1st layer problems - in depth look at software PINDA problems (and solutions!)
 
Notifications
Clear all

1st layer problems - in depth look at software PINDA problems (and solutions!)  

Page 15 / 15
  RSS
stahlfabrik
(@stahlfabrik)
Honorable Member
Topic starter answered:
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

I think they shortly wanted to use the "pinda" function in a development commit, but then switched back to the "bed" function. I guess, the pinda function might be used in the future - I have not looked in the details of it yet.

The temperature is taken into account for each mesh bed leveling measurement. That is why it would be a bad idea to slow bed leveling down or something like this.

If your bed is not flat - have you looked into a mechanical improvement? http://prusaowners.com/wiki/index.php?title=Bed_Leveling_with_Wave_Springs

Posted : 10/07/2018 4:29 pm
H311b3nt
(@h311b3nt)
Active Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

does updating firmware (i see that 3.3.0 is out) wipe the values we have manually entered? just curious.

Posted : 12/07/2018 2:58 pm
stahlfabrik
(@stahlfabrik)
Honorable Member
Topic starter answered:
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)


does updating firmware (i see that 3.3.0 is out) wipe the values we have manually entered? just curious.

I think not - Just do "M861 ?" before and after the update to check. I think only when you do a factory reset (which I tend to do after every update) the EEPROM is rewritten completely.

Posted : 13/07/2018 10:34 am
stahlfabrik
(@stahlfabrik)
Honorable Member
Topic starter answered:
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

We wrote a how-to article on the prusaowners.com wiki on that topic.

Have fun:

http://prusaowners.com/wiki/index.php?title=Manual_print_based_PINDA_temperature_calibration

Posted : 21/08/2018 2:06 pm
Spacemarine
(@spacemarine)
Estimable Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

Link isn't working anymore

Posted : 30/12/2018 5:38 pm
bobstro
(@bobstro)
Illustrious Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)


Link isn't working anymore

I believe this page is the one.

My notes and disclaimers on 3D printing

and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan

Posted : 30/12/2018 8:58 pm
stahlfabrik
(@stahlfabrik)
Honorable Member
Topic starter answered:
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)



Link isn't working anymore

I believe this page is the one.

Oh it is! Thank you! The wiki was moved to GitHub:-)

Posted : 31/12/2018 12:04 am
bobstro
(@bobstro)
Illustrious Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)


[...] Oh it is! Thank you! The wiki was moved to GitHub:-)
Thank you for the great content!

My notes and disclaimers on 3D printing

and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan

Posted : 31/12/2018 4:17 am
Spacemarine
(@spacemarine)
Estimable Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)


Oh it is! Thank you! The wiki was moved to GitHub:-)

You might also want to adjust the link on the first post in this thread? 🙂

Posted : 31/12/2018 11:35 am
Spacemarine
(@spacemarine)
Estimable Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

I've dived into this topic very deep in the last few days and I must say that the discussion with bobstro was extremely helpful. And reading all the content from stahlfabrik made me understand the problems much better.

But here is what I don't fully get: What is the technical advantage of manually calibrating the first layer hight as described in the wiki ( https://github.com/PrusaOwners/prusaowners/wiki/Manual_print_based_PINDA_temperature_calibration ) compared to the built in method in the prusa firmware?

I just ran the built-in calibration routine on firmware 3.5.1-1778 and watched the temperatures to get a clue what the printer is doing, as I didn't find any documentation on the procedure (Or did I miss it?)

It looks to me like the printer is doing the following:
1. Let pinda cool down to 35 °C
2. move extruder to induction position and move Z-axis repeatedly to get multiple pinda trigger events
3. heat the bed to heat pinda to 40 °C and repeat step 2
4. heat the bed more to heat pinda to 45 °C and repeat step 2
6. heat to 50 °C etc. etc.

In my understanding, if we don't loose any motor steps in between, this will result in a table which contains triggering hights for temperatures from 35 to 60 °C in 5 °C steps.

This is also the method I would habe chosen intuitively if I had to program the firmware.

As I see it, this method has several advantages:

- it runs fully automated
- it eliminates user-error, as there is no uncertainty of dialing in different first-layer hights on each temperature step

What is your opinion on this?

Posted : 31/12/2018 11:57 am
stahlfabrik
(@stahlfabrik)
Honorable Member
Topic starter answered:
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

The sad thing is that this forum does not let me edit my initial post anymore:-(

The thing about the build in automatic method is that the results it yields are bad for my printer. Really different from the results I get when doing the "print based, manual approach".

Other than that: The way it is implemented (idea wise) is also what I would do instinctively - no idea why it works so bad:-)

Posted : 02/01/2019 4:15 pm
Spacemarine
(@spacemarine)
Estimable Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

Ok, that makes sense. Maybe it is just a simple programming error in the algorithm? Who knows. In theory it should work great.

Do you have a comparison of your "hand-optimized" results and the results of the integrated algorithm? Maybe that could give us a clue.

Posted : 02/01/2019 4:51 pm
stahlfabrik
(@stahlfabrik)
Honorable Member
Topic starter answered:
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

No recent data. I think I described it in the first post

Posted : 02/01/2019 7:12 pm
Spacemarine
(@spacemarine)
Estimable Member
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)

Here is an interesting idea, that might explain why the integrated routine fails: It relys on the assumption, that the bed shape doesn't change with temperature, which is not true.

https://github.com/prusa3d/Prusa-Firmware/issues/1454

Posted : 15/01/2019 11:29 pm
This_Dude_Jay
(@this_dude_jay)
Eminent Member
RE: 1st layer problems - in depth look at software PINDA problems (and solutions!)

Hey there,

very informative thread first of all, great work 👍

If some of you are still interested in the PINDA probing topic, I did a little research on it and had some surprising results. Temperature dependency works way other than I thought and the current compensation method is not really up to it.

https://forum.prusa3d.com/forum/original-prusa-i3-mk3s-mk3-hardware-firmware-and-software-help/an-empirical-insight-into-a-major-flaw-of-pinda-v2-bed-level-probing-and-why-7x7-makes-things-even-worse/

Part 2 is not really necessary to read, but I thought some people love to dig into raw data.

What I recommend to you is part 3, there everything is explained and summarized.

You might find this interesting as well I hope, as it could possibly be a little gamechanger to consistent and accurate probing.

Cheers

Posted : 01/02/2021 3:52 pm
Page 15 / 15
Share: