Weird noise and print failing at same spot  

  RSS
dakiki
(@dakiki)
Eminent Member

Hi,

 

I've downloaded Eclipson Free RC Plane files and started printing them. I've successfully (and very nicely) printed 5 parts, when I started to get some failures. I tried to print the first part using their generated gcode files instead of slicing myself, to see out it goes. As the print seemed very fine to me (no artifact, no error, really good print), I kept using the generated gcode files for the remaining parts.

When printing the last successful I made, the printed started to make some noise when the Y-axis moves, so when the print completed I used the provided lubricant, as I always do when I start hearing this kind of noise, which always had solved the "issue".

Then I started printing the next part which failed around 45mm height (not sure exactly), I heard the nozzle rubbing against the part so I checked the print and canceled it (left part on the picture below). I printed again and the part started to fail at the exact same height (again, not sure exactly but very close to 45mm, as I quickly canceled the print due to the same rubbing noise I heard). Second failing part is on the right of the picture.

I also took 2 videos to record the printer's sound as it doesn't seem right to me. Actually the first 5 minutes are really quiet, then I started to hear a bit more noise (first video), then a few minutes later (just a few layers before the print fails) much more noise, like if the rods are not lubricated enough.

First video:

Second video:

I didn't try to slice it myself, nor to try a 3rd print, but I'm currently printing the next part and I don't hear any noise at all (very very quiet), and the print is already at 90mm of height.

Any ideas?

I checked belts, ran a self-test, calibrated again, lubricated all rods.

This is the failing gcode file, I loaded it in slicer but couldn't seem to find anything weird around 45mm

Fus3.gcode

 

Thanks a lot for your insight.

Posted : 09/05/2021 4:04 pm
--
 --
(@)
Illustrious Member

Unless you slice the gcode yourself for your specific printer, the chances of it working well are pretty slim. And it gets even more slim when there is a line like this in code being sent to a Prusa Mk3:

; Ender 3 Custom Start G-code

 

Posted : 09/05/2021 4:28 pm
dakiki
(@dakiki)
Eminent Member

@tim-2

That's what I thought, and that's the reason I started with a sample to see if it was ok.

I actually printed all parts OK, except this one... I'll try to get the correct settings then and see if I can slice like they did. As it's for an RC plane, the slicing has to be tuned very precisely if you want the stuff to fly 🙂

And regarding the noise, any idea? Do you think it's related to the gcode being generated for another printer?

Posted : 09/05/2021 5:46 pm
--
 --
(@)
Illustrious Member

@dakiki

I'm not in front of your printer, so no clue about the noise. But I'll guess based on: motor currents are wrong, K factors are wrong, the bed isn't being leveled, travel and print accelerations and speeds are wrong, bed limits are wrong, there are probably gcode commands the Prusa doesn't understand,  ... do I need to go on?

You can open up the gcode and look at the settings - I think I saw some of that in there. Prusa's Slicer embeds the settings at the end of the file. If not, you'll need to create a profile based on whatever the designer says: or, take apart the gcode and makes all the necessary mods for it to be Prusa compatible.

ps: and yes, in some cases there are slicing tricks to get the desired result because the model was designed for a specific slicer and printer in mind.

This post was modified 1 month ago 2 times by --
Posted : 09/05/2021 6:33 pm
dakiki
(@dakiki)
Eminent Member

Hi @tim-2

just a quick follow up.

I managed to slice myself the STL files, and I'm achieving almost the same result (not exactly but I think it should be close enough).

I managed to correctly print the part that was failing without any strange noise nor errors, thanks for the tips 🙂

Now, I sliced another part and it printed fine, just during the last few minutes I heard the same kind of rubbing noise, but even louder than on my video. I managed to watch the layer at which it started (not the exact one but close enough I think). The noise suddenly stopped before the end of the print, but this time I saw the exact layer it ended.

When I looked, the print was at 183.00mm so it started a bit before. It ended at exactly 185.20.

Looking at the gcode after the print (attached) using Slicer, I'm seeing some changes between height 182.20 and 185.20 which seems to match exactly my observation.

It looks like the speed changes and the volumetric flow rate (don't know what this is but I guess it's the amount of plastic being pushed out of the extruder).

Do you have any idea what caused this changes in slicer, and if it's bad for the printer. The high pitch sound seems to come from Y-axis but it's hard to tell.

Also attached is the 3mf file used to slice.

Thanks for any advice, I don't fully understand the impact of all parameters and I don't want to damage the printer.

Wing1L.3mf

Wing1L_0.2mm_PLA_MK3S_4h8m.gcode

Posted : 10/05/2021 11:10 pm
--
 --
(@)
Illustrious Member

@dakiki

First off - it's pretty difficult to physically damage the printer with settings changes. About the worst you can do is set the nozzle temp over 280c and leave it there for hours - you'll toast the thermistor. Other than that you'd probably have to try to do damage.

The speed changes are due to cooling settings : specifically layer print time below 15 seconds. I'd change print speed to 35 or 40 mm/s and set cooling to 10 seconds. This will cause the entire section to print at a fairly steady rate.

Filament:Cooling:SlowDown: 10
Print:Speed:External: 35 mm/s

Also - I'd set reduce elephant foot comp to 0.1 mm - enable a brim of 8 mm, change skirt to 2 loops. The brim is important on that wing section and when EFC is set at 0.2, the brim isn't well attached to the part and may detach during printing. A skirt is - by design - is not attached to the part and is doing nothing for print stability. 

Print:Advanced:ElephantFoot: 0.1 mm
Print:SkirtBrim:Brim width: 8 mm
Print:SkirtBrim:Skirt: set all back to defaults

My guess the noise is coming from where the wing support tabs begins at 182.2. How does that section look printed? Printing a concave curve without support is nigh on impossible.

I don't see any simple way to improve how it prints without modifying the stl, and minimum fan speed is already 100%. 

Or, the wing section may have separated from the bed allowing the part to twist - a single extrusion doesn't have a lot of grab. Hard to say - I wasn't there - lol.

I'll finish with stating the simple fact I have never printed an RC wing. So I can't say my advice is based on anything except observation and tool use. 

 

Posted : 11/05/2021 5:05 am
dakiki
(@dakiki)
Eminent Member

@tim-2

Thanks for all the advices.

Regarding the piece that came out the printer, it's just perfect. No warping at all, the piece didn't even move when printed.

Actually, it's designed to withstand flight, so that would explain why it's pretty stable when printed. I even had to bend a lot the magnetic sheet for the piece to came out, so it was really nicely attached to the bed. Also, the tip of the wing was printed exactly how I saw it after slicing, no issue at all.

I'll take a look at the settings you recommend, there are still (quite) a few parameters I don't understand 🙂

Regarding EFC, it was actually pretty hard to separate the wing from the brim (had to you my nails! :D), which would explain why the piece was really nicely attached to the bed.

Out of curiosity, what's the need for a skirt if a brim is already added? I mean what purpose does it serve when printing?

Thanks again, learning a lot 😉

Posted : 11/05/2021 7:54 am
cwbullet
(@cwbullet)
Illustrious Member

@dakiki

You can adjust your settings to make it easier to separate.  I have found this is especially tough with supports.  

--------------------
Chuck H
3D Printer Review Blog...
Posted : 11/05/2021 3:43 pm
--
 --
(@)
Illustrious Member

@dakiki

A skirt can help - according to guides I've read on the subject - to reduce print variation due to convection around the base of the part. Probably more important for higher temp plastics printed in enclosures than PLA printed in free air. And a full height skirt can be built up along with the part to form a blanket that allows the part to cool more slowly to prevent warp.

I use the default 1 or 2 layer skirt as an additional purge line to further reduce nozzle cling-ons and drool from affecting the first layer. I doubt it has any real effect on other print qualities. 

Posted : 11/05/2021 4:54 pm
dakiki
(@dakiki)
Eminent Member

Ok thanks for your answer, it's exactly how I understood it as well (and I also use it to wipe the nozzle after bed leveling :)).

So in my case with a very large brim, it has no use.

Just a note actually, I switched from a brim to a skirt in my latest prints, as I noticed the brim would also be printed inside my wing (no bottom layer, it's hollow). So I configured a 1 layer skirt, 0mm from model, 4/5mm width... same effect than my previous brim, except it's only printed on the outside of the model, which is what I want (just to have a better grip on the bed).

PS: No more sound issues with your suggested speed. I actually thought I was printing at 50, that's why I increased the speed to that value in my settings.. you helped me understand it never printed at that speed due to the filament cooling 15 sec value! Thanks!!

Posted : 11/05/2021 5:23 pm
Share:

Please Login or Register