Brand new prusa mini: Z needs to adjust on every new print.
 

Brand new prusa mini: Z needs to adjust on every new print.  

Page 5 / 6
  RSS
Karl Herbert
(@karl-herbert)
Famed Member

I am working on the Mini with Octoprint and a Meshbed Leveling Plugin. This shows with a difference of up to 0.3mm between cold and warm printing bed when leveling. For this reason I always let the printer warm up for about 10 minutes (bed and nozzle) and then do a bed leveling. I have to mention that my printing bed is mounted on springs and therefore I can easy level the bed (0.05 to 0.1mm).
I don't have any adhesion problems, on the contrary - the printed parts can often hardly be removed from the printing bed. Nn the textured sheet it is easier to remove the parts after cooling down.

 

Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen....
Posted : 09/09/2020 1:52 pm
sth.udoma liked
rdanielo
(@rdanielo)
Active Member

Holy cow, I'm reading this thread and I'm freaking out. I wish I read it before because I would have never bought a prusa mini.

Now I have it at home and I was very happy with the first prints: I just adjusted the first layer, and tried some of the examples on the USB using the provided filament samples.

Problems has started to show when I wanted to print different things and using different filaments. At first, I tried to re-adjust the first layer just to freak out when the maximum of `-2`mm was not enough. A loose MINDA? I thought, so I raised it up and tried again. I got a decent first layer calibration so I moved on.

Since then, I had to calibrate the first layer on every new print, and I was cursing the filament I was using, now I see what the problem is.

I have another printer (a witbox) for 5 years, and I was sick of it being a niche printer (small community) and to have to level the bed every few prints for a not so great print quality. So I decided to "step up" and buy a prusa. "Just assemble it and print without issues from first day", I read about Prusa printers, and that is what I was expecting from my prusa mini. 

I'm very very disappointed and angry, because I don't want another printer to waste my time on tuning, fixing and testing just to get average printing results.

I'll check with support if I can send it back, because if I have to fine-tune it I already have a bigger printer at home.

Posted : 11/09/2020 6:30 pm
rdanielo
(@rdanielo)
Active Member

Also, what is the point on having a probe if you have to calibrate on every print? I was thinking to buy a probe for my old printer because I was sick of calibrating it. If I had a probe I want to calibrate it once (to check the distance between hotend and probe) and don't have to touch it ever again.

To preheat it for 5 minutes? To calibrate on every print? To monitor the first layer and adjust live even after performing all the previous steps ? What joke is this?

Posted : 11/09/2020 6:40 pm
bobstro
(@bobstro)
Illustrious Member
Posted by: @rdanielo

[...] To preheat it for 5 minutes? To calibrate on every print? To monitor the first layer and adjust live even after performing all the previous steps ? What joke is this?

Are you having to re-calibrate Live-Z even with the 5 minute warmup pause? 

The original PINDA on the Mk3 had similar issues. It has the ability to monitor the PINDA temp though, so you can have it continue on once the probe is at a consistent 35C. I print across a range of temperatures in my home, so actually like the warmup to get the bed evenly heated and warmup the ambient temp a bit.

If you're not happy, definitely get in touch with support and let them know. A lot of users aren't having any issues. Day-to-day ambient temp variations might account for some issues, but it sounds like some users are experiencing more persistent issues.

If you're not happy with it, get the refund. Otherwise, Prusa is pretty good at dealing with issues over time. If you can live with a work-around for a bit, the wait should be worth it.

 

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 : 11/09/2020 7:08 pm
jhenley01
(@jhenley01)
Active Member

Pre-heating for a few minutes when cold has definitely fixed the issue for me. I also checked the mesh with a bed visualizer and squared up the z axis, not sure if that made any difference or not with first layer issue or not.

Hard to believe it's been this long without an official fix but then again it's been almost a year and we still don't have any of the promised updates like wi-fi, power loss recovery, farm software or stealth chop.

 

This post was modified 1 month ago by jhenley01
Posted : 11/09/2020 7:44 pm
sth.udoma
(@sth-udoma)
Eminent Member

Just received a written info from PRUSA support that "Regarding the MINDA - our developers are working on a proper solution". We should wait a little more giving them a chance to fix it properly.

Posted : 11/09/2020 10:16 pm
jeremy.naus
(@jeremy-naus)
Active Member

For me the fix was to add in the startup gcode the warm up of the MINDA for 3 minutes.

 

But I understand the frustration. What is the use of a sensor if you continuously have to manually verify the distance with each print.
Better have a mechanical sensor like the BL Touch.

Posted : 11/09/2020 10:59 pm
crawlerin
(@crawlerin)
Honorable Member

I am printing MINI plastic parts from PETG now and it looks like I get more consistent results when I let the printer and enclosure cool down rather than pre-warming. No issues with first layer bouncing up and down and second layer attaches to first properly and does not crumble.

Posted : 12/09/2020 3:54 pm
herbert-1
(@herbert-1)
Active Member

@sth-udoma

That is very good news in my take.  Most importantly it restores my faith in Prusa being an ultimately customer friendly company that is constantly trying hard to drive the 3D printing experience forward. I have to say I was rather disappointed when it seemed for months that Prusa was intentionally ignoring the issue. I believe it is a fair request now from their side to ask for the time it takes to solve this problem properly.

Frankly, I had worked around the problem by giving the printer significant time after a completed print to cool down, and in addition used the warm-up procedure before each print to even take the room temperature influence out of the equation. A possible, but annoying procedure, considering that the Ender3Pro with added BL touch, that stands just beside my Prusa Mini, is faster now to start printing despite is slower warm-up capabilities. Not to speak about the fact that I can start a new print on the Ender3Pro only seconds after I have ripped off the printed part from the still hot rubber bed. I seriously hope that the Prusa Mini will ultimately win these races in the future 😉

 

Posted : 12/09/2020 4:45 pm
rdanielo
(@rdanielo)
Active Member

So, now that I have added the warmup code to my slicer profile I noticed that starting a print from a "cold prusa" produces a perfect first layer and a perfect print. Not sure if the warmup code made a difference (it is just one line, right?) but starting from cold definitively did.

Not sure how, but I misread the suggestion on this thread about leaving a 5 minute cool-down between prints, not sure if 5 min is enough though.

Posted : 14/09/2020 6:01 pm
Radmaster
(@radmaster)
New Member

Thanks everyone for the information on this post this is what I have been looking for I have been having real inconsistent first layers after adding this to the start gcode it has definitely helped out

Posted : 20/09/2020 3:03 pm
thomas.b13
(@thomas-b13)
Eminent Member

Anyone had the issue crop up again after working fine when doing a preheat? My z has been fine for months, zero change however just today it’s been all over the shop and no amount of preheating is fixing it. 
A constant z offset was about the only reliable aspect of this printer for me, and now that’s gone.

...
Posted : 21/09/2020 12:16 am
rdanielo
(@rdanielo)
Active Member
Posted by: @thomas-b13

Anyone had the issue crop up again after working fine when doing a preheat? My z has been fine for months, zero change however just today it’s been all over the shop and no amount of preheating is fixing it. 
A constant z offset was about the only reliable aspect of this printer for me, and now that’s gone.

If you mean, if after some days working fine using the preheat trick the issue appears again, yes, it is happening to me. Also, as a nice nasty bonus the prusa mini is killing USB memory sticks.

Posted : 21/09/2020 8:22 am
jeremy.naus
(@jeremy-naus)
Active Member

I've been printing ABS for 2 weeks now, with sometimes a break of a day. And my 1st layer is still spot on.

I do let the minda heat up for 3 minutes when the bed is at temperature.

Posted : 21/09/2020 8:24 am
rdanielo
(@rdanielo)
Active Member
Posted by: @jeremy-naus

I've been printing ABS for 2 weeks now, with sometimes a break of a day. And my 1st layer is still spot on.

I do let the minda heat up for 3 minutes when the bed is at temperature.

At which height? Because sometimes the preheat happens with the head way up, sometimes happens close to the bed.

Posted : 21/09/2020 8:25 am
marcel.mayer
(@marcel-mayer)
Active Member

Guess this will be the reason for that:

https://forum.prusaprinters.org/forum/hardware-firmware-and-software-help/analysed-problems-with-minda-probe/

Posted : 22/09/2020 8:51 am
ktdon83
(@ktdon83)
New Member

@rick-m12

Rick, Thanks so much for such detailed information. However, I am very new to all of this and this is definitely not my area of knowledge/experience. I have downloaded, saved and manipulated a print using Prusa Slicer and they were downloaded as one type of file that I had to save as G code, but I don't really understand what you mean by "copying the text and paste into a new file". If you have the time, could you "walk me through" that process? Do I copy the entire text starting at:     ;Preheat temperature extruder... all the way to M84 ; Disable motors?   and how/what do I save it into? 

Posted : 25/09/2020 4:21 pm
skreelink
(@skreelink)
Active Member

Still waiting on my Mini and have read a lot about the potential problems and all with it. I've no experience with 3D printing yet, but I have an idea if anyone wants to try implementing it. The main issue seems to be temp fluctuations in the MINDA sensor, so the "fix" would be to normalize it and the preheat method seems to be the go-to. If you do a Z-cal after preheating the MINDA, then subsequent preheats should in theory make it about the same... but this wastes time, so my idea is instead to keep the MINDA from warming.

 

I have a few ideas on how to do this since the main culprit is likely the hotend, since its location never changes compared to the MINDA and is the hottest part vs the bed only being close at the first few layers. Part of the solution could be a simple silicone sock, this insulates the hotend and should keep temps level, but also keep the heat from directly going to the MINDA probe. I've read the e3d socks fit, but maybe a bit loose because of the chamfer, but at the same time, I've read they fit fine. Again, I don't have the Mini yet to test.

Second part of the fix would be to take the MINDA probe holder files and modify it a little. Bring the plate on the fan side down, creating a heat shield between the hotend and probe. Maybe create a 45 degree angle toward the probe side so it has less direct hotend facing edges. This should block the heat from the hotend from getting to the probe, so its calibration doesn't change from cold to warm, it should stay relatively room-temp (except the very beginning of the print when it's close to the bed, but sufficiently cool down in longer prints as it moves away from the bed). Possibly attaching a small bit of aluminum foil on the hotend facing side to deflect heat could be implemented as well, but that'd reflect it directly into your print likely.

Another idea would be building a part from scratch (for you people that are decent with CAD) that maybe mounts at the wire bolt at the top (or maybe clips onto the brass bowden connector). Idea here is to make a duct that catches maybe one vent worth of air passing through the heatsink on the front and directs it down to the end of the probe to keep it cool. While this may be slightly warmer air (though at the edge it should be the coolest vs center air), it would keep it reasonably even. At the same time, maybe the moving air would be much cooler by the time it reaches the probe and just keep it from heating up at all.

 

Either of these methods should instead keep the MINDA at the "cool" side, thus eliminating the preheat time and making prints faster. If anyone is interested, I can see about making the heatshield version of the MINDA holder and they can print it.

 

EDIT: These methods also remove the requirement to modify/add G code.

This post was modified 3 weeks ago by skreelink
I think I know, I don’t think I know. I don’t think I think I know, I don’t think I think....
Posted : 03/10/2020 7:57 pm
marcel.mayer
(@marcel-mayer)
Active Member

@all: As I wrote in my Thread linked here... If you have to do such „warm up hacks“, your Minda Probe is not ok.

With the new one I got, I made the calib once and that’s it. I can switch between filaments without doin it again or any warm up gcodes. Just print and being happy ...

Posted : 03/10/2020 8:17 pm
skreelink
(@skreelink)
Active Member

I went ahead and mocked up the heatshield on the MINDA holder. The height of it is not correct, naturally, since I don't have my mini yet to measure. However, it's a visualization if anyone else wants to actually measure the distance and print it.

I've uploaded the .stl file here: https://www.prusaprinters.org/prints/42167-prusa-mini-minda-heatshield to showcase the idea.

I think I know, I don’t think I know. I don’t think I think I know, I don’t think I think....
Posted : 03/10/2020 8:28 pm
Page 5 / 6
Share:

Please Login or Register