Welcome to the SeeMeCNC Community

[edit - added summary]
tl;dr -- The bowden tube wasn't all the way down. The extruder was partially plugged. It gave inconsistent, stringy results. Fix that and the Rostock Max can do the 3dlabprint planes (using S3D - MatterControl may also work but I'll need to go back and test it).

I just finished building a Rostock Max V3. I'm am new to 3d printing. I was thinking I had it dialed in fairly well.

Then I tried to print a 3dlabprint aircraft and I have voids in the print after a filament retract. I've spent all day tinkering with retract distance, extra distance, speeds and temps but I cant' get the void to close without having loads of strings and globbing.

A layer finishes at the trailing edge of the wing, retracts the filament and flies back to the leading edge. The PLA doesn't appear for the first 10 to 15mm of extrusion on the leading edge.

So, any tips on how to get the extruder to start on dime?

Anyone with a V2 or V3 able to print an aircraft? What settings did you use?

Also, is there any way to control the G-Code produced in the slicing? I would be good to have it work the perimeter then alternate through ribs, keeping all the retract hops short, and alternate back to the start again. You could almost avoid retracting if you did that.

Dewey, what do you have your retract distance, retract speed and hot end temp set to? One thing that's recommended with the HE280 hot end is keeping the retracts short and slow. A high and fast Z-lift is also beneficial (4mm/sec @ 250mm/sec).

g.

I've tried a wide range of settings. The default started at:
Retract 2.5mm, Extra 0, Retract speed 100 -- that does the Retract Cubes test really well but doesn't do the wing well at all (major void after the long flyback).

I've tried(showing just the extremes, I did a sweep through the values):
Retract 1mm, Extra 0, Speed 100 -- stringy, still has the void
Retract 3mm, Extra 1.5mm, Speed 100 -- not stringy but blobby, still has the void
Retract 2.5mm, Extra 1.2mm, Speed 25 -- stringy and blobby

I tried temps 210, 215, and 220 (using PLA) - 220 looks a little glassy but other than that no change I noticed.

So would you recommend high and fast z-lift and a short, slow retract with a little extra un-retract?

Very curious as I plan to print a P-38 once my Rostock is running.

You don't need any extra beyond whatever the original default was. Try 2.5mm retract @25mm/sec with a 4mm z-lift @250mm/sec.

g.

Dewey, when I read your post this morning I wanted to reply but didn't know anything about 3dlabprint. By this afternoon I had my first attempt at a Spitfire nose using MatterHackers clear blue PLA on my Rostock V3. It took about 4 attempts to get started without horrible fails, but I managed to get a decent first rough part.

[img]https://c6.staticflickr.com/1/637/31760193581_5f902dc430.jpg[/img]

[img]https://c8.staticflickr.com/1/711/31067398663_9c4cf5791d.jpg[/img]

It weighted 33.6 grams, well below the 48 gram target set by the 3dlabprint guys. So now I have an extrusion multiplier set to 1.46 and am running another copy. Here are the settings I found to work well. Printing these parts which are essentially outlines without any infill requires some unorthodox settings (ok, that's my disclaimer )

SS nozzle 0.4mm, extruder 225C, Glass bed 85C, fans 100% after first layer, speed 30mm/s or SLOWER, 200% first layer width, retract 6mm, extra length on restart -0.5mm, retract speed 30 mm/s, Z retract 0mm (use wiping instead).

I'm going to experiment with slower speeds. It is an 8hr part at 30 mm/s but slower speeds usually end up with a better print.

If you PM me I can email you the MatterControl profile I'm currently running.

Thanks for the help.

My goal is to get a solid leading edge on the wing with a minimum of stringing. So far I haven't found a group of settings that work because the wing involves a retract, full bed transition slew, and re-extrude. It is missing the first 10-15mm.

I'm going to start a print with:

Test#1
PLA
Hair spray
Extruder Temperature 215
Bed Temperature 65

Retraction 2.5
Extra Length On Restart 0
Speed (Retract) 25
Z-Lift 4mm

Extrusion Multiplier 1.00
Extrusion width (First layer) 200%

Cooling fans off

Speed for Perimeters (Inside) 40
(Outside) 30
Speed for non-print moves 250

Perimeters 1
Top Layers 0
Bottom Layers 0

Fill Density 0%


Result: Extrusion isn't starting up on the LE. I'm going to switch to a faster "two cylinders at 200mm apart" test print and see what I get.

I have only had my V3 for a month but I can tell from the gooey-ness of your part that you should try two things: print slower at 20 mm/s and turn the fans 100% on after the first layer. Forget about the strings for now, you need to keep the integrity of the part.

Once you get the speed down, while it is printing turn down the extruder temperature in 5 degree steps and wait a layer or two. Keep turning down the extruder temp 5 degrees at at time until the the lines look bad or the extruder motor skips. Then add 10 degrees. This is how I get my first settings.

Since your V3 is new, I strongly recommend running at 20 mm/s for all extrusion until you get the feel for your printer. In my case I am going to slow back down to 20 or even lower to try to get some clean airplane parts.

@timskloss - I'll have a go at that.

As an additional test (to save time and plastic) I created two cylinders 200mm apart and 15mm dia x 3mm tall (that re-creates LE to TE distance). Same settings as above. They look fairly good to me (the photos show the near and far cylinders). The only blobbing I'm getting is at the "fly in" point as it alternates between the cylinders. Could matter control have a bug in the g-code? This seems like something specific to the wing g-code.

I wondered if tool path ordering is contributing to this problem. So I rotated the wing 180 on the bed (guessing that would generate a very different tool path). The first 10 layers drew very differently and worked great. I thought I had it licked. Then it "reverted" to "fly off half way through the LE and fly back later" and it was messed up again. This is really strange. I wonder if the filament is binding when the arms are near the extreme edge of the table.

The second try looks better but still has the voids in the skin near the raised text and underneath. I don't think the PLA is flowing properly in the hot end.

Expert question: How much force should be required to get PLA to flow from the nozzle I need to turn the extruder by hand about a 1/4 turn and hold the pressure for the PLA to come out the nozzle. It definitely is not 'free flowing'. How can this be fixed? I have tried temperatures from 190 to 225 and it only seems to flow sluggishly near the high end. Am I looking at a HE280 rebuild?

Here are some pictures of the part. The weight is now higher but still shy of where it should be: ~48 grams.

[img]https://c7.staticflickr.com/1/545/31849437406_263c13a861.jpg[/img]

[img]https://c5.staticflickr.com/1/454/31739298212_3ed25b0401.jpg[/img]

[img]https://c1.staticflickr.com/1/759/31849494656_033f3ca27d.jpg[/img]

[img]https://c8.staticflickr.com/1/655/31770376031_64d664a618.jpg[/img]

What PLA are you using? I got some Pro PLA from matter hackers that has been a huge pain in the ass, in the same color as your plastic. (transluscent blue) I ended up increasing the temp to 225 and getting rid of retractions and it seems to print OK now... I would suspect my themistor settings are wrong, but ABS prints fine at 230....

Qdeathstar wrote:What PLA are you using? I got some Pro PLA from matter hackers that has been a huge pain in the ass, in the same color as your plastic. (transluscent blue) I ended up increasing the temp to 225 and getting rid of retractions and it seems to print OK now... I would suspect my themistor settings are wrong, but ABS prints fine at 230....

The non-pro type. It was actually the first spool I bought. What PLAs work well for you?

Well, matterhackers stuff is usually pretty good, i have been using their ABS for a while, but i needed some transparent blue and all they had for that was PLA....

I had used hatchbox filament before, and i got good results, but i found with ABS the matterhackers non-pro abs performed better than the hatchbox.

That fly away you're talking about is mattercontrol letting the last layer cool as it thinks the layer is too hot still from the last go-round. I forget what setting it's called though.

minimum layer time and likely also enable extruder lift

I'm running out of ideas. I can get simple test cases to print great but the wing is a complex mess of retracts and moves that has voids. Is there a slicer that gives more control over the slicing process? I think if the outer skin could be kept as one continuous series of moves it could work.

Try Kisslicer. The free version should do a good job.

g.

Already tried it - I cant figure out how to get it to slice AND include the ribs/spars. So how do I set it up to for a hollow object, that contains hollow interior objects, with no top or bottom surfaces?

Here is what a mean; from Kisslicer it doesn't seem to know about the interior ribs/spars. When it generates the G-code it creates a segmented perimeter with loads of retracts on the filament and nothing on the interior. I can't figure out what I'm doing wrong.

Here is someone who had success but he had to use S3D to pull it off:

http://forum.seemecnc.com/viewtopic.php ... int#p74148

If Kiss is omitting the ribs, I'd run the model through netfabb, it's probably bad geometry in the original model. Slicers vary a lot on how well they deal with various errors in the original mesh.

Slicers also vary in how well they deal with extremely thin features. I don't know how thick the internal structure is, but it certainly could be thin enough to be problematic.

So 3dlabprint has some test parts that people can try. They have videos (links in the pdf manuals) that show how to configure the various slicers for their prints.

http://www.thingiverse.com/thing:1875834 or http://www.thingiverse.com/thing:1207209

They use Simplify3D at work so I spent the morning making the fff file for my Rostock Max V3. Printing with S3D totally fixes the voids problem. Now I have a bit too much stringiness but no globbing or zits. I'm going to let one of the wing sections run to completion to see how it turns out.