My Prusa i3v 10" Build === Photos, Comments & Questions

[printbus]

I would reduce the retraction distance to something like 1mm. Cura defaults to a high distance value since Ultimaker uses a Bowden extruder that requires it. 1mm to 1.5mm should be adequate for most filaments on a direct extruder. If I understand the min extrusion before retraction setting, the default of 0.02mm seems awfully small. Combined, the settings will reduce the number of times retractions can cause the same section of filament to be rolled past the hobbed bolt.

BTW - my settings are being used with 1.75mm filament. 3mm could require something different, if that's what you have.

[jtice]

Did a few more prints last night, and things seem to be a bit better.

I took alot of the advice here.
Lowered the retraction speed and distance. Also been playing around with temps a bit, went from 230 to 234
I dont get nearly as many loud cracks and creaks anymore, just little ones now.
It seems to do it when the extruder starts and stops alot very quickly, not during retractions.

I have not got around to the second calibration test yet, where you print the solid cube.
But I was playing around with the tune section on the RAMPS.
I adjusted the flow from 100 to 104 and I think it helped with filling gaps better, before my fill seemed to be broken up a good bit.
I also think it was making my layers stick a bit better, but this still needs tweaked.
I am having some layer separation not during the prints, so I dont think its really an ABS contracting issue.
The parts are just weak, I can break them along the layers in spots.
The last print I did was at 104 flow, 234 head and 96 bed, that seemed to help some.

Here is the pic of the hob marks on the filament. Though, I have it sliiiightly looser than this now.


I will upload a vid today of the "creaking" noise I still have.

[jtice]

Here is the video of the creaking.
3 or 4 times you will hear a higher pitched creak/tick. Thats the relay kicking on and off.


Also been checking motor temps, all seem fine, the extruder motor gets fairly warm, but can still comfrotably hold my hand on it.
It reads about 43c, X and Y motors read about 40c
My heat bed reads about 75c at the edges and about 86c toward the middle when set to 90c, guess thats normal.

[gmay3]

I also think it was making my layers stick a bit better, but this still needs tweaked.
I am having some layer separation not during the prints, so I dont think its really an ABS contracting issue.
The parts are just weak, I can break them along the layers in spots.

Hmm, I've had these issues with PLA which I'm not very experienced in printing, but not with ABS. In addition to temperature tweaks, you also could try slicing at very slightly different layer thicknesses as an example, instead of .20mm maybe .22 or .19. Maybe there is a region between the mathematics of the slicing software and the steps of the z motors that is not meshing completely.

I think what could be more likely is that the blend of polymers in this specific ABS filament may not be the most ideal for 3d printing. From what I've heard, ABS plastic is a generic "recipe" that can vary greatly between manufacturers. I have only used Makerfarm filament so far, and I feel like their ABS is very good quality since I haven't had any filament issues so far. If you continue to have these issues it might be worth trying a roll from them or even another vendor to see if this could be your issue.

Here is the pic of the hob marks on the filament. Though, I have it sliiiightly looser than this now.

I will upload a vid today of the "creaking" noise I still have.

I think even just seeing the picture it says to me that your hobbed bolt is biting into the filament too much. Those teeth marks look like they are digging into the filament about 35-40% where you would probably want 10-15%. When I have reversed my filament the depth of the hobbed bolt markings are only around .2 mm max.

Like you said, I think loosening those two screws that control this amount of "bite" could help. I think the general mindset is that they should be as loose as possible without allowing the filament to slip.

[printbus]

I agree with gmay that the hobbed bolt is cutting into the filament way more than it should have to. The deep cuts will weaken the filament, making it easier to fold instead of extruding. Run the filament back and forth a few times for closely-spaced retractions and you'll definitely have a weakness in the filament.

The 0.20mm layer height shouldn't be an issue. Some printers (like Printrbot) require strange optimal layer height values since they use Imperial threaded Z-rods and the layer height values have to compensate for the Imperial-to-metric conversion. 0.22, 0.20, and 0.18mm would all come up with integer step counts to achieve those layer heights on the i3v. 0.21 and 0.19 do not. (You can play with the optimal layer height calculator at http://prusaprinters.org/calculator/).

jtice, you're not indicating how much force is required to break the layers apart. This may not be unusual for any print based on FDM. With ABS, if I don't have a thick enough wall or an infill, I tend to get layer separation as the print cools. If I know I'm going to need to drill holes in the finished print, I know to still use ABS though since the drill bit will tend to shear the layers apart in PLA more than with ABS.

On the tick sound, it's hard to say if it's similar to what I had. Somewhere in the following I fixed mine: 1) My hobbed bolt is a loose fit in the bearings. I have some kapton tape wrapped around the hobbed bolt where the bearings ride to help keep the bolt from possibly slapping back and forth inside the bearings. 2) the inner races of the bearings of the hobbed bolt might have been binding up on the plastic of the extruder base. My original install had the taped hobbed bolt very snug inside the bearings, and I had a theory that the binding of the bearings would stick and break free, stick and break free, etc. with a click each time. In my latest rebuild, I reduced the amount of tape on the hobbed bolt so it isn't as tight in the bearings and didn't tighten the hobbed bolt up as much so the bearings aren't pulled as tight to the base. 3) I added a bit of lubricant on the extruder base where the bearing inner races ride so they would be less likely to "stick". 4) I'm also now running with new bearings.

[gmay3]

The 0.20mm layer height shouldn't be an issue. Some printers (like Printrbot) require strange optimal layer height values since they use Imperial threaded Z-rods and the layer height values have to compensate for the Imperial-to-metric conversion. 0.22, 0.20, and 0.18mm would all come up with integer step counts to achieve those layer heights on the i3v. 0.21 and 0.19 do not. (You can play with the optimal layer height calculator at http://prusaprinters.org/calculator/).

Ah it makes sense now that this is the reason. I think I was reading this from an older slic3r calibration guide which suggested this could be an issue. Thanks for the correction printbus!

[printbus]

Ah it makes sense now that this is the reason. I think I was reading this from an older slic3r calibration guide which suggested this could be an issue. Thanks for the correction printbus!

We'll just call it a clarification and not a correction. You do see the optimal layer height issue brought up a lot. IMO, I think it's good for everyone to understand the reason for the possible issue. For me, it was quite enlightening to experiment with different threads and layer heights in the prusa calculator. I thought about sending Colin a thank you for using metric rods after I did that.

[usarmyaircav]

Printbus, why did you decide to use new bearings?

[printbus]

Putting in new bearings was part of a brute-force, balls-to-the-wall, do-everything-I-could-think-of approach to eliminating the annoying click/creak I had in the area of the filament idler on the hobbed bolt.

[jtice]

Once again guys, thanks for all the help and input.

I think my photo exaggerates how deep the marks are in the filament, but still from your descriptions I think I may have it too tight still.
I will try loosening it a bit more, and get to a point where I am just tight enough for it not to slip.

As far as the layer separation.
After printing a spool bearing holder, just grabbing it and barely leveraging it over to see if it would pop off the bed, it broke.
Way less force than I would say is reasonable.
The second one I printed a bit hotter, and turned the flow up from 100 to 104 (how much is that? wasnt sure what an increase of 4 actually was)
It seems stronger, but still not great. I think I am going to use some acetone to glue the weak layers and try to salvage them.

So far I have only used Winbo ABS from eFilament.
I was going to try some other ABS I got off eBay, but its a wider spool and wont fit the Prusa spool holder. grrrrrrr
Thats why I was printing the bearing spool holder, well that and it just seems like a better holder option.
Very curious to see if it prints any stronger.
My other prints seemed to be a bit stronger, but the bearing holder was my second tallest print.
The Clough42 fan shroud is my tallest so far, it looks ok, but I can tell that I would easily break its layers if I wanted to. So might reprint it.

BTW, the bearing holder was printed with these settings, far as I can remember...
Cura Slicer
50mm/s speed
0.2mm layer height
1.2mm top/bottom/shell (I should make top/bottom a multiple of layer height, and shell a multiple of the 0.4mm nozzle.... correct?)
234c head
96c bed
Forget what fill, probably 15 or 20%
no additional cooling fan
flow on Prusa set to 104
F/R (speed) on Prusa set to 105%