RE: Lots of initial trouble
Well, the first print of one of my models came out perfect, easily better quality than my other cheapo printers. Looking forward to many more. 🙂
RE: Lots of initial trouble
I had one more thought about all this... when I was having that initial "some sort of issue" issue... probably due to a little something under the print plate as a poster noted above... I did a full wizard recalibration which would have erased any live Z settings they may have made before they shipped it. So that would explain my all my printing and adhesion problems. All that is sorted now that you guys have got me on the straight and narrow. And I learned a lot about the printer and its settings.
RE: Lots of initial trouble
Only complaint is that this was supposed to come fully calibrated and had a supposed print on the bed to show that.
True, but don't forget that during transport things can get shaken up--and depending on the carrier, badly--and we're talking distances of less than a millimeter here. For the same reason, I also routinely tighten all screws and check the belt tension of new printers, just to be sure.
Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...
RE: Lots of initial trouble
I would go as far as to say that even a factory-new printer is very seldom truly plug-and-play.
A 3d printer is not a toaster oven. It requires some calibration and tweaking under the best of circumstances.
RE: Lots of initial trouble
Ding! Ding! Ding! We got a winner. It took me a good 3 months to get my printing like I wanted.
I would go as far as to say that even a factory-new printer is very seldom truly plug-and-play.
A 3d printer is not a toaster oven. It requires some calibration and tweaking under the best of circumstances.
--------------------
Chuck H
3D Printer Review Blog
RE:
In this case I now believe they probably had it calibrated correctly, and I erased their Live Z setting when I did the full system calibration. So, really just operator error.
And I have been tweaking it back using a little different method for now. I have printed the PRUSA badge at different Live Z settings and compared the results to the original one that came on my printbed. In all cases the parts printed and adhered fine, the only difference being the first layer smoothness where the part contacts the printbed. The original was totally smooth whereas those printed with insufficient negative Live Z offset show progressively more roughness due to the individual runs not being squished together quite enough. In the photo below the one on the left is the original from Prusa. The ones on the right were printed with LiveZ offsets of -0.7, -0.9, and -1.1 from bottom to top. The -1.1 is very close to the Prusa original. Looks like maybe another -0.1 will get it there.
Ding! Ding! Ding! We got a winner. It took me a good 3 months to get my printing like I wanted.
I would go as far as to say that even a factory-new printer is very seldom truly plug-and-play.
A 3d printer is not a toaster oven. It requires some calibration and tweaking under the best of circumstances.
RE: Lots of initial trouble
Getting Z adjustment right is probably the most difficult but most important part of a printer setup to get good looking and reliable prints. It's not that it's hard to do, it's just learning the proper way to do it which only happens with experience of doing it. Once you get it right and achieve good results it's easy to do it again. It's just a technique you have to learn.
A technique I use is to run the z calibration and keep decreasing the number until you get nice square corners and lightly brushing your finger across the printed areas to make sure the filament sticks with very light pressure. The top will still look rounded. Write the z number down. You might have to do this several times, so be a bit patient here and make sure the top of the filament is still rounded but the corners are sharp.
The next step is to run the z calibration again and quickly adjust the live z until the filament is flattened to the point it is starting to look translucent. Write this number down. If you have to do this several times... that's fine.
Once you have the high and low values, split the difference of the z numbers and you should be really close to a very usable z number. At the very least you will be in a comfortable range to start producing good reliable prints and can dial in further in very small increments if you want to get z cal perfection.
I have used this for not only Prusa printers but other brands as well. It has always allowed me to get good z cal numbers very quickly. Once you have done this a time or 2 you will see the results of a good z cal compared to a bad z cal and things become very easy to repeat because you will know what a decent z cal looks like. This is just a way to gain experience with proper z height for printing. If changing nozzles or anything that can change your z cal you will need to do this again, but after doing this once it becomes really easy to repeat a good z cal.
Before starting make sure to have the proper sheet selected since each sheet will save this z number saved to that sheet profile and you will need to do this for each sheet you intend to use since the numbers will be different for each one. Good Luck.
Experience definition: Experience is what you get when you don't get what you thought you were going to get
RE: Lots of initial trouble
Getting Z adjustment right is probably the most difficult but most important part of a printer setup to get good looking and reliable prints.
It's a lot easier on the Prusa than on some printers which have only a mechanical bed adjustment and leveling mechanism!
RE: Lots of initial trouble
I remember buying my first Prusa. Couldn't believe I didn't have to futz around anymore with those stupid corner screws.
But yes, getting a good Live Z is arguably the number 1 answer to 90% of all questions asked here, right next to clean the sheet with hot water and soap, and RTFM 😀
Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...
RE: Lots of initial trouble
In this case I now believe they probably had it calibrated correctly, and I erased their Live Z setting when I did the full system calibration. So, really just operator error.
Ding! Ding! Ding! We got a winner. It took me a good 3 months to get my printing like I wanted.
It is really common. I went through this same thing almost four years ago.
--------------------
Chuck H
3D Printer Review Blog
RE:
The thing is it never would have occurred to me that a machine on default settings wouldn't or couldn't at least print successfully, if not optimally. Not a design choice I would personally make, but perhaps there is a good reason Prusa did this? Maybe I would understand if I did what fuschr recommends and RTFM! Nah....
In this case I now believe they probably had it calibrated correctly, and I erased their Live Z setting when I did the full system calibration. So, really just operator error.
It is really common. I went through this same thing almost four years ago.
RE: Lots of initial trouble
The reason is, that first layer calibration depends on the mounting height of your PINDA and the type of sheet you use. It needs to be set according to this two variables, there is no generally correct z-height offset…
If at first you don't succeed, skydiving is not for you.
Find out why this is pinned in the general section!
RE: Lots of initial trouble
I have used atleast 30 printers that eight different manufacturers assembled. None of them arrived without needing some leveling and tinkering. None of them printed well out of the box. All of my Prusas needed some tinkering with the PINDAs and calibration. They must shift some in transit.
The thing is it never would have occurred to me that a machine on default settings wouldn't or couldn't at least print successfully, if not optimally. Not a design choice I would personally make, but perhaps there is a good reason Prusa did this? Maybe I would understand if I did what fuschr recommends and RTFM! Nah....
In this case I now believe they probably had it calibrated correctly, and I erased their Live Z setting when I did the full system calibration. So, really just operator error.
It is really common. I went through this same thing almost four years ago.
--------------------
Chuck H
3D Printer Review Blog
RE: Lots of initial trouble
That makes sense, especially considering that they offer a textured plate that has to be somewhat thicker. Any setup of the probe and such during building would need to leave room for that... hence the need to go negative on the z adjust for the others.
The reason is, that first layer calibration depends on the mounting height of your PINDA and the type of sheet you use. It needs to be set according to this two variables, there is no generally correct z-height offset…
RE: Lots of initial trouble
All of my Prusas needed some tinkering with the PINDAs and calibration.
One thing that you rarely hear about from people who have assembled and used the Prusa MK3 series, at least those who wired things correctly and RTFM'd, are incidents of the print head crashing into the build plate, gouging the surface of the build plate, or even damaging the heat bed. (Let alone the damage to the brass nozzle.)
The initial setup (with the sheet of paper and the zip tie as a feeler gauge) sets up things on the high side with enough margin so cumulative tolerances and such will still not allow a collision with the build plate.
As I think back to the reported issues here, I can remember only a single case, a video of a machine with either a Pinda failure or some electromechanical issue that caused the extruder to ram downward onto the build plate, visibly deforming it on impact. Things like that appear to be very rare on the Prusa.
RE: Lots of initial trouble
That makes sense, especially considering that they offer a textured plate that has to be somewhat thicker.
Actually, the textured plate is thinner than the stock smooth build plate. It usually needs a significantly lower value on Z calibration.
I do remember reports of people scratching the surface of the smooth PEI plate by forgetting to change the profile from textured to smooth and then launching a print.
RE: Lots of initial trouble
Yeah, when I did the full system cal wizard there was a step where it prompted me to place a sheet of paper on the bed and to stop immediately if the nozzle pressed up on that enough to grab the paper. I had no such problems.
Interesting how the satin and textured plates are thinner than the smooth. Counterintuitive, to me anyway. Goes to show that you can assume nothing. RTFM! OK, I'll try. 😉
All of my Prusas needed some tinkering with the PINDAs and calibration.
One thing that you rarely hear about from people who have assembled and used the Prusa MK3 series, at least those who wired things correctly and RTFM'd, are incidents of the print head crashing into the build plate, gouging the surface of the build plate, or even damaging the heat bed. (Let alone the damage to the brass nozzle.)
The initial setup (with the sheet of paper and the zip tie as a feeler gauge) sets up things on the high side with enough margin so cumulative tolerances and such will still not allow a collision with the build plate.
As I think back to the reported issues here, I can remember only a single case, a video of a machine with either a Pinda failure or some electromechanical issue that caused the extruder to ram downward onto the build plate, visibly deforming it on impact. Things like that appear to be very rare on the Prusa.
RE: Lots of initial trouble
The Pinda doesn't know where the nozzle is, it just knows its own distance from the steel sheet. Here's a good description of how it works and what live z adjustment does, and it should be quite obvious from that that even minute changes during transport can affect it: https://www.reddit.com/r/prusa3d/comments/9kab1l/can_someone_explain_the_concept_of_the_pinda_to_me/
i should also note that one can tackle this issue differently. The Prusa XL's Nextruder has a load cell, so the nozzle itself is used to detect contact with the steel sheet. No more Pinda or live z adjustment required. At least on pape you should always have a perfect first layer.
Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...
RE: Lots of initial trouble
The Prusa XL's Nextruder has a load cell, so the nozzle itself is used to detect contact with the steel sheet. No more Pinda or live z adjustment required. At least on pape you should always have a perfect first layer.
I think that's how the Ultimaker does it. There's no obvious Pinda-equivalent on the movable head assembly.
It does a very similar multi-position 'sounding' of the build plate before each print, but it appears to concentrate on the areas of the build plate that will be occupied by the print and skips most of the rest.
Occasionally, something on the build plate, such as an old blob of filament in one of the probe positions, will cause it to error out and display a message to check the plate and nozzle for debris.
The first layer has been spot on all of the time.
The Ultimaker does have some manual leveling adjustments, but TFM has admonitions not to futz with these unless it's absolutely necessary.
RE:
Which also begs the question: how stable is this setting and detection? Is there any electronic drift in the system so that the Live Z needs to be recalibrated periodically? Or maybe it would just be a good practice to do so. I know, this is probably addressed in TFM.
The Pinda doesn't know where the nozzle is, it just knows its own distance from the steel sheet. Here's a good description of how it works and what live z adjustment does, and it should be quite obvious from that that even minute changes during transport can affect it: https://www.reddit.com/r/prusa3d/comments/9kab1l/can_someone_explain_the_concept_of_the_pinda_to_me/
i should also note that one can tackle this issue differently. The Prusa XL's Nextruder has a load cell, so the nozzle itself is used to detect contact with the steel sheet. No more Pinda or live z adjustment required. At least on pape you should always have a perfect first layer.