[Solved] Surface Quality Issue with E3D Cyclops

[gestalt73]

So, I've been looking for other Cyclops owners who have posted pics of their prints, and it looks to me like everyone is having the same issue.

Depending on lighting and focus, it's possible to make the print look better or worse. I'm calling out the ones that are lit in a way that you can make out the surface finish.

And please don't misunderstand my intent, printing with two colors/materials out of a single nozzle is really neat, it's just that the print quality isn't what you can get out of a simple V6.

Here's the album from E3D showing detail shots of their prints
http://e3d-online.com/index.php?route=e ... blog_id=25

The surface finish is clearly shown in almost every shot:
http://files.e3d-online.com/Cyclops/Pho ... 212301.png
[img]http://files.e3d-online.com/Cyclops/Pho ... 212301.png[/img]

Here is an album from 3dKreashunz
http://imgur.com/a/SIUv0

You can see the irregular finish in the following shots
http://i.imgur.com/sAo7vAn.jpg
[img]http://i.imgur.com/sAo7vAn.jpg[/img]

http://i.imgur.com/ruUp3Za.jpg
[img]http://i.imgur.com/ruUp3Za.jpg[/img]


And here is an album from immersedn3d
http://imgur.com/a/WI9pj

You can see the irregular finish in these shots
http://i.imgur.com/abMaAMz.jpg
[img]http://i.imgur.com/abMaAMz.jpg[/img]

http://i.imgur.com/bFfxh2D.jpg
[img]http://i.imgur.com/bFfxh2D.jpg[/img]

[JFettig]

look at that close-up, it looks more like something is flopping around, something got loose while changing the hot ends maybe.

Your heaterblock is as close as mine were and the platform definitely warped, the artifacts I had weren't the same though.

Build yourself a mount that drops the hot end below the platform, way way far away - if for nothing else, just testing. I hated giving up some Z but I then realized I have never printed anything to the height I still had(By 1 inch).

[Nylocke]

Those pictures from E3D are stated as the first cyclops prints ever, and it was done on one of their "worst" printers. You seem really eager to jump to the conclusion that the Cyclops is the culprit, and even if it is its still a Beta product, there are going to be design kinks.

I think KAS and Jon are right, or at least onto something. You likely get extrusion issues that pronounced, it seems like it could very easily be a mechanical issue.

I would say something about the other prints, but I can't view Imgur links at school.

[gestalt73]

Hey KAS,
You do have a sharp eye.
With the V6 print, every defect you've called out is something that machine always does. Even the line in the base, which is a type of defect I see when the density of the print changes for some layers (that's where the top starts to render solid)

With the Cyclops, the striations are in different locations, and a different angles for every print. I'm looking at them side by side, and there are no consistent patterns between them.

I'm lighting the parts specifically to call out the surface finish. The benchmark part the isn't perfect, but serves to show the differences.

I'm not sure what printing a 0.4mm high 75mm diameter disk will tell us, but I might be able to put it on the printer later on this evening.

Hey JFettig,
Thank you for your help.
Unfortunately nothing is flopping around on the Cyclops hot end. All balls are tight and secure, and the platform was reinforced with bracing about half way through the day.
Also when I pulled the hot end apart, the effector platform appeared flat and uniform.

I'll rework the bracket today to mount upside down, or adapt your mount to my effector platform. I should be able to run the test either tonight or tomorrow.
I'm not really worried about lozing Z, I never print anything very high, but I've had my concerns in the past about how trig will exaggerate errors in nozzle position from delta calcs resulting in the slightest tilt to the platform. It's worth a shot though, at least to rule out platform warping.

Hey Nylocke,
Thank you for your input.
I'm definitely not "eager" to jump to any conclusions, but I've already been messing with this for about a week. I started this thread *after* I was out of ideas.

I won't bore you with everything I've tried changing, but nothing seems to have any impact on the print.

I'll rework the effector platform to mount the Cyclops under the effector, and report back when I complete another test print.

Thanks everyone for your help!

[geneb]

So you've covered the bottom face of the heater block with silicone tape? That should have eliminated any radiated heat coming from the heater block. If all you did was wrap it around the perimeter of the heater block, that ain't gonna work.

The big square you printed does make me think it's heat related though.

g.

[JFettig]

Belts and cheapskates right too?

[gestalt73]

Hey geneb,
It's a really good suggestion.
Yeah, I first just wrapped the sides of the heaterblock and started the print before I realized that you were thinking about downward heat radiating into the part, and not the heat radiating onto the effector platform.
So I ran a second test both with the sides and bottom of the heater block wrapped. There isn't alot of space on the bottom with that narrow nozzle, so I was only able to get two layers of wrap across the bottom.
Both tests turned out the same. Plus for that part I have all 3 parts fans running, so there is some airflow underneath the effector plate anyways. The benchmark part was also printed with all 3 parts fans running.

I've really been studying the big square, and I'm seeing the same surface variations but because they're not so tightly packed, they're weren't quite as obvious on initial inspection.
I've pretty much mastered the "overlit shot" technique, here's a new shot of the square showing the surface variations. The vertical bands are a known issue, so I don't worry about them.
http://i.imgur.com/PKOAa74.jpg
[img]http://i.imgur.com/PKOAa74.jpg[/img]

While I don't have any shots to share, on the single walled calibration squares I've printed, you see a similar pattern. What's interesting is that if it's a "valley" on one side, it's also a "valley" on the other side. It looks kinda like the plastic isn't being extruded evenly.

Hey JFettig,
That's a good point.
Yeah, the belts and cheapskates are looking good, with just the right amount of tension. I actually replaced the belts and cleaned all the internal and external pulleys late last year (that was fun). I'm seeing good tension on the belts, and looks like good tension on the cheapskates as well. Also the benchmark print was printed just before and after the Cyclops prints, so any obvious Rostock issues would have shown up in both prints. I'm only seeing sloppy prints from the Cyclops.

[Nylocke]

http://i.imgur.com/bFfxh2D.jpg

That one you posted is really interesting. The cone on the left, with the solid color switches, printed as well as I could print it with my Kraken or anyone else could with a V6 in mono color, while the one with the twisted interleave of sorts has the surface irregularities. If it were a Cyclops in general problem, wouldn't it manifest itself in both?

[gestalt73]

Hey Nylocke,

Yeah, on that one it's not as obvious because of the lighting, but I think I can still see layer sloppiness on the left cone. It's not as pronounced as the right cone, but it's definitely there.
I think I found a way to contact him through thingiverse, I'm going to message him now and see if he has any input.
If I light up that large square I printed just the right way, it looks almost as perfect, so I'm not sure if that's a great example or not.

I don't think anyone is intentionally misrepresenting the quality of their prints, but after getting such great prints out of my V6, I was kinda hoping to get comparable prints out of the Cyclops.
It's just easier to see in all my shots because I'm intentionally overlighting the pics from a high angle to call out the surface detail.

I'm not sure if you'd see this issue on a Kraken or Chimera, because they both have a straight shot through the heater block and through the nozzle.
In the Cyclops after exiting the heat break, the molten filament makes a 90 degree turn to the center of the block where it makes another 90 degree turn to flow out through a very short nozzle. The nozzle isn't a V5 or V6 nozzle.

[Nylocke]

I've seen the drawings.

I got to thinking over lunch. What extruder design are you using? It could be running out of torque or building and releasing pressure unevenly if it's a direct drive.

[gestalt73]

Hey Nylocke,

I'm running Direct Drive EZStruders for both filaments.

That is actually a really interesting hunch.

I have all the hardware for Greg's Wade Extruders for my second printer, just never got around to printing out the rest of the parts and assembling.
I should be able to complete a Wade geared extruder this week and give it a shot.

[Nylocke]

I would be willing to bet that this might be your issue. you mentioned it taking extra force to extrude, the EZstruder isn't the greatest with torque.

[TFMike]

Hey Nylocke,

I'm running Direct Drive EZStruders for both filaments.

Direct Drive? Didn't I see a bowden tube? What am I missing here?

[Nylocke]

Hey Nylocke,

I'm running Direct Drive EZStruders for both filaments.

Direct Drive? Didn't I see a bowden tube? What am I missing here?

there are 2 types of direct drive. Direct drive in the sense of direct vs geared extruders, and direct as in direct vs bowden. Here we are referring to direct vs geared.

[TFMike]

Is there a link to a place where I can read more on that?

[gestalt73]

Oh crap, must have used the wrong term. I meant that I'm using the non-geared ezstruder extruders.

[Tonkabot]

Okay, I received my Cyclops, and then I realized that I'll need another extruder.

Now I hear tell that the ezstruder may not be the only good choice.

Along those lines I bought a really cool stepper motor with a 5.2:1 planetary gearbox, that I think would make a decent extruder motor. (it looks cool and I have other projects I could use it on)

What else should I be getting to make this new extruder?

[gestalt73]

Hey Tonkabot,

Hang tight for a few hours, I'm running a test with a Greg's Wade Geared Extruder. Before you make a purchase, let's see if the extruder is part of the problem or not.
I should have something in a few hours. I'm running a new V6 benchmark print with the Wade, then I'll swap out the V6 for the Cyclops.

[gestalt73]

Holy Smokes Nylocke, I think you nailed it!

Swapping out the non-geared extruder for the geared extruder seems to have done the trick.

Unfortunately my benchmark part on the V6 is far from perfect. It's actually kinda crappy.
(I have a fair amount of work to ahead of me to tune for the geared extruder)

But... Comparing the V6 benchmark print (left), to the Cyclops print (center), and the original Cyclops print (right)

You can see that the quality is comparable.
(click on the link to see the unresized version)

There are a few more surface defects on the Cyclops print, but they have a similar character to the defects on the new V6 "benchmark print" and are very minor compared to before.

I'll spend some more time this evening to get the settings right for the new extruder, and I'll post some updated shots soon.

Thanks!

http://i.imgur.com/tsgLxIG.jpg
[img]http://i.imgur.com/tsgLxIG.jpg[/img]

[Holy1]

I am following this thread with great interest. I have been holding off buying a cyclops to see if they are working out for people. Thanks for effort gestalt73!

Along those lines I bought a really cool stepper motor with a 5.2:1 planetary gearbox, that I think would make a decent extruder motor.

@ Tonkabot. I am looking for a gearbox stepper. Do you have a link to the one you bought?

[JFettig]

I wonder if the issue isn't that the plastic isn't melted enough by the time it takes a sharp turn. It then takes huge forces to get the plastic around that corner.

You can make a nema 17 adapter for the geared stepper and use an 8mm bore drive gear with the ezstruder. I have a couple gear motors for another project that I contemplated doing using on my v2 for a test.
http://www.tridprinting.com/Mechanical- ... Drive-Gear

http://www.tridprinting.com/Electronics/

[Nylocke]

I'm glad your issues seem to be resolved. I'm still determining a good substitute for the EZStruder, I'm actually thinking of making a mix of a greg's and my EZFlex. Probably will aim to allow people with greg's to easily adapt to this design and those with EZstruders as well, so 2 versions, hopefully I can design it so they have a common body.

[bvandiepenbos]

Certainly the cyclops is heavier than the v6... could the added weight cause mechanical variances in effector platform movement?

Are the mag-ball joints coming apart ever so slightly?

Another thing, the bowden tubes are off center, and bowen tube exerts some force, could that be rocking the platform?

[Glacian22]

Hey Nylocke, I finally posted my planetary geared variant of your EZFlex! I am super pleased with it, the thing has been absolutely bullet proof so far. http://forum.seemecnc.com/viewtopic.php?f=36&t=8063

[geneb]

What I wanna know is this - when you're done testing, what are you going to do with 300 fan shrouds?

g.