Notifications
Clear all

[Solved] Z-level calibration varies with temperature  

Page 2 / 4
  RSS
DRKMSTR
(@drkmstr)
Trusted Member
RE: Z-level calibration varies with temperature

@avrcore

I've talked to Prusa, their official statement so far is "we will not ever support the pinda v2 probe on the mini printer" 

Right now I'm just using a lot of custom gcode because I don't want to  void my warranty by writing code to support the pinda v2 probe.

 

 

Posted : 02/05/2020 2:32 am
Texy
 Texy
(@texy)
Reputable Member
RE: Z-level calibration varies with temperature
Posted by: @avrcore

Same issue here. Every Mini will have this issue. Is there a way to automate the 5min preheat?

Yeah - change the Start G-code to include this extra line :

G90 ; use absolute coordinates
M83 ; extruder relative mode
G28 ; home all
G1 X100 Y100 F4000
G1 Z1.5 F50 ; park position
M104 S170 ; set extruder temp for bed leveling
M140 S[first_layer_bed_temperature] ; set bed temp
M109 R170 ; wait for bed leveling temp
M190 S[first_layer_bed_temperature] ; wait for bed temp
G4 S300; wait 300 secs (5 minutes) when heated up <--------------------
G28 ; home all without mesh bed level
G29 ; mesh bed leveling
Posted : 02/05/2020 7:57 am
Thunder1312
(@thunder1312)
New Member
RE: Z-level calibration varies with temperature

Hi!

I had the same issue with the z-level. the value starts (cold) at around -1.95 and when it getting hotter and hotter i have to adjust up to -0.8. I already had some scratches on my steelplate while i still was unaware of this issue. I have written some emails to the support, which seems not to understand the problem and chaseing the wrong ends. (Cable broken, bed not level, axis not level, ... And i provided pictures with prove that these options are not the case).

This is my first time prusa printer and when this is not fixed properly also my last time prusa printer. How could i recommend the printer if such an issue is not properly fixable with a price which is already above average compared to other printer. If the sensor is so expensive, why then they not made it optional like the filament sensor, but providing a solution which is not working properly and could damages the printer is really no way to go in my opinion.

Is the preheating solution working reliable enough, so that i really don't need to manually adjust it and can the printer let work "unattended" after start?

Posted : 17/05/2020 8:45 pm
avrcore
(@avrcore)
Active Member
RE: Z-level calibration varies with temperature

Indeed. They have to give us the option to at least buy a better sensor and support in in the firmware. It eats so much time and material to recalibrate with changing temperature. We need to get this message to prusa. 

Posted : 17/05/2020 8:53 pm
Texy
 Texy
(@texy)
Reputable Member
RE: Z-level calibration varies with temperature

The 5 minute pre heat helps but doesn’t  fix the issue fully for me. I find I need to restart many prints before getting the first later as good as I’d like it. 

Posted : 17/05/2020 9:04 pm
mark
 mark
(@mark-3)
Reputable Member
RE: Z-level calibration varies with temperature
Posted by: @thunder1312

Is the preheating solution working reliable enough, so that i really don't need to manually adjust it and can the printer let work "unattended" after start?

The 80 second version warmup is working reliably for me. I have been printing 16 hours a day, every day, and have not had to reset the Z axis adjustment in several weeks. It has varied over the 2 1/2 months I've been using the printer, from mechanical changes, but seems to have settled down lately. For example, taking apart the hot end required redoing the adjustment, but I have not had to do it since I also added G-code to park the head in a position for the load and unload to work better.

All the tricks I've found are listed here:

https://sites.google.com/view/newbieprusaminireview/home

I'm pretty satisfied now.

Regards,

Mark

Posted : 17/05/2020 9:12 pm
kobazik, With_Maltodextrin, Tracy and 2 people liked
DRKMSTR
(@drkmstr)
Trusted Member
RE: Z-level calibration varies with temperature

@thunder1312

@avrcore

@texy

The only preheating that helps is by:

1. Select Preheat
2. Select "Auto-home" - calibration
3. Select "Disable Stepper Motors" - settings/configuration
4. Move the print head, centering the PINDA V1 (MINDA) probe over the center of the build plate.
5. Wait 5 minutes
6. Print, but watch first layer to make sure it hasn't shifted too much

I'm beginning to log first layer shifts, with this preheat it only shifts in the 10s of microns each print. 

WARNING: This only works when re-printing the same material again and again. Once your temperature changes for a different material, you have to redo z level again and again until you figure out the temperature shift. 

 

Posted : 19/05/2020 1:41 am
DRKMSTR
(@drkmstr)
Trusted Member
RE: Z-level calibration varies with temperature

@marklgoldberg

We shouldn't have to make custom GCODE to get the printers to work reliably. It's a friggin "beginners" printer. It is just as complicated as my Ender 3 but costs 2x as much.

Posted : 19/05/2020 1:42 am
Texy
 Texy
(@texy)
Reputable Member
RE: Z-level calibration varies with temperature

The custom gcode does those steps 1-5 for you. 

Posted : 19/05/2020 5:57 am
DaveH
(@daveh)
Active Member
RE: Z-level calibration varies with temperature

I just got my Mini a couple of days ago, and I’m having this same problem.

I’m not impressed with having to mess about with first layer heights before every print. I’ve tried pre-heating and that doesn’t really help. The only thing that helps is re-printing my 1 layer test square over and over until the printer ‘gets up to temperature’, and then I can run the print.

I spoke to support and they’re going to send me a new MINDA but I’m not confident that will help but we’ll see. This looks like a poor decision from Prusa by not using a PINDA with the Mini. 

Posted : 17/08/2020 12:59 pm
Tracy
(@tracy)
Trusted Member
RE: Z-level calibration varies with temperature

@dave1234

Here's what I do: Keep paper and pen near the printer, learn the hot and cold Z heights and write them down. You'll generally have to change the Z only twice a day. It's annoying but at least I get good prints all the time now.

Also, the cold Z seems to work on the first print of the day only. If I let the printer sit for a couple hours I usually use the average of the hot and cold number.

Posted : 17/08/2020 1:23 pm
bobstro
(@bobstro)
Illustrious Member
RE: Z-level calibration varies with temperature

FWIW - PINDA 1 had similar issues. I found that the process of waiting for the PINDA to warm up had other benefits with warming the print bed to a consistent temperature as well as the entire printer. The loss of a few minutes at the start of a print is annoying, but bothers me less as I move into longer prints.

I did find that using 80C for the bed temp while warming sped things up and was far more effective on cold days. I switch the bed temp back to the first layer temp immediately after warmup and find it's usually close when the prime line prints.

Not sure how well this approach might translate to the Mini but wanted to share experience and observations from having similar issues in the past.

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 : 17/08/2020 1:31 pm
mark
 mark
(@mark-3)
Reputable Member
RE: Z-level calibration varies with temperature
Posted by: @dave1234

I’ve tried pre-heating and that doesn’t really help. The only thing that helps is re-printing my 1 layer test square over and over until the printer ‘gets up to temperature’, and then I can run the print.

I spoke to support and they’re going to send me a new MINDA but I’m not confident that will help but we’ll see. This looks like a poor decision from Prusa by not using a PINDA with the Mini. 

Did you try the custom g code? That has worked for me. I just start printing with no issues. My printer is in an enclosure and we are running HVAC almost all the time as we are in Arizona.

Regards,

Mark

Posted : 17/08/2020 1:36 pm
DaveH
(@daveh)
Active Member
RE: Z-level calibration varies with temperature
Posted by: @tracy

@dave1234

Here's what I do: Keep paper and pen near the printer, learn the hot and cold Z heights and write them down. You'll generally have to change the Z only twice a day. It's annoying but at least I get good prints all the time now.

Also, the cold Z seems to work on the first print of the day only. If I let the printer sit for a couple hours I usually use the average of the hot and cold number.

 

Thanks for that, that's very helpful.

Posted : 17/08/2020 5:42 pm
DaveH
(@daveh)
Active Member
RE: Z-level calibration varies with temperature
Posted by: @marklgoldberg
Posted by: @dave1234

I’ve tried pre-heating and that doesn’t really help. The only thing that helps is re-printing my 1 layer test square over and over until the printer ‘gets up to temperature’, and then I can run the print.

I spoke to support and they’re going to send me a new MINDA but I’m not confident that will help but we’ll see. This looks like a poor decision from Prusa by not using a PINDA with the Mini. 

Did you try the custom g code? That has worked for me. I just start printing with no issues. My printer is in an enclosure and we are running HVAC almost all the time as we are in Arizona.

Regards,

Mark

 

Thanks, I haven't yet, but I will try this next. 

Maybe my expectations were too high, I have been running my mk3 now for a a couple of years and it's running pretty much all day long and some days into the night - most days, and I don't think i've touched the z height in well over a year! It just plods away. I was expecting the Mini to be the same but a bit smaller! Maybe I should have bought another mk3.

Posted : 17/08/2020 5:49 pm
vital.g
(@vital-g)
New Member
RE: Z-level calibration varies with temperature

For me, the custom gcode with positioning the head always in the same position and heating everything up for five minutes saved the MINI. With this trick, it went from not usable to a workhorse 🙂

Posted : 17/08/2020 7:13 pm
loglow
(@loglow)
Active Member
RE: Z-level calibration varies with temperature

The fix for this problem, whatever it is, must be incorporated into a firmware revision.

Custom G-code is a temporary and brittle solution. This is an entry-level printer, and even if it requires a 5 minute process to achieve reliable results, that reliability is the difference between a useable piece of equipment and a toy. New users should not be expected to discover this unreliability on their own, search the forum for a fix, and then implement it manually.

Also, has anyone actually tried replacing MINDA with PINDA yet?

Posted : 18/08/2020 1:27 am
Malcolm
(@malcolm)
Trusted Member
RE: Z-level calibration varies with temperature

@vital-g

I've had the same experience (2 min warmup) but would also like to add-- you need to let it cool down between prints for it to be reliable. Someone said it before and I've followed the same advice: let it cool until the hotend fan stops, that seems to be 'enough'

Posted : 18/08/2020 1:31 am
Malcolm
(@malcolm)
Trusted Member
RE: Z-level calibration varies with temperature

@loglow

Someone has done this successfully, it requires you physically breaking a buddy board connector that allows you to flash unsigned (modified) firmware, which also voids warranty. I don't have any issues with voiding warranty but I would rather not have to go through all the trouble, it is a known issue with the devs and there is a thermistor slot on the buddy board so I think a better fix will come eventually. 

Here is the someone I'm referring to: https://forum.prusa3d.com/forum/user-mods-octoprint-enclosures-nozzles/printing-abs-on-the-mini-with-pinda-and-110-bed/

Posted : 18/08/2020 1:38 am
loglow
(@loglow)
Active Member
RE: Z-level calibration varies with temperature

@malcmcm

That's a good writeup, thanks for the link.

 

Prusa, you guys really need to do the following:

  • Implement official Mini firmware fix for reliable operation with MINDA.
  • Implement official Mini firmware support for PINDA compatibility.

 

Neither of these things should be difficult. Voiding the warranty to fix this is not an acceptable solution.

It'd also be a good look to offer existing Mini customers free or reduced-cost PINDA components. If manufacturers don't stand behind their designs—and their mistakes—why shouldn't their customers turn to generic alternatives?

This post was modified 4 years ago 2 times by loglow
Posted : 18/08/2020 1:57 am
kobazik liked
Page 2 / 4
Share: