RE: Unable to determine what's causing this issue
Your live-Z is looking a bit low - you should not have ridges on top. A guesstimate is to make the value about 0.05 more positive, and try again. Adjust upwards util there are no ridges between strands. It might also be your extrusion multiplier is way off since even your top layers are overfilled. In Plicer, make sure you are using the default 1.00 multiplier and 1.75 mm filament diameter. It also wouldn't hurt to measure your filament diameter: measure the filament in about six places at different angles (filament isn't perfectly round) make sure it is 1.75, and if it isn't adjust the filament diameter in slicer to match.
And a check of extrusion rate is probably in order:
Preheat the extruder to 215c. Mark the filament 95, 100, and 105 mm above the extruder filament entrance. Be as precise as you can. Use Settings/Move Axis/Move E to increase exactly +100 mm - do not overshoot or you'll need to do the test again. The filament should extrude to the 100 mm mark. If the 100 mm filament mark is past the entrance, your Bondtech gear may be offset. If it is centered, then you will need to adjust the Plicer extruder ratio according to your measurement. If the extruder moved 103 mm, then adjust the ratio to 0.97.
If the extruder moved 108 mm, then 0.92, etc. If the extruder moved less than 100 mm, it should be corrected, too: 96 mm == 1.04 multiplier. It doesn't fit the symptoms so something else is causing over extrusion.
Try printing the sample BENCHY gcode on the SD card, and not your own slice. The samples are tuned to the MK3 and are useful as a reference point for finding printing problems. Home slices leave just too many variables to guess at.
And at print time, the slicer has been calibrated for a normal layer compensation of 95% as shown on the LCD. Don't change that value.
As for the part popping off, the bed is not clean. I'm not sure what the mechanism is that causes the human brain to fight simple solutions. But washing the bed with soap and water works. Really. And you aren't the first person to swear the bed is clean when it isn't. Until you accept the fact the bed is contaminated with oils, you will continue to have problems.
Don't take any shortcuts here. Do NOT use your dish rag, dish sponge, or dish towel in any of this. Only use new, fresh, unused plain paper towels without any hand softeners or fragrances. After washing, the water should bead and sheet off leaving only a few drops here and there to dry.
Hot Water wash: often, as needed (and is helpful after any of the following methods)
- Handle the bed only by the edges.
- Wash the bed in hot water, use a fresh paper towel as a wash cloth, with a few drops of plain dish soap (Dawn, unscented, no anti-bacterial, etc.).
- Rinse well in hot water - if you have very soft water, rinse a bit longer. Water should bead and sheet off.
- Dry the bed with a fresh paper towel.
- Handle the bed only by the edges.
- Place bed on printer.
A new sheet should not need this, but a scrub with acetone can remove PEI oxides that over time develop on the surface. Follow the acetone scrub with soap and water to remove any residue from the scrub. Alcohol can help remove PLA residue after dozens of prints. Again, follow with soap and water.
Again - until you get the adhesion problem solved, you aren't going to be happy with the printer. And you can ask anyone with experience - a drop or two of soap, a good long scrub with a paper towel, and plenty of very hot water during rinse is the cure for cleaning the PEI sheet.
Well, there is that one segment that says sanding with 400 grit every other print is required.
So back to the basics: clean bed is mandatory; test only using sample gcodes.
RE: Unable to determine what's causing this issue
Thanks so much for the continued advice!!
I'll try your suggestions hopefully tonight.
And a clarification, I used the gcode benchy directly for my printer model and did not slice it using PrusaSlicer.
For other STL files (like the test circle you provided): I know all of the slicer settings are the defaults as shipped with the software (and via the profile updates) -- as I've never touched any of them (other than supports on/off). Nonetheless, I'll look at them to confirm they're as you described.
To be clear, I did clean the surface with soap as described earlier with hot water and fresh paper towels. I will definitely do it again before the next tests, and be even more thorough.
I thought that the part may have popped off when it came in contact with some of those errant strands of filament that had built up on the top.
RE: Unable to determine what's causing this issue
I've had some serious nozzle to part collisions: I've had parts break loose from supports, small detailed parts with less than a square cm of contact surface come loose from collisions, but have never had a large part like a benchy knocked off the bed. The dino skull was problem, but that was fingerprint oil on the bed.
You may want to try an acetone wash... who knows, maybe the PEI is old. Acetone is helpful once in a while, but not something needed every day. Monthly at most in my experience and only after lots of printing. Use gloves.
I have had one spool of filament that was contaminated: the entire first layer was oily from the factory and would not stick. Once I worked through that first layer adhesion came back. But that was one out of many dozens of spools. My guess is whoever packed it at the factory had just applied hand lotion. When testing it, the filament was coming out of the nozzle like foam. Not a solid 0.4mm thread, but one that expanded to a 0.8mm stream of foam. It looked similar to this contaminated filament (this sample was from my over handling):
RE: Unable to determine what's causing this issue
Status:
Bed cleaned thoroughly.
Extrusion measured to be close to 97-98, so I modified extrusion modifier settings for filament (1.025)
Adjusted the live-z to be +.05.
Printed the 60mm circle.
It did print. 🙂 I tried without the extrusion modifier and that print failed.
Extrusion modifier set to 1.0:
RE: Unable to determine what's causing this issue
After seeing what your bed looks like, you are way past due for a good acetone scrub; even a new PEI sheet - they are replaceable.
Since the test pad stuck, though. The circle looks slightly over extruded - either too much material flowing or Live-Z a bit low. Or is that photo showing two layers of material? For layer 1 tests - you should only care about setting Live-Z and stop the print before a second layer starts.
As for why there are loose strands on the second print? A 3% change in extrusion probably won't cause that - all that should do is reduce the width of the trace being extruded a fraction. Since the issue started with reconstruction, did you happen to replace anything in the hot end? If you did pull the hot end apart, did you follow the E3D-V6 assembly guide? There are some steps that are critical: setting the nozzle and heat break depths and proper torque at temperature procedure are key.
How about posting a photo of the nozzle and heater block from underneath and side. And describe what you did when you repaired the printer.
Another factor is age and how the filament has been stored. But I don't see any signs the filament is 'bad' ... so it's most likely a temperature control or flow issue. You aren't hearing any loud clicking at the extruder as this is happening are you?
RE: Unable to determine what's causing this issue
The 60mm circle you provided is 2 layers high. I let the print finish with the extrusion modification after seeing the first one failed (I hadn't noticed when slicing it that it wasn't just a single layer high).
I don't know why there are loose strands unfortunately. The strands aren't always adhering to the previous layer. I only made the one change between the two prints, the extrusion multiplier.
I think these issues were starting to happen before disaster struck and caused the damage that resulted in me rebuilding parts of the extruder assembly. I did follow the guide for the E3d-v6 that was provided by Prusa to me.
What visual signs should I be looking for to determine I'd need to do the acetone scrub? I need to wait on buying more parts but could try the scrub.
There's no audible clicking during extrusion that I've heard in some YouTube videos.
Here's what I bought to replace:
| -- |
|
|||
When trying to disassemble it, the heatbreak snapped in two pieces (there was a giant mass of hardened PLA filament). So, I also bought a
Prusa MK3 Heatbreak
Prusa, via chat, provided me with the links and clarification I needed to reassemble the parts.
I took this photo last week when trying to troubleshoot the issue with Prusa chat. I don't have one right now without the silicone sock. (I've not used the sock during these recent tests).
FYI - There's some filament "residue" baked in at the back -- it's not leaking, it's from the giant filament ball that destroyed the printer. It's cosmetic -- I scraped away what I could safely.
RE: Unable to determine what's causing this issue
Old filament that concentrates on the nozzle can cause problems and increase the chances of new layers sticking better to the nozzle than the part. But I'd rate it lower on the list of causes. But you can use a paper towel and heat the nozzle and wipe most of the plastic off the nozzle pretty easily. I also will heat the nozzle to about 100c and push an old part into the nozzle - it sticks to the warm plastic and then I let is cool and everything pulls off (think external cold pull).
The only thing that looks odd about your hot end is the HEX on the nozzle seems embedded to far into the heater block. You might check to make sure the HEX portion does not touch the block. It should stand off a 1/4 turn minimum. If it is too high, too tight to the block, the heat break is too high and might cause problems by not seating fully against the nozzle.
This is the gap that should be there.
RE: Unable to determine what's causing this issue
And - since you've replace the hot end components - have you run a PID calibration? This compensates the printers thermal response to thermal ramps and step changes, like layer 1 to layer 2 where the nozzle temp is changed 5 or 10c and the print fan kicks on. There are cases where this causes a dramatic drop in nozzle temp and extrusion problems for quite a while after the temp change.
RE: Unable to determine what's causing this issue
Still no luck. 😕
I adjusted the nozzle.
I ran the PID calibration and tweaked the Live-Z.
I cleaned and cleaned the Prusa printing surface and did a print ... it failed and became unstuck eventually.
I also tried a textured 3rd party printing surface I'd bought a few months ago when the printer was still chugging away -- I'd never used it though (and adjusted the live-z for that surface). I followed the directions for it exactly. The print failed and became unstuck.
I was trying a retraction test print this morning and noticed that the surface of the print on layer 3 or 4 was looking rough and fuzzy... I'm not sure if that's indicative of anything. I've got the extrusion modifier still set to 1.025 for the filament I'm using. The part was starting to unstick as it printed. It prints so fast it's hard to tell, but it's almost like it was pushing into the existing filament rather than just putting a layer of filament down.
RE: Unable to determine what's causing this issue
Text advice is sometimes problematic. The nozzle now looks a bit extended. The E3D-V6 instructions say 1/4 to 1/2 turn out. Your nozzle is looking like a full turn. But that small delta probably isn't the root cause.
PLA sticks to a clean PEI sheet very well. Perhaps it is time to buy a new sheet, do the initial cleaning and see how it goes.
Some report that a light sanding of the sheet with 1000 grit sand paper works for those who can't get the sheet clean. I don't recommend it because I've always been able to get the sheet sticky again with soap and water and once in a while resort to acetone to remove the white oxide residue that happens with time.
At this point I don't believe I am helping. Maybe someone else has suggestions.
RE: Unable to determine what's causing this issue
I did a full turn with the nozzle because of this text in the instructions:
Then, unscrew the nozzle a full turn. This will leave a little space to tighten after screwing in the heat break.
As I mentioned, I did have a new textured PEI sheet that I tried ... and that also failed in nearly an identical way.
Since adjusting the nozzle didn't change the outcome, it does seem like there's some other problem unfortunately. 🙁
Unfortunately, it seems like this is a giant waste of time and useless printer at this point. It's too bad that these things can be so difficult to troubleshoot and fix consistently. I've spent a phenomenal amount of time with it. 😖 I wish Prusa could help more. I'm glad I got quite a few good prints out of it over the last year.
Thanks for trying.
RE: Unable to determine what's causing this issue
The textured sheets have a known PLA adhesion issue - and most owners have resorted to an initial soap and water wash to remove whatever factory residuals they ship with. Several threads on that.
All I can end with is your PEI sheet looks dirty and oxidized. I am not surprised PLA is not sticking to it.
My nearly two year old PEI sheet looks like this - and it's the only side I use... I used a brim on that part because I had modified the model to have a hollow under the base for adding weights - the base only had a 2 mm rim.
RE: Unable to determine what's causing this issue
A better shot of the sheet ...
RE: Unable to determine what's causing this issue
🤔
(I followed the instructions of my PEI textured sheet that said "DO NOT WASH" in big red letters. )
RE: Unable to determine what's causing this issue
yep, and most of us agree that a careful wash is the best thing you can do to help a PC sheet work with PLA - just make sure you never leave the sheet moist when storing it to prevent rusting