Hyperfine bed leveling?  

Page 9 / 22
  RSS
3d-gussner
(@3d-gussner)
Reputable Member



Is there a chance that the finer adjustment capability will be fed into a future release?

There are two pull request to get Hypefine tuning in the Prusa firmware, hopefully one get in:
https://github.com/prusa3d/Prusa-Firmware/pull/211
https://github.com/prusa3d/Prusa-Firmware/pull/220

Posted : 10/10/2017 1:47 pm
PJR
 PJR
(@pjr)
Moderator




Is there a chance that the finer adjustment capability will be fed into a future release?

There are two pull request to get Hypefine tuning in the Prusa firmware, hopefully one get in:
https://github.com/prusa3d/Prusa-Firmware/pull/211
https://github.com/prusa3d/Prusa-Firmware/pull/220

Just a quick note to say a big THANK YOU for your work on this and to make one further suggestion:

One Z micro-step is 2.5 microns. Therefore, when saving the values in E^2, why not store the adjustment step count? This would allow an adjustment of up to ±250 microns (±100 steps) to be stored in a single byte.

On-screen adjustments would then be the same increment as that for "Live Z" (2.5 microns, truncating the .5 on display).

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…...
Posted : 10/10/2017 2:19 pm
3d-gussner
(@3d-gussner)
Reputable Member


Just a quick note to say a big THANK YOU for your work on this and to make one further suggestion:

No thank YOU and Jeff Jordan to make this possible! This all your idea, and i am just trying to get the code right.


One Z micro-step is 2.5 microns. Therefore, when saving the values in E^2, why not store the adjustment step count? This would allow an adjustment of up to ±250 microns (±100 steps) to be stored in a single byte.

On-screen adjustments would then be the same increment as that for "Live Z" (2.5 microns, truncating the .5 on display).

Peter

That sound interesting. Did you check the pull request from Thess? He has a different approach storing the values.
BTW: I don't care if it is abc, d cntr e, fgh or something else, just needs to be easy to understand for everybody.

Posted : 10/10/2017 2:48 pm
PJR
 PJR
(@pjr)
Moderator


Did you check the pull request from Thess?

Not really, don't understand any of that stuff!


BTW: I don't care if it is abc, d cntr e, fgh or something else, just needs to be easy to understand for everybody.

Doesn't matter how the points are organised; but I think the storage in E-Steps is the way to go, with G80 parameters in microns.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…...
Posted : 10/10/2017 3:18 pm
Nomad965
(@nomad965)
Trusted Member

@JeffJordan

Would you be kind enough to post the stl files for the mini A-H squares please.

Thank you

Prusa i3 MK2s MK2.5s Mk3 Mk3s BearPrusa Zaribo 320 - Kit Build - Rc Driver - Tweak freak...
“I know nothing about surpassing others. I only know how to outdo myself.” -Bushido-...
Posted : 13/10/2017 2:06 am
3d-gussner
(@3d-gussner)
Reputable Member

Hi,

if somebody is interested in the firmware 3.1.0-RC1 including hyperfine bed calibration, filament runout sensor via LCD menu and some other mods/fixes, checkout https://github.com/3d-gussner/Prusa-Firmware/releases

Waldemar aka 3d-gussner

Posted : 16/10/2017 5:35 pm
PJR
 PJR
(@pjr)
Moderator

It seems as though the beds on my printers are getting old and becoming more variable due to heat cycling.

Printing calibration squares fairly regularly started to become somewhat tedious, so I wrote a non-printing calibration routine which requires either a 0.2mm feeler gauge or a double thickness sheet of copy paper...

The attached file simply heats the bed and nozzle and then moves the nozzle to all 9 probe points and pauses for measurement.

The results I obtained using this method were pretty accurate and only required a single calibration square print to confirm.

Please read the notes in the G-code file.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…...
Posted : 16/10/2017 6:38 pm
3d-gussner
(@3d-gussner)
Reputable Member


It seems as though the beds on my printers are getting old and becoming more variable due to heat cycling.

Printing calibration squares fairly regularly started to become somewhat tedious, so I wrote a non-printing calibration routine which requires either a 0.2mm feeler gauge or a double thickness sheet of copy paper...

The attached file simply heats the bed and nozzle and then moves the nozzle to all 9 probe points and pauses for measurement.

The results I obtained using this method were pretty accurate and only required a single calibration square print to confirm.

Please read the notes in the G-code file.

Peter

Great Idea, thanks for sharing.
Using my feeler gauge (0.3,0.1,0.04,0.05,0.06,0.07,0.08,0.15,0.2mm) i can now measure all nine points and adjust the hyperfine values as a start point before using the hyperfine bed calibration g-code.

Posted : 16/10/2017 7:10 pm
JeffJordan
(@jeffjordan)
Moderator

💡 some people asked me about the single *.stl files to slice the little calibration squares themselves (e.g. for the use with other filament materials).
➡ so here they are, just for your convenience:

dem inscheniör is' nix zu schwör......
Posted : 16/10/2017 11:37 pm
jacob.b9
(@jacob-b9)
Active Member

Well... I know what I'm doing tonight...!

Thanks for everyone's hard work on this.

Posted : 24/10/2017 9:45 pm
richard.p13
(@richard-p13)
Active Member

This is exactly what I've been looking for - thanks PJR and JJ.
I mistakenly thought that the newer firmware 3.1.0-RC1 included this 9-point bed correction, so I went ahead and spent a day printing (and re-printing) all the circle and square calibration prints, working out my final values etc only to find that I still only had the old 4-point bed correction options.

@JeffJordan - are you able to PM me the recompiled firmware (HEX) with the 9-point calibration? Unfortunately I don't have enough thread replies to PM you myself.

Cheers,
Rich

...
Posted : 29/10/2017 7:22 am
JeffJordan
(@jeffjordan)
Moderator

3d-gussner has the 3.1.0-RC1 with hyperfine bed levelling online here.

dem inscheniör is' nix zu schwör......
Posted : 29/10/2017 9:50 am
richard.p13
(@richard-p13)
Active Member


3d-gussner has the 3.1.0-RC1 with hyperfine bed levelling online here.

Thank you. I must have D/Led the incorrect version of 3.1.0-RC1.
Cheers, Rich

...
Posted : 29/10/2017 10:08 am
3d-gussner
(@3d-gussner)
Reputable Member


Thank you. I must have D/Led the incorrect version of 3.1.0-RC1.

Hi Richard,

what do you mean incorrect? i must have D/Led?

Waldemar

Posted : 29/10/2017 1:16 pm
JeffJordan
(@jeffjordan)
Moderator


...what do you mean incorrect? i must have D/Led?

guess D/Led = downloaded 🙄

dem inscheniör is' nix zu schwör......
Posted : 29/10/2017 1:26 pm
3d-gussner
(@3d-gussner)
Reputable Member


guess D/Led = downloaded 🙄

Okay, thanks

Posted : 29/10/2017 1:28 pm
Sidbitters
(@sidbitters)
Eminent Member

I have tried that link before and it just comes back to the same page. I used the chrome browser and edge and same thing. I also joined GitHub. I am not sure what my problem is.
edit: Ok, I believe I have here now.

Posted : 29/10/2017 7:46 pm
richard.p13
(@richard-p13)
Active Member



Thank you. I must have D/Led the incorrect version of 3.1.0-RC1.

Hi Richard,

what do you mean incorrect? i must have D/Led?

Waldemar

Hi, sorry for the confusion. In my previous post "incorrect" version meant a copy of the 3.1.0-RC1 firmware that did not have the 9-point hyperfine dbed correction settings (it only had the 4 L/R/F/B inputs). And "D/Led" meant downloaded.

...
Posted : 30/10/2017 3:48 am
Sidbitters
(@sidbitters)
Eminent Member

Anyone know what could be the cause of this error? The Prusa version loads fine and I've gone back and forth a few times.

Posted : 31/10/2017 2:22 am
3d-gussner
(@3d-gussner)
Reputable Member


Anyone know what could be the cause of this error? The Prusa version loads fine and I've gone back and forth a few times.

Hi Ray,

which hex file did you try to upload and where exactly did you get it from?

Waldemar

Posted : 31/10/2017 8:27 am
Page 9 / 22
Share:

Please Login or Register