Notifications
Clear all

Perplexed ??  

  RSS
Sink
 Sink
(@sink)
Estimable Member
Perplexed ??

Have been working with support, but it seems they are as perplexed as I am right now.. New machine, prints are more than reasonably good except the benchy.  Always prints with this line across the cabin door. Belt adjustments, vibration testing, all have not affected this line when testing with support over days.

Support doesn't think this is a shift, though it looks a bit like one, other side is shifted as well, but it can't be reproduced on other models of similar or greater height. . They are analyzing gcode to see if something there, it is the code that came on SD, but have run it on other machines and not getting results with line this bad.

The bishop next to the benchy was printed on same machine right after the benchy for comparison and test whether this was a Z thing, but it is not; verified by later prints of other models as well. 

Anyone ever see something like this  before?

Posted : 01/06/2021 5:33 am
cwbullet
(@cwbullet)
Member
RE: Perplexed ??
Posted by: @sink

Have been working with support, but it seems they are as perplexed as I am right now.. New machine, prints are more than reasonably good except the benchy.  Always prints with this line across the cabin door. Belt adjustments, vibration testing, all have not affected this line when testing with support over days.

Support doesn't think this is a shift, though it looks a bit like one, other side is shifted as well, but it can't be reproduced on other models of similar or greater height. . They are analyzing gcode to see if something there, it is the code that came on SD, but have run it on other machines and not getting results with line this bad.

The bishop next to the benchy was printed on same machine right after the benchy for comparison and test whether this was a Z thing, but it is not; verified by later prints of other models as well. 

Anyone ever see something like this  before?

Two questions:

  1. Any changes from the other 3 sides?
  2. Did you slice this yourself or is this from the SD card from Prusa? 

 

--------------------
Chuck H
3D Printer Review Blog

Posted : 01/06/2021 9:09 am
Sink
 Sink
(@sink)
Estimable Member
Topic starter answered:
RE: Perplexed ??

@cwbullet

The benchy is from the code on the SD that came with the printer. The bishop is my own and sliced by prusa slicer, all stock settings with .15 quality layer and .15 infill.

The bishop is near identical on all sides. The benchy from the front has the line and what appears to be a slight layer shift left, the other side of the benchy has a slightly less defined line, but it is there.

Interestingly, I just received an email from support and they ran the gcode and while they did not get the line, they did get some serious ghosting and bad stringing. They suggested that the bishop was of high enough quality as to drop the hardware problem search altogether.

They did suggest some changes of speed might be contributing, though I am not quite sure if they meant some unexpected  changes on my side or something within the gcode itself..

I might run one more test using a reduced print speed, but the line remains a mystery, maybe I will pull the extruder and see if a reset has any affect..

Currently running some tests across all MK3S+ machines with seam, infill, perimeter, and min shell changes to see affect on ringing and if there is an optimal set of settings. Will try the lower speed test but if to no avail, will pull the extruder and try a reset.

Posted : 01/06/2021 10:59 am
cwbullet
(@cwbullet)
Member
RE: Perplexed ??
Posted by: @sink
 will pull the extruder and try a reset.

Sounds like a good idea.  

--------------------
Chuck H
3D Printer Review Blog

Posted : 01/06/2021 12:05 pm
bobstro
(@bobstro)
Illustrious Member
RE: Perplexed ??
Posted by: @sink

Have been working with support, but it seems they are as perplexed as I am right now.. New machine, prints are more than reasonably good except the benchy.  Always prints with this line across the cabin door.

It's worth noting that the line is occurring at a common failure line for Benchy prints. Having the print detach from the build plate as the arch is printed is very common. A bridge span is being printed at that point, and if there is any sort of movement, the nozzle can easily snag the print and pop it loose. (Been there, have the t-shirt myself.)

What I suspect is wrong is that your Benchy is simply not adhering and you're getting a slight wobble accounting for distortion at that specific spot. Otherwise your prints look very good.

Examine the bottom of the Benchy and post a picture here. Look for any slight curling or lifting. Two factors are critical to success with the Prusa printer:

  1. Bed cleanliness. The PEI must be properly clean to give good, even adhesion.
  2. Live-Z adjustment. You need just the right amount of "squish" of the filament to get a good grip.

These are the 1st issues anybody new to a Prusa are likely to encounter. I'm going to provide a big info dump on these topics below that you can look through. I suggest trying both techniques, then printing your benchy again.

Info dump:

Your underlying issue is bed adhesion. Bed adhesion should be strong enough to hold the print flat and securely against  the forces of filament warping and nozzle friction. Filament should lay down cleanly on the PEI print surface and hold throughout the print.  If adhesion is poor, your print is likely to fail or have problems at higher layers. Common adhesion-related problems include:
  • Part warping or lifting along edges.
  • Bumping or knocking noises as the part moves under the nozzle.
  • Uneven vertical surfaces and bulges caused by print lifting and compression of overlying layers.
  • Spaghetti resulting from parts moving during printing.
  • The dreaded "blob of doom" caused by the part lifting and sticking to the nozzle as it extrudes filament.
 
There are 2 main causes of bed adhesion problems: cleanliness of the PEI print surface and Live-Z height adjustment. Even if you're dubious that these are the cause of your problems, they need to be addressed as 2 fundamental troubleshooting steps to rule out the basics:
 
  • A dirty PEI print surface. Even if you don't think this is the cause, it's always a good idea to make sure your PEI surface is clean before trying any other fixes. If it's a smooth PEI sheet, take it to the kitchen sink and give it a good dunk with Dawn (original formula, no vinegar or hand softener variants) dishwashing soap or your local equivalent (e.g. Fairy in UK). Use a clean paper towel to clean it off, and another to dry it. Avoid using any sponge or cloth that has been in contact with grease. Above all else, avoid touching the PEI print surface. Once it's good and clean, you should be able to use 91%+ isopropyl alcohol between prints, 100% acetone when that fails , and another dunk when acetone fails. Worst-case, use a 3M 7445 ScotchBrite (1200-1500 grit) pad or equivalent on smooth PEI to give it a very light buff, but only infrequently. If you've got a textured powder-coated PEI print surface, the official instructions are to use 91%+ isopropyl alcohol on it only. Rumors persist that some have had good luck getting started with these sheets by giving them a wipe with 100% acetone and a dunk with Dawn. Either way, there's no real warranty on these sheets. YMMV.
  • If you have not already done so, try using Jeff Jordan's "Life Adjust" procedure for calibrating your Live-Z setting. It is much easier to use and understand than the onboard routine and much easier to make mid-print adjustments accurately with. In general, start high (less negative) and work lower (more negative) in large increments (e.g. 0.1mm) until the filament starts to stick on its own. When you've got your Live-Z setting adjusted properly, you should be able to gently rub the extruded lines on the PEI surface without dislodging them.  Then start lowering (more negative) the level until there are no gaps between layers. You should not be able to peel the lines apart after printing, but the top should be regular. 
 
Remember that the effectiveness of a solvent such as isopropyl alcohol is going to depend on concentration and volume relative to the amount of grease you're trying to remove. 71% pads work... on very tiny amounts of grease. A squirt of 91% works better, but if there's a lot of grease, you need a lot more alcohol. This is why a wash under the sink with Dawn is so effective: There's a much larger volume of Dawn-and-water attaching to grease and rinsing stuff away instead of just moving the broken-down grease molecules around.
  
And above all else, avoid touching the PEI print surface.
 
Glue stick is NOT necessary for PLA on the PEI print surface. You paid the big bucks for a Prusa with a removable spring still sheet with PEI, so clean it up and take advantage of it. Save the glue stick for printing sticky stuff like PETG that might adhere too well. Even then, I only find I need glue stick when printing high-temp PETG at 260C+ temps.
 
Try those 2 steps. If you're uncertain of the Live-Z results, post pics here of the 75x75 print bottom and you'll get quick help. If you want more detail, I've consolidated my notes on Prusa PEI adhesion, bed cleaning, and Live-Z calibration.
 
If, after doing all this, you still find narrow or tall parts popping off or warping you might try adding mouse ears to corners and thin parts,  adding a brim, or applying some additional adhesion booster to your bed. I'd go in that order for best consistency and ease of cleanup.
 
A bit of trivia - The reason higher concentrations of alcohol seem to be harder to find is that isopropyl alcohol is most effective against bacteria at roughly a 71% concentration. A bit of distilled water helps slow evaporation, making the alcohol more effective. If you're looking in pharmacies, that's why 91% and higher seem to be hard to find.
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 : 01/06/2021 1:52 pm
Sink
 Sink
(@sink)
Estimable Member
Topic starter answered:
RE: Perplexed ??

@bobstro

Thank you for your detailed suggestions. Did clean the plate with dishwasher liquid, paper towels, and the 99.9 IPA. Redid the live-z adjust and did actually moved the Z down a bit, 25 pts or so. It made no difference, same problems with both pre-packaged gcode and my own sliced.

I tried adjusting MVS and reset the  Prusa PLA default from 15 down to 10 for a trial run and did see some improvement, not much but some, but that is not surprising as the benchy is small.

Trying one more run at 80% speed to see affect.

After this, not sure how to go forward. Maybe acceleration and jerk settings, will see results of speed reduction test first.

Wish I could understand what I am doing wrong, Have 3 MK3S+ machines in production, and a 4th under construction on my workbench, but none of the 3 is reaching the print quality of the earlier MK3S or MK2.5 models. 

Had a fifth MK3s+ in shopping basket, but have cancelled it and ordered a new printer from a different vendor, if I am going to get ratty quality, my fault or not,  might as well pay a ratty price.

 

Posted : 02/06/2021 5:03 am
Sink
 Sink
(@sink)
Estimable Member
Topic starter answered:
RE: Perplexed ??

Speed test results, reduced speed to 80%, almost no difference. Must be a build problem, but no idea what, so it is offline to be disassembled. 

72 hours on this printer, and nothing but frustration to show for it.. losing my religion..

Posted : 02/06/2021 7:38 am
cwbullet
(@cwbullet)
Member
RE: Perplexed ??

@sink

I wish I could help.  I have 8 MK3S and MK3S+ and 4 Minis.  Never experienced anything like this.  I suspect you will not either.  It is probably something simple that your printer is attenuating.  

--------------------
Chuck H
3D Printer Review Blog

Posted : 02/06/2021 9:25 am
Share: