Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Interesting! How to you dial in the height where the sensor trigggers? With live z adjust I guess? Nice idea!
Edit: I reread your post and now fully understand. Genius! I will compare after holidays my printed based values with your method.
It is not possible to dial in values above 60 or below 35C.
That would be so awesome!
I did not know that the probe gets hotter than 60C. Lower than 35C on the other hand is so obvious.
It would require a few small changes in the firmware to install increase the range. And then it would require a factory reset for everybody to accommodate the new eeprom layout after the update. At least if done in the most obvious way. It would not be rocket science by far:-)
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Thank you 🙂
Do you think that this process can be implemented using GCODE or as a firmware add on?
BTW, I forgot to mention one thing: when I calibrated, I always took the PINDA top down - not sure if there are any minor backlash or hysteresis issues that should be considered so I went along on the safe side.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Well the automatic calibration is somehow doing an related thing. The only problem is that the values it produces do not work for me at all.
I do not know if live adjust z can be set by gcode. Last time i looked i did not find it.
Do you know if using normal move gcode we could move z for 0.001 mm - I mean the fine granularity the live adjust z menu offers?
I am sure one could fully automate your approach but I still wonder how it differs from the build in automatic approach.
I just think your method is really logical and I will test it on the coming weekend. If it gives similar values as my approach it would be a real time saver.
Edit: if the necessary gcodes would be all there one could implement the process for example as an OctoPrint plugin. Maybe just as a serial talking Python script in the first attempt.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
The issue I had with trying to use Gcode is that it wouldn't move below 0, so I had to use live Z. Tried using relative and absolute coordinate systems, no luck there for me...
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
stahlfabrik, thanks for your heroic efforts with this issue!
I thought I'd share my results -- my manually set adjustments appear to be reasonable close to the old factory defaults which explains why I only needed minor adjustments to my live-Z in the past. However, on updating to the 3.2.1 firmware I initially tried the auto tune option -- what a disaster! Clearly it isn't working as intended.
I'm hoping I never have to adjust again. Fingers crossed.
These were my final adjustments:
M861 S15 I0
M861 S42 I1
M861 S64 I2
M861 S82 I3
M861 S127 I4
Edit: Pinda probe has black body and black cable
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
hey all i just want to know if the calibrate temperature has been fixed as i tried manul doing it but it still seem to be off if there anywhere i can find in detall why of doing it
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
See my last post -- yes the automated method seems completely broken.
The method for manually setting is documented in this wiki: https://github.com/prusa3d/Prusa-Firmware/wiki/Manual-temperature-calibration as well in several posts in this thread. I suggest you create gcode files for each temp: 35.gcode, 40.gcode, 45.gcode, ... 60.gcode with prerequisite setting for PINDA temp and bed temperature as detailed in the wiki using a single layer reasonable large model. Then start with printing 35.gcode, record z-offset, cancel print, clean, repeat with 40.code. It's not hard, but you should budget about 90 mins - 2 hours to complete.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
The issue I had with trying to use Gcode is that it wouldn't move below 0, so I had to use live Z. Tried using relative and absolute coordinate systems, no luck there for me...
I did a first quick attempt to do your method. I have the problem of being slow:-) When I have dialed in live Z until the PINDA triggers, the PINDA is getting warmer and warmer. So I think the PINDA already had like 39C instead of 35C.
I have no idea how one could make that quicker... Any practical tips?
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Two tips that I hope would help:
1. Do not fully heat the bed or nozzle. I stated with only heating the bed to roughly 5 degC higher then the PINDA temperature I wanted to calibrate. Took a few extra minutes but I got it accurate give or take...
2. Look at the PINDA indicator light. I played with it a bit before I started the calibration and could roughly estimate when the PINDA will trigger accordingly.
BTW -
1. I am sure you are aware of it (i didn't could have save some precious time), but the PINDA temperature is showing on the pronterface screen so you don't have to jungle between LCD menus...
2. Was the reading you got on 39degC met your previous calibration results?
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
What about manual calibration for 80 degrees? There is no need to set 'correction'? I have issues with first layer printing PTEG (decreasing offset does not help), where my PLA is spot on :/
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
What about manual calibration for 80 degrees? There is no need to set 'correction'? I have issues with first layer printing PTEG (decreasing offset does not help), where my PLA is spot on :/
I am not sure I understand - does your PINDA get 80C warm? You can only calibrate the range 35 - 60C.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
If anyone finds it useful - I attach the manual print based calibration gcodes that work great for me.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
I just did the manual adjustment test on mine and the resulting offsets came as a bit of surprise to me.
35C - 0.520
40C - 0.525 (correction = 2)
45C - 0.537 (correction = 7)
50C - 0.585 (correction = 14)
55C - 0.680 (correction = 64)
60C - 0.862 (correction = 137)
As you can see the values for >= 55 went skyrocket. I guess this could be an indication of a poor linearity of PINDA (I have the black one). Will ask support for sending me one from the new batch and see if that helps.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Guys... look I don’t know coding or GCoding at all. I’m still learning. Even though I had a Flashforge for over a year and used it everyday. I didn’t realize how much Flashforge was doing for me. This Prusa MK3 is so frustrating. I thought it would be better than my cheap Finder. But I’m struggling between learning a new slicer(s) and this Printer.
Sometimes I get perfect first layers, and sometimes I don’t. Nothing is “just print and walk away”.
However... I was scared to play with this firmware... I don’t understand what I’m supposed to do. Load a firmware? Load G-Code? Anyway... I centered the extruder on the plate and turned on the bed and nozzle to PETG temps. I let it heat all the way up, then homed, then printed a test and did a Z adjustment. It was totally different than my best adjustment so far? Anyway, I readjusted, and then started a print... wow. It seemed to make such a huge difference.
If someone could make instructions for a total newbie that would be awesome. Or a video. Don’t know. I read most of this, and I still am not clear. Too much info to understand.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Sometimes I get perfect first layers, and sometimes I don’t. Nothing is “just print and walk away”.
...
If someone could make instructions for a total newbie that would be awesome. Or a video. Don’t know. I read most of this, and I still am not clear. Too much info to understand.
That sounds like your printer needs that calibration.
First step is to install the latest firmware (3.2.1 as of today).
Then you need to ask more specific - because I do not even now where to start. Have you used Pronterface before? You need that or something similar like the terminal in Octoprint for communicating to the printer. There you could do a "M861 ?" and look at your current calibration values.
For a quick and easy "solution" to your problem: Modify your startup gcode to include "M860 S35". Examples have been given. That will make sure that you always start a print with a PINDA temperature of 35C. If your PINDA is not WARMER when you start a print like this - you DO NOT NEED to calibrate (so make sure your PINDA is colder than 35C before starting a print). Because if you make sure that the PINDA is always the exact same temperature during mesh bed leveling - you do not need to calibrate for the other temperatures.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Sometimes I get perfect first layers, and sometimes I don’t. Nothing is “just print and walk away”.
...
If someone could make instructions for a total newbie that would be awesome. Or a video. Don’t know. I read most of this, and I still am not clear. Too much info to understand.
That sounds like your printer needs that calibration.
First step is to install the latest firmware (3.2.1 as of today).
Then you need to ask more specific - because I do not even now where to start. Have you used Pronterface before? You need that or something similar like the terminal in Octoprint for communicating to the printer. There you could do a "M861 ?" and look at your current calibration values.
For a quick and easy "solution" to your problem: Modify your startup gcode to include "M860 S35". Examples have been given. That will make sure that you always start a print with a PINDA temperature of 35C. If your PINDA is not WARMER when you start a print like this - you DO NOT NEED to calibrate (so make sure your PINDA is colder than 35C before starting a print). Because if you make sure that the PINDA is always the exact same temperature during mesh bed leveling - you do not need to calibrate for the other temperatures.
Thanks for the summary. I haven't been following this thread because I have not had any problems with first layers. I have a black PINDA and it was generally set and forget with Live Z. But as of 3.2.x it has become less consistent. Now I need to understand this problem.
Does M830 wait for >= or only ==? I would imagine the latter causing it to hang indefinitely if you started hot and close to a hot bed.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Here's a thought - stop calibrating on start of every print. Simple calibrate once (per day/per week/per something or when you move the printer) when cold, and remove bed mesh leveling from the start of the gcode.
What do you guys think?
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Sometimes I get perfect first layers, and sometimes I don’t. Nothing is “just print and walk away”.
...
If someone could make instructions for a total newbie that would be awesome. Or a video. Don’t know. I read most of this, and I still am not clear. Too much info to understand.
That sounds like your printer needs that calibration.
First step is to install the latest firmware (3.2.1 as of today).
Then you need to ask more specific - because I do not even now where to start. Have you used Pronterface before? You need that or something similar like the terminal in Octoprint for communicating to the printer. There you could do a "M861 ?" and look at your current calibration values.
For a quick and easy "solution" to your problem: Modify your startup gcode to include "M860 S35". Examples have been given. That will make sure that you always start a print with a PINDA temperature of 35C. If your PINDA is not WARMER when you start a print like this - you DO NOT NEED to calibrate (so make sure your PINDA is colder than 35C before starting a print). Because if you make sure that the PINDA is always the exact same temperature during mesh bed leveling - you do not need to calibrate for the other temperatures.
Hey! Thanks for responding. So I was up until 5:00am today working on this. So I’m not new to printing just new to this advanced of printing!
I hooked up Pronterface last night, and kind of figured it out.
I followed your instructions as best I could. I installed your firmware, and did all the tests. Wrote my results down and did the math. Then I tried to set my values with the M667 (I can’t remeber is this was the command) Sxx xx commands to no avail. I could get them to result.
So I restored my current firmware and went to bed.
My issues are exactly as you state. I get it all dialed in, and then the next print it changes. Also my bed isn’t level, so I almost always have to do bed correction.
I use S3D not Slic3r and I am not pretty familiar with Pronterface. I have OctoPrint set up too.
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Sometimes I get perfect first layers, and sometimes I don’t. Nothing is “just print and walk away”.
...
If someone could make instructions for a total newbie that would be awesome. Or a video. Don’t know. I read most of this, and I still am not clear. Too much info to understand.
That sounds like your printer needs that calibration.
First step is to install the latest firmware (3.2.1 as of today).
Then you need to ask more specific - because I do not even now where to start. Have you used Pronterface before? You need that or something similar like the terminal in Octoprint for communicating to the printer. There you could do a "M861 ?" and look at your current calibration values.
For a quick and easy "solution" to your problem: Modify your startup gcode to include "M860 S35". Examples have been given. That will make sure that you always start a print with a PINDA temperature of 35C. If your PINDA is not WARMER when you start a print like this - you DO NOT NEED to calibrate (so make sure your PINDA is colder than 35C before starting a print). Because if you make sure that the PINDA is always the exact same temperature during mesh bed leveling - you do not need to calibrate for the other temperatures.
Thanks for the summary. I haven't been following this thread because I have not had any problems with first layers. I have a black PINDA and it was generally set and forget with Live Z. But as of 3.2.x it has become less consistent. Now I need to understand this problem.
Does M830 wait for >= or only ==? I would imagine the latter causing it to hang indefinitely if you started hot and close to a hot bed.
Hi JLTX!
If the old behavior just worked for you, than you could load the old values back into the EEPROM by "M861 !". Do not forget to enable temp cal afterwards in the settings menu.
You can verify that the EEPROM has been loaded by "M861 ?"
Re: 1st layer problems - in depth look at software PINDA problems (and solutions!)
Sometimes I get perfect first layers, and sometimes I don’t. Nothing is “just print and walk away”.
...
If someone could make instructions for a total newbie that would be awesome. Or a video. Don’t know. I read most of this, and I still am not clear. Too much info to understand.
That sounds like your printer needs that calibration.
First step is to install the latest firmware (3.2.1 as of today).
Then you need to ask more specific - because I do not even now where to start. Have you used Pronterface before? You need that or something similar like the terminal in Octoprint for communicating to the printer. There you could do a "M861 ?" and look at your current calibration values.
For a quick and easy "solution" to your problem: Modify your startup gcode to include "M860 S35". Examples have been given. That will make sure that you always start a print with a PINDA temperature of 35C. If your PINDA is not WARMER when you start a print like this - you DO NOT NEED to calibrate (so make sure your PINDA is colder than 35C before starting a print). Because if you make sure that the PINDA is always the exact same temperature during mesh bed leveling - you do not need to calibrate for the other temperatures.
Hey! Thanks for responding. So I was up until 5:00am today working on this. So I’m not new to printing just new to this advanced of printing!
I hooked up Pronterface last night, and kind of figured it out.
I followed your instructions as best I could. I installed your firmware, and did all the tests. Wrote my results down and did the math. Then I tried to set my values with the M667 (I can’t remeber is this was the command) Sxx xx commands to no avail. I could get them to result.
So I restored my current firmware and went to bed.
My issues are exactly as you state. I get it all dialed in, and then the next print it changes. Also my bed isn’t level, so I almost always have to do bed correction.
I use S3D not Slic3r and I am not pretty familiar with Pronterface. I have OctoPrint set up too.
Hi!
You set the values with the command "M861 Sxx Iy" - for example, to set the 40C offset to 5 you do "M861 S5 I0". The 35C "offset" cannot be set as it is the reference value. So 45C Offset to -8 would be "M861 S-8 I1"