Stream of I3V 12" Build Consciousness

[voodoo28]

Looks good! ....the bits of roughness you see on the edges could be that your retractions settings need some tweaking..

Was looking for something with a little height. I really picked the wrong thing, but it ended up being an interesting experiment anyway.

At the very least, I need to secure the printer to a rigid and stable surface. Still, if this is the worst it does in less than ideal circumstances..., it's not too bad.

https://youtu.be/BAD4ikaqFhE

Will be interesting to see what mods things like this might lead to...

[lakester]

Thx! That reminds me: what I should do is look for calibration shapes appropriate to this printer....

Looks good! ....the bits of roughness you see on the edges could be that your retractions settings need some tweaking..

[lakester]

So I needed a knob for the LCD panel. Who doesn't?

While I was assembling the extruder, I knew what it had to be, and that it might be a good opportunity to begin learning FreeCAD.

On the first go around, I simply doubled the size of the extruder motor pinion, closed up some holes, and put in a 6mm hole.

The print produced something like a 5.715mm hole, and it didn't fit.

Of course, 6mm wouldn't have fit either..., since I didn't measure the shaft particularly carefully. Went back and mic-ed it for real and found it was more like 6.1mm diameter. Adding a 5% allowance seemed like too much, so I kinda split the difference and tried a 3% shrinkage allowance (and added a bevel to the hole to match the shaft). That one fits and is the one in the picture.

Turns out, doubling the size of the part was a bit much. The rotary encoder is not mounted on the PCB in a particularly robust way, and I think this knob gives too much leverage, increasing the possibility of breakage. So..., I'll probably shrink it down to something 1.5x w/ a 6.3mm x 10mm hole.

Was a little disappointed w/ FreeCAD on Linux (Mint 17). The version in the package repo is 0.14, and is flaky as hell. Unusable for my purposes actually. So I'm using the most recent Mac version (0.15) and that seems to be OK, though I'd much rather be using the desktop for this stuff.

Anyways..., here's a vid and a pic...

https://youtu.be/X0VV6cXjzKU

20150513_085756.jpg

[lakester]

Per Voodoo's comments, I decided to try running some retraction tests.

For context:

• 3mm PLA (tan) filament.
• hexagon HE w/ .4mm nozzle.
• stock MF HE cooling shroud.
• heTemp=225c, hbTemp=70c.

Unfortunately, the final set of photos don't really present very well, so I'll only show some intermediate photos of the first two tests (the 30->20 comparison).

I ran the tests with retraction rates of 30mm-s(per MF slic3r config file for my setup), 20mm-s, and 10mm-s.

The most obvious change in quality came from the 30->20 test (which is what the photos cover). The 20->10 test showed some small improvements, but nowhere near as dramatic.

The article with the equally spaced pylons of all the same height showed the most improvement. With the exception of the base, all the layers are identical, and all segments are relatively close together.

The article with the tall corner pylons behaved much the same way, UNTIL the tall pylons rose above the rest, and whose rendering involved retracts w/ large X/Y moves. I.e., they spent the least amount of time in the downwash of the HE cooling fan. Once the 4 tall pylons, each spaced a significant distance from one another, began to be the only things rendered, things got worse.

Or put another way: the sections that spent the most continuous time under the HE cooling downwash did better than the ones that spent less continuous time in the downwash.

This makes me think that the addition of a part cooling fan might be useful.

All Short Pylons, 30mm-s on left, 20mm-s on right

20150513_210034.jpg


Tall Corner Pylons, 30mm-s on left, 20mm-s on right
20150513_210026-small.jpg

[voodoo28]

Im not sure how slic3r retraction settings go..but in simplify3d there is a retraction distance and a retart ...retraction distance on mine is set to 1.75mm..that's how far up the filament is retracted into the extruder before the next layer... Then restart distance tells it when to start extruder again before new layer.

[lakester]

Slic3r allows specification of the following related to retraction:

• Length (set to 1mm)
• Lift Z (0mm)
• Speed (defaults to 30mm-s)
• Extra length on restart. (set to 0mm)
• Minimum travel after retraction (2mm)
• Retract on layer change (true)
• Wipe while retracting. (false)
• A couple of multi-extruder relted items.

I only wanted to play w/ retract speed initially, since my sense was that if I started fiddling w/ length, I'd have to balance that against the "extra length on restart" value.

Lift Z would probably be a last ditch sorta thing, since you'd take a pretty big performance hit playing w/ Z on a Y-bed printer like a Prusa. Printers w/ a Z-bed MIGHT do better w/ this, and deltas I assume would.

FWIW.., Clough42's part cooling fan shroud (stretched version w/ ears added) is printing now..., hopefully will get to install it in the next 2-3 days. Heh..., this is my first ABS part. Now I know what curling is all about, hence the ears!

Im not sure how slic3r retraction settings go..but in simplify3d there is a retraction distance and a retart ...retraction distance on mine is set to 1.75mm..that's how far up the filament is retracted into the extruder before the next layer... Then restart distance tells it when to start extruder again before new layer.

[voodoo28]

You can try setting the length from 1mm to 1.75...that will help with the tiny blobs. I had mine set to. 1mm and was having the blobs too.

Slic3r allows specification of the following related to retraction:

• Length (set to 1mm)
• Lift Z (0mm)
• Speed (defaults to 30mm-s)
• Extra length on restart. (set to 0mm)
• Minimum travel after retraction (2mm)
• Retract on layer change (true)
• Wipe while retracting. (false)
• A couple of multi-extruder relted items.

I only wanted to play w/ retract speed initially, since my sense was that if I started fiddling w/ length, I'd have to balance that against the "extra length on restart" value.

Lift Z would probably be a last ditch sorta thing, since you'd take a pretty big performance hit playing w/ Z on a Y-bed printer like a Prusa. Printers w/ a Z-bed MIGHT do better w/ this, and deltas I assume would.

FWIW.., Clough42's part cooling fan shroud (stretched version w/ ears added) is printing now..., hopefully will get to install it in the next 2-3 days. Heh..., this is my first ABS part. Now I know what curling is all about, hence the ears!

[printbus]

10 or 15mm/sec is a more reasonable speed for the extruder. Also, view the gcode in a viewer like gcode.ws. Blobs that occur somewhere other than moves or layer shifts can't be due to retraction.

[voodoo28]

I second that on the Retraction speed...I used to have my set to default by 30mm...never worked correctly until i dropped it to 15mm.

10 or 15mm/sec is a more reasonable speed for the extruder. Also, view the gcode in a viewer like gcode.ws. Blobs that occur somewhere other than moves or layer shifts can't be due to retraction.

[lakester]

Thx all on the retraction comments.

ATM, I have ABS loaded and printing. I figure that while I have it in the machine, might as well as print off some ABS specific stuff I've had in mind.

While it's there, will run the retract test on the ABS once the various jobs have been finished.

Thx for the pointer to the gcode viewer..., cuz my eyes just don't hack it anymore when it comes to sussing out when things go wrong in tiny little places.

When I get back to PLA, I'll try 1.75mm for the retract distance at 20mm-s. If that gives a positive indication, will try at 15mm-s.

I'm also VERY interested to see if the parts cooling fan makes a difference on the "4 towers" part of the test. Not sure if I have all the parts handy to get that wired as soon as I'd like..., we'll see.

Anyways, will get back to more fun w/ retraction after some bla bla bla on my part on ABS once tonight's/tomorrow's stuff gets done.