Lint and such on filament, wiper on UM2?

[aaronalai 2]

So while I am waiting for my UM2 to arrive I ordered some Colorfabb filament that arrived today. I was looking at the filament, because that's pretty much all I have at this point, and noticed some little black flecks in the interior of the spool and just a little on the filament itself (the filament is white), it's only held on by static cling; this is probably minuscule to all the cat hair and lint that stuck to the filament after I dropped it on the ground and a good bit of the filament unfurled while it rolled away.

Now there is a reasonable amount of non-plastic debris on the filament itself stuck there by static cling. Does the UM2 have a piece of foam or something to wipe off the exterior of the filament so it doesn't get trapped inside the bowden tube? I know I've seen something on thingiverse related to filament wiping, I don't imagine its a super big deal in the beginning but I can definitely see cat hair getting inside the bowed tube and eventually down into the extruder over time without some type of wipe.

 

[IRobertI 520]

There is no factory mounted wiping system, no. There are tons of different designs available to print and use but since I made my own I have to pimp that one of course http://www.thingiverse.com/thing:190118

 

[aaronalai 2]

Downloaded! Thanks, seriously you probably prevented a lot of problems for my cat an I in the future.

 

[mnis 11]

Look out for such a matching cake bell (cheese-cover) or a small bucket. The things are also transparent and flexible available. Cut a hole in the center and a wide slot in the outer region. Sliding the whole thing on the spool holder. We have three cats ...

[aaronalai 2]

Shoot, this one is a bit more difficult to figure out. I'm very interested to hear what has worked for you; my cat sheds much more than a normal cat. I'm also interested in reading that you use your spool holder. I've read in several threads that the stepper motors are not strong enough to unspool the filament thread while the spool is on the holder, too much friction? Perhaps you could send a photo, I'm sure there are many Ultimakers out there with cats.

 

[aaronalai 2]

You know I think I get it, a clear plastic bag over the printer with a hole cut out the top.

 

[aaronalai 2]

A picture would still be nice though, a person with three cats must have perfected the design.

 

[illuminarti 18]

I wouldn't say that the motors aren't strong enough... just that when good quality prints depend on very precisely controlling the movement of the filament into the printer, and there is no feedback loop to ensure that the requested amount was actually delivered, anything that you can do to make the system more predictable is a good thing.

 

[mnis 11]

Currently there is unfortunately only an idea.

Maybe I'll take a rectangular container for the entire back wall.

A round cover on the spool holder is easier to attach and is not rotated. It has to turn the filament role only.

I think all the problems with the filament transport relate to very high print speeds. Also incorrect combinations of print speed and extruder temperature quickly lead to problems. I print mostly with 35 to 75mm /s and have few problems. I use the Ultimaker filament at 205-210 degrees.

[aaronalai 2]

Thanks for the printing suggestions, mnis. I want to contain my UM2 somehow as well, I have a few ideas but if I actually make something I'll definitely ping you.

 

[illuminarti 18]

I think all the problems with the filament transport relate to very high print speeds. Also incorrect combinations of print speed and extruder temperature quickly lead to problems. I print mostly with 35 to 75mm /s and have few problems. I use the Ultimaker filament at 205-210 degrees.

 

While it's true that good quality prints depend on not exceeding a temperature-dependent maximum speed, it's also important to recognize that that speed is primarily a volumetric one (at least until you reach the limits of the mechanics, at several hundred mm/s linear speeds).

The limit is the volume of plastic per second extruded which depends not only on linear print speed but also the layer height that you are using (and, technically the extrusion width - which isn't always equal to the nozzle size, but probably should be).

It's also sadly not true that filament feed problems only occur at high speeds - especially on the UM2; I've had it happen on 50mm/s 0.1mm layer prints.

 

[aaronalai 2]

I wouldn't say that the motors aren't strong enough... just that when good quality prints depend on very precisely controlling the movement of the filament into the printer, and there is no feedback loop to ensure that the requested amount was actually delivered, anything that you can do to make the system more predictable is a good thing.

 

I have been seriously thinking about this a lot! Is something like a laser emitter receiver sensor not used in a feed back loop for a reason? Is it because the spool moves too slowly and the sensor drifts? Have you run across anyone ever trying to implement this feature?

 

[mnis 11]

@illuminarti

Aha, this is for me as a beginner very serious matter.

I have just found for me that it is apparently worse at higher speed. And with the settings I mentioned almost no problems occurred. The results suggested me I'm doing it right. The click was almost never hear. It is also nonsense ever want to print at high speed, if you do not yet understand the relationships. By that I mean mostly me :smile:

[nick-foley 5]

Robert's filament dust wiper is great. In terms of simplicity+functionality I think it's the best thing I've ever downloaded from Thingiverse.

 

[illuminarti 18]

mnis - I did some testing on the UM1 last year, regarding the relationship between requested and actual throughput. See here:

http://www.extrudable.me/2013/04/18/exploring-extrusion-variability-and-limits/

In recent comments on that post, Erik van der Zalm (lead developer on the Cura firmware) verifies my findings based on the use of such a sensor, and talks a bit about his plans to incorporate that into the firmware in future.

The UM2 faces similar challenges, although if anything the throughput is lower than the UM1. While the clicking of the extruder indicates an absolute failure to extrude, I would expect that you could still get under extrusion without experiencing clicking, similar to what I found in my UM1 tests, when trying to extrude too high of a volume per second.

 

[illuminarti 18]

I also just corrected the last sentence of my earlier post, to avoid confusion - of course problems CAN occur at high speeds - if you're already straining the capabilities of the extruder, anything that makes it harder is goign to cause problems; but my point was that even at what should be quite easy speeds for the printer, I've found that problems with getting the filament off the spool can cause print quality issues.

 

[mnis 11]

Probably you do not see Under-extrusion as inexperienced on the finished object. So if it is rare, it may not noticeable. Only with frequent Under-extrusion can also clearly be seen, right?

[illuminarti 18]

If it's consistent, it's hard to see. You can get 10% or 15% less plastic than intended, and not really see much of a difference in the finished print. Top surfaces might not close up quite as well, but in general the walls will probably still look just fine.

Where you really notice it is if it is excessive - if extrusion basically stops for a while (as happens when the UM2 extruder clicks back a bit, or the head empties during a long, leaky move), and/or if it happens for a relatively short period of time, in an obvious spot, where the rest of the print is fine.

For instance if there is a brief under-extrusion while printing internal infill, no one will ever know. But if there's one perimeter line on the front of the print that is under-extruded a bit, while the rest of the print is perfect, it will stand out - while the same amount of underextrusion on every line would barely be noticed.

 

[mnis 11]

I've found problems did with getting the filament off the spool can cause print quality issues.

 

May suffice a small curved groove between the extruder stepper and filament role. So a curved channel of PLA?

 

[illuminarti 18]

That's part of the solution, but it goes further - sometimes the loops on the roll can get partly crossed over one another - not knotted, but the 'top-most' loop of plastic can end up buried under earlier ones in the spool, and needs to get pulled past them again to reach the surface and feed freely into the printer.

My proposed solution is to have loose coils of filament hanging in space for the extruder to pull easily into the printer with minimal effort, combined with a powered spool motor that forcibly unwinds the filament at the same rate it is being fed into the printer, so as to keep a roughly constant-sized coil of plastic waiting to be fed in to the printer. (The spool wouldn't need to be unwound constantly - it would be sufficient to unwind a few cm at a time, one that amount had been fed in to the printer).

 

[mnis 11]

Phew...

Thank you for the effort that was very interesting. But for now it was enough and I have to digest it first.

Did you all have a beautiful time.

@ AaronAlai

Let me know if you have found a cover.

Markus

[IRobertI 520]

Robert's filament dust wiper is great. In terms of simplicity+functionality I think it's the best thing I've ever downloaded from Thingiverse.

 

Seeing as I just a few days ago sat and went through your design process of the UBIS clamp step by step trying to learn from you that means a lot, thanks (you also made me order a couple of those hotends... bastard... )

 

[aaronalai 2]

@ AaronAlai

Let me know if you have found a cover.

Markus

 

Definitely! I had a project a couple years ago using large pieces of very clear plastic sheets about 1/4 inch thick. I mistakenly ordered twice as much as I needed and the company wouldn't let me return the product. I found them last night and they are perfect for making an entire cover for the printer, I looked at the full dimensions in the user manual. I already have a design drawn up and am going to actually do the cutting and assembly when the printer arrives so I can make sure to include extra filament spools.

 

[gr5 2,167]

cat hair and dust can really ruin your day if it doesn't melt at 240C or cooler.

Regarding the spools getting stuck. 99.9% of the time it's fine. But if you are doing a 5 hours print with 1000 lines of height and one line is underextruded you get this annoying horizontal line in your print running all around every detail at that level.

Without measurements - I would say the extruder can pull/push 10 to 20 pounds of force. It takes about 1/4 pound of force to turn the filament normally. But sometimes something sticks a little. It's very rare. And might take a whole 2 pounds of force to get things moving again. Meanwhile that layer looks different from all the other layers.