20mm square now. So I guess my question is: Would changing the steps for X and Y to different values (and different from each other) create that effect or is something else going on? I did some hardware troubleshooting and I can't find anything, and other than the extruder nothing has changed on the printer in a long time. Thoughts?

Edit: Actually I lied (a little) - on the right side it actually went further 'right' rather than 'left'. i.e. if front-left is origin then as it traveled to front-right it went 'back' a bit. Then on the right side (i.e. front-right to back-right leg) it actually went 'right' a bit. Then it corrected itself on the remaining two legs leaving me with a pretty perfect parallelogram...

Edited 1 time(s). Last edit at 06/30/2018 03:39PM by MarkLogan.
Re: New experimental firmware: all kinematics in host
July 01, 2018 08:34AM
Quote
MarkLogan
Kevin - I got a new extruder so needed to update my extrusion steps. I took the opportunity to update to the latest Klipper. Seemed to be working fine after the update. I did an 20mm XYZ cube and noticed my X and Y were a little long so I went back and recalculated the steps for them. I changed the X and Y steps and restarted and suddenly my square was way off. If you consider front/left as origin, the cube was drawn from left to right. That front face went back a bit so it was say 1mm back by the time it hit the corner. Same thing for the right side - from where it started to where it ended, it moved 'left' about 1mm. Reverse was true for back and right sides. So I ended up with a parallelogram. I played with the x and y steps some more and it was worse. Printed something bigger and the effect was even more pronounced. I went back to my original steps and it was 'almost' right - probably like its always been honestly and I just never noticed. Trouble is that I'm .2 to .4mm out in each direction on a 20mm square now. So I guess my question is: Would changing the steps for X and Y to different values (and different from each other) create that effect or is something else going on? I did some hardware troubleshooting and I can't find anything, and other than the extruder nothing has changed on the printer in a long time. Thoughts?

Hi. If you're having an issue with Klipper, it's best to open a github issue following the directions at: https://github.com/KevinOConnor/klipper/blob/master/docs/Contact.md
In particular, we need the Klipper log file in order to assist.

-Kevin
Re: New experimental firmware: all kinematics in host
August 05, 2018 07:05AM
Can I used board Gen7 with AVR ATMEGA32 on it ?
Re: New experimental firmware: all kinematics in host
August 06, 2018 10:54AM
Quote
Adjoe
Can I used board Gen7 with AVR ATMEGA32 on it ?

The ATMEGA32 isn't currently supported. At a quick glance, it does not look difficult to add support.

-Kevin
Re: New experimental firmware: all kinematics in host
August 09, 2018 08:50PM
Edited. Problem solved in irc.

Edited 2 time(s). Last edit at 08/09/2018 10:32PM by obelisk79.
Re: New experimental firmware: all kinematics in host
September 01, 2018 03:14PM
Hello, does Klipper support auto bed leveling feature?
SEA
Re: New experimental firmware: all kinematics in host
September 01, 2018 10:56PM
Hi KevinOConnor,

I tried to connect display 12864 anet3d (aka RepRap Discount Mini graphical controller) with Tevo Tarantula (MKS Base 1.4)
I used UC1701 configuration, but it seems not yet implemented.
I assume it is planned. If so, when I can expect it to be done?

Edited 1 time(s). Last edit at 09/01/2018 10:57PM by SEA.
Re: New experimental firmware: all kinematics in host
September 02, 2018 11:03AM
Quote
Kespa
Hello, does Klipper support auto bed leveling feature?

Yes. See the config examples in the config/ directory for a description of the parameters needed to configure it.

-Kevin
Re: New experimental firmware: all kinematics in host
September 02, 2018 11:05AM
Quote
SEA
Hi KevinOConnor,

I tried to connect display 12864 anet3d (aka RepRap Discount Mini graphical controller) with Tevo Tarantula (MKS Base 1.4)
I used UC1701 configuration, but it seems not yet implemented.
I assume it is planned. If so, when I can expect it to be done?

If you're experience an issue with Klipper, please follow the directions at: https://github.com/KevinOConnor/klipper/blob/master/docs/Contact.md

-Kevin
Re: New experimental firmware: all kinematics in host
September 08, 2018 02:33PM
Just experienced another issue with Klipper:

MCU 'mcu' shutdown: Rescheduled timer in the past
This generally occurs when the micro-controller has been
requested to step at a rate higher than it is capable of
obtaining.

Running Klipper on Arduino Mega2560, it is an old file I printed before with another firmware. I do not remember what the slicer settings were, but do not think that the speed goes above 120 mm/s (95K steps/sec).

klippy.zip
Re: New experimental firmware: all kinematics in host
September 09, 2018 10:51AM
Quote
Dimitar
Just experienced another issue with Klipper:

MCU 'mcu' shutdown: Rescheduled timer in the past
This generally occurs when the micro-controller has been
requested to step at a rate higher than it is capable of
obtaining.

Running Klipper on Arduino Mega2560, it is an old file I printed before with another firmware. I do not remember what the slicer settings were, but do not think that the speed goes above 120 mm/s (95K steps/sec).

The log indicates that an XY diagonal move with F9000 was requested just prior to the error. That's 150mm/s, and on your config that translates to ~120K steps per second with two active steppers. That is greater than the benchmark rate for the atmega2560. (To be clear, the benchmarks are with the micro-controller doing nothing but stepping - you should not expect to be able to sustain the full benchmark rate during a real-world print.)

In the future, please use github to report an issue with Klipper. See: https://github.com/KevinOConnor/klipper/blob/master/docs/Contact.md

-Kevin
Re: New experimental firmware: all kinematics in host
September 09, 2018 11:53AM
Thank you about the clarification, if understand correctly the requested speed exceeded the MCU capabilities, as well as it cancelled the print. It was good that it was on the beginning, so no much material was lost, but in such case shouldn’t there be some error checks, to prevent cancelling of prints and loss of material and time ?

Regarding GitHub, as I stated there, my accounts were deleted and I submitted a complain with the Bulgarian authorities regarding their practices. There many problems with GitHub but they are unwilling to change their practices etc.
Re: New experimental firmware: all kinematics in host
October 24, 2018 02:15PM
Again issue with Klipper. This time using Arduino Due (SAM3X8E):

MCU 'mcu' shutdown: Stepper too far in past
This generally occurs when the micro-controller has been
requested to step at a rate higher than it is capable of
obtaining.

F6000 suppose to be ok ?

Also would it be possible to add alternative commands ? For example, instead of RESTART and FIRMWARE_RESTART etc. I would prefer: RESET, FIRMWARE_RESET etc.

klippy.zip
Re: New experimental firmware: all kinematics in host
October 28, 2018 11:40AM
@KevinOConnor Is it possible to use an Arduino nano with an ATmega328P (old bootloader) as an aux-board


Why you might think...
Well I have one lying around and I need some analog pins for temperature readings...

/Martin

Edited 1 time(s). Last edit at 10/28/2018 03:35PM by Nitram.
Re: New experimental firmware: all kinematics in host
October 29, 2018 12:02PM
Quote
Nitram
@KevinOConnor Is it possible to use an Arduino nano with an ATmega328P (old bootloader) as an aux-board

The atmega328p is basically the same as the atmega328 so it should work fine. I added the 328p to the "make menuconfig" selection (just so avrdude doesn't complain when flashing). Grab the latest Klipper code to use it.

Separately, I don't monitor this forum much - for Klipper topics it is best to use the contact methods described at: https://github.com/KevinOConnor/klipper/blob/master/docs/Contact.md

Cheers,
-Kevin
Re: New experimental firmware: all kinematics in host
October 29, 2018 04:39PM
Thanks @KevinOConnor

I will try the mailing list the next time...


/Martin
Re: New experimental firmware: all kinematics in host
December 10, 2018 12:58AM
I wish to make a tool changer type printer.
Is there any cheap board (below 100$) which I can use?
Would klipper be helpful to substitute the lake of 32bit board thereof.

Kindly suggest.
Re: New experimental firmware: all kinematics in host
January 25, 2019 12:18PM
Would it be worth getting TMC2130 drivers and using sensorless homing? Specifically on the Zaxis
Re: New experimental firmware: all kinematics in host
January 29, 2019 12:49PM
@kdodman, not so sure about TMC2130, probably you have to test it by yourself. On Zaxis quite unlikely those to fit ... . From what I read about Z-axis not recommendable at all to use senselessness homing. But probably someone with more experience can share their thoughts too. I use LV8729, not as silent as TMC2208, still they work fine.

@KevinOConnor, again problems:

DELTA_CALIBRATE

18:12:27.946: // Internal error on command:"PROBE"
18:12:27.946: // Once the underlying issue is corrected, use the
18:12:27.947: // "FIRMWARE_RESTART" command to reset the firmware, reload the
18:12:27.947: // config, and restart the host software.
18:12:27.947: // Printer is shutdown
Attachments:
open | download - klippy.zip (113 KB)
Re: New experimental firmware: all kinematics in host
January 29, 2019 03:33PM
@kdodman the TMC2130 need a certain minimum speed for the homing to work. So if the axis moves to slowly, then the back EMF, that is used for the homing, is a too weak signal.
Re: New experimental firmware: all kinematics in host
January 29, 2019 10:02PM
FYI, I don't monitor this forum much - for Klipper topics it is best to use the contact methods described at: https://github.com/KevinOConnor/klipper/blob/master/docs/Contact.md

@kdodman - for sensorless homing you may wish to check out: https://github.com/KevinOConnor/klipper/blob/master/docs/Sensorless_Homing.md

-Kevin
Re: New experimental firmware: all kinematics in host
January 31, 2019 04:09AM
@KevinOConnor, the described from you "contact methods" are not ok. Also you tend to close issues on Github without they to be solved etc.

As it was stated earlier, there are still problems with Klipper, currently when issues: DELTA_CALIBRATE command, I'm getting this respond:

// probe at 0.000,-0.000 is z=387.459954

Yesterday, it did disjoint the arm of a switch I use as a Z-probe etc.

Before begin to use Klipper, I was using Repetier, had some issues with it. Anyone tested the latest versions of Repetier or Marlin, which might be better for Delta 3D printer ?
Attachments:
open | download - klippy.zip (53.8 KB)
Sorry, only registered users may post in this forum.

Click here to login