Welcome! Log In Create A New Profile

Advanced

Critique This Torture Test Please

Posted by Idolcrasher 
Critique This Torture Test Please
March 21, 2013 02:24PM
Good Day All,

I recently built a hybrid RepRapPro Mendel / Ecksbot. I am getting some hit or miss results with it. I thought I would print out a torture test and ask for what adjustments you think are in order smiling smiley

I am using Slic3r 0.9.8. I won't contaminate your thought process by telling you what settings I am using exactly winking smiley

I am particularly interested in your thoughts on how to smooth the cylinders.

I would also like to hear your methods for reducing lifting overhangs.

Thanks for your thoughts smiling smiley














FabberForge - Printing Happiness

[www.fabberforge.com]
Re: Critique This Torture Test Please
March 21, 2013 02:34PM
I think you are over-extruding 2-5%, based on the cylinders and the corners of the box.

You may also be printing slightly too hot, based on the stringing and bumps on the cylinders. The amount of curling might indicate you're running your fan(s) too high.

I still see some slight Z wobble present.

Did you print the arch with support?

Here's my best torture test to date, printed on my work-in-progress H-bot. [i.imgur.com]
Re: Critique This Torture Test Please
March 21, 2013 02:42PM
I tend to agree with everything you say.

I did a few extrusion calibration tests and everything came out perfectly, which then leads me to think that my extrusion width is actulaly a bit low for my nozzle.

I am now adjusting my layer width from 0.44 to 0.51 based on a free extrusion that I just did. With a layer height of 0.231818 my width vs. height ratio is now 2.2

I am running a bit hot because the PLA I am using is having a little bit of trouble adhering to itself.

I will test out this new height and width, and then dial in the extrusion multiplier using a single walled object and a pair of digital calipers.

Then I will revisit the temp.

On Z-Wobble: I am currently using ACME Spindles and have them attached to the Z-Motors with printed U-Joints. I am not sure if I see z-wobble or over extrusion... or both.

Edited 2 time(s). Last edit at 03/21/2013 02:50PM by Idolcrasher.
Re: Critique This Torture Test Please
March 21, 2013 02:43PM
That was a nice pic BTW smiling smiley

Wait, your box is not closed lol.

No support on the arch

Edited 1 time(s). Last edit at 03/21/2013 02:43PM by Idolcrasher.
Re: Critique This Torture Test Please
March 21, 2013 02:45PM
The best z coupler, hands down, is a short length of flexible aquarium tubing or other tubing. Leave a 2-4mm gap between your motor shaft and the threaded rod in the tube.
Re: Critique This Torture Test Please
March 21, 2013 02:47PM
Those are in vogue right now.

I saw one of the admins talking them up years ago. They seem to be gaining popularity since Prusa used them on the i3.

Edited 1 time(s). Last edit at 03/21/2013 02:54PM by Idolcrasher.
Re: Critique This Torture Test Please
March 21, 2013 03:20PM
Anyone have a picture of this aquarium tubing? IS it hose clamped?
Re: Critique This Torture Test Please
March 21, 2013 04:18PM
im in the middle of testing 4mm tubing on my i3 vs cooper compression couplers.

From what iv seen so far i think the tubing might be better giving the Z a bit of give
Re: Critique This Torture Test Please
March 21, 2013 05:18PM
This i3 is so much better looking that the previous generations. They are still using grossly overpowered stepper motors for some reason though. The stepsticks provided with my prusa provide 1A output, and the steppers they provided take 1.66A current. They are running way below capacity.

I didn't realize the connector was a press fit. I do not trust it as much now since once it starts to slip it won't grip again. It would be better with glue or hose clamps or something, even a twist tie.
Re: Critique This Torture Test Please
March 21, 2013 06:06PM
I wouldn't mess with the torture test until I have the single wall part in the slic3r calibration wiki to come out right and until I can get one of the small bridge test like Caru's or Starno's to print right.

Those prints suggests extruder problems to me. Retract setting not optimal or extrusion temperature being too low can result in strings.

Quote
crispy1
The best z coupler, hands down, is a short length of flexible aquarium tubing or other tubing. Leave a 2-4mm gap between your motor shaft and the threaded rod in the tube.

If the motor is in the bottom, doesn't that gap allow for the rod to move up and down? This can also result in z-wobble effect as the layer thickness can vary.

Edited 1 time(s). Last edit at 03/21/2013 06:09PM by brnrd.
Re: Critique This Torture Test Please
March 21, 2013 06:10PM
The best z-coupler, hands down, is a real motor coupler, not make-shift pieces of tubing.
Re: Critique This Torture Test Please
March 21, 2013 06:19PM
Quote

If the motor is in the bottom, doesn't that gap allow for the rod to move up and down? This can also result in z-wobble effect as the layer thickness can vary.

I don't think anyone has seen that issue, but Josef would be the person to ask, since his i3 motors are at the bottom and he uses tubing. If you use the right size tubing the motor shafts and threaded rods are very tightly held. If you are worried about slip, you can always add a couple of zip ties around the tubing to give it some extra grip.

Quote

The best z-coupler, hands down, is a real motor coupler, not make-shift pieces of tubing.
Please show me side-by-side prints that prove your point. Many people tried them, including most of the original ORD bot kits, and most of them had issues with Z wobble.
Re: Critique This Torture Test Please
March 21, 2013 06:21PM
"Josef would be the person to ask"

Not really. Joseph is, like you, an amateur. He's is the last one I'd ask for technical advise.
Re: Critique This Torture Test Please
March 21, 2013 06:30PM
xiando Wrote:
-------------------------------------------------------
> The best z-coupler, hands down, is a real motor
> coupler, not make-shift pieces of tubing.

Yep
Re: Critique This Torture Test Please
March 21, 2013 06:45PM
@xiando

Luckily I wasn't suggesting we ask him for technical advice, I was suggesting we ask him if he's encountered issues with his setup. You know, actual usage history.

This may shock you, but I'm not an amateur - I have a ME degree, I just don't wave it in peoples' faces all the time.

I don't know who wedged that stick up your ass, but please remove it and work on your "I know more than all of you" attitude. Because even the brightest person on this forum doesn't know more than everyone else on here all the time.
Re: Critique This Torture Test Please
March 21, 2013 07:37PM
calm down friends smiling smiley

Everybody chillax winking smiley

I've done and redone the Slic3r single wall calibration. That is my go-to calibration. The single wall pieces look fine and dandy.

Some of my prints are still getting these little dots. Is this a retraction problem, or an over extrusion problem?

Increasing retraction did not seem to get rid of the dots. Who has thoughts on the dots? winking smiley





One side smooth, one side, not so much.
Re: Critique This Torture Test Please
March 21, 2013 08:03PM
I've seen this. The extruder is loose on the bushings, wavying side to side
Re: Critique This Torture Test Please
March 21, 2013 08:59PM
ok, the parts on this machine are looking much better now. thanks for all your advice.

One of the primary culprits was the "Randomize starting points" feature in Slic3r. It was causing some weirdness with starting and stoping locations in curved objects.
Re: Critique This Torture Test Please
March 21, 2013 09:47PM
Quote

I've seen this. The extruder is loose on the bushings, wavying side to side

Pretty sure that is belt resonance from the abrupt changes in the velocity vector. Decreasing the XY jerk value in the firmware will mitigate it.
Re: Critique This Torture Test Please
March 21, 2013 10:40PM
Give it a try with KISSlicer too. I got smoother cylinders with KS, all settings as equal as I could get them. Also brim is probably the only reason the arch is standing. Every version of Slic3r knocked it over for some reason, but KS and Skeinforge didn't.
Re: Critique This Torture Test Please
March 22, 2013 05:34PM
Another thing that might cause issues on one side is a nozzle that dies not have a completely symmetrical opening. Perhaps the nozzle is not perfectly normal to the build platform, or not completely square on the end. I fave heard of people placing very fine grit sandpaper in the bed and running the extruder back and forth over it to ensure everything is flat and square.
Re: Critique This Torture Test Please
March 22, 2013 06:48PM
Greg Frost Wrote:
-------------------------------------------------------
> Another thing that might cause issues on one side
> is a nozzle that dies not have a completely
> symmetrical opening. Perhaps the nozzle is not
> perfectly normal to the build platform, or not
> completely square on the end. I fave heard of
> people placing very fine grit sandpaper in the bed
> and running the extruder back and forth over it to
> ensure everything is flat and square.

OOPS. That is 100% the problem. I remember now because I once sanded a nozzle back and it was uneven and caused weird issues on one side but not the other grinning smiley.
Re: Critique This Torture Test Please
March 22, 2013 11:17PM
Quote

One of the primary culprits was the "Randomize starting points" feature in Slic3r. It was causing some weirdness with starting and stoping locations in curved objects.

Looking at the bumps on the wall, all this would have done would have been to line them up. I expect that the bumps, which are due to retract-restart problem, would still be there.
Re: Critique This Torture Test Please
March 22, 2013 11:23PM
Quote

Quote

The best z-coupler, hands down, is a real motor coupler, not make-shift pieces of tubing.

Please show me side-by-side prints that prove your point. Many people tried them, including most of the original ORD bot kits, and most of them had issues with Z wobble.

I would also like to see proof of this claim. I agree with cripsy1.
Sorry, only registered users may post in this forum.

Click here to login