MakerGear Forum

I noticed on my prints, that it seems there's a vertical line of very thin plastic hairs at some of the corners that connects from one side of the object to the other. I'm assuming that the extruder is leaving behind thin strands of plastic when it travels from one side to the other. My other printers, Makerbot Replicator 2Xs, don't have this issue. How do I resolve this issue with the Makergear M2? I'm going to clean them up manually, but I'd rather not have this mess to begin with because it seems like there's other extraneous material that it leaves behind that leaves the surface looking fairly imperfect after cleanup.

Are you talking about top surface marring, or open air threads? Top surface marring can be fixed by using a small amount of z-lift on retraction. Threads can be reduced by increasing retraction distance and/or lowering extrusion temperature. Fine threads are very easy to remove afterwards by giving them a quick pass with a lighter flame, they will burn rapidly and shrink to nothing.

I was talking about threading, it seems that it leaves a line of threads on the same corner as the extruder travels from one side to other. So I guess...

Lowering temperature.
Increasing retraction distance.

Are the solutions. My current retraction distance is 1.00mm on the M2. Does that seem a little low? (Coast is 2.8mm and Vertical lift is 1.00mm. 2.8mm coast seems to produce the best results for minimizing the seam/z-scar.).

My temperature on the pink hatchbox ABS is 225. That seems to produce the best results as far as the best compromise between bed stickiness and the least amount of "burnt plastic". Too low, doesn't stick well to the bed, too high, burns too easily leaving black marks on the plastic.

In an entirely different matter I previously solved, solving the top surface marring also needed me to uncheck the "Only retract when crossing open spaces." which seems like the default setting.

You can try increasing the retract distance. Your temperature is already quite low for ABS. One other thing you can use is Wipe, which will retrace a segment already laid down before travel, to get rid of any stringing.

On the other hand, your coast and vertical lift are quite extreme. Coast cuts off extrusion that many mm before the end of a segment, which means you should be seeing nearly 3mm gaps everywhere. A few tenths of a millimeter is more normal. And only enough vertical lift is required to clear any over-extrusion or warp, generally a single layer height or two should be sufficient.

Definitely try Wipe. Also retract during wipe under Advanced.

In an entirely different matter I previously solved, solving the top surface marring also needed me to uncheck the "Only retract when crossing open spaces." which seems like the default setting.

Yes, I should have mentioned that.

Interesting, will try the wipe. I noticed that all of objects that have these threads were all part of multi-piece prints (4 pieces), and a multi-piece print seems to have different travel patterns than when printed individually. I wonder if printing them one at a time will eliminate this issue?

I tried replicating this issue on a "small test object", but I can't seem to get the extruder to travel on something small. Wait, lemme try shrinking one of these particularly problematic guys down.

Nope, seems to be that small things don't travel, but large ones do. Then again, my large thing had infill going up. Maybe I need infill to get my extruder to travel?

Creating a new test model. Will take about 10 minutes.

Just got it. Yep, confirmed my hypothesis. Something small with infill will travel alot. Infill = travel.

Time to test these suggestions. First, let's create the oh-so-hairy control and then experiment!

The control wasn't that hairy. It seems like I need inward-pointing corners to invoke lots of "long distance travel" and thus threads; it doesn't seem like outward pointing corners do much of anything. Creating a new control w/ lots of inward pointing corners.

Did it and, yes, lots of long distance travel. Let's see the hairs! (Model will be printed in 25 mins or so.)

Printed it out and... it did a lot of "long distance" travel and though I can see hairs, they aren't nearly as prominent as on the larger models. I'm guessing that the larger models have a lot more printing before traveling long-distances, and thus there's more 'momentum' in the filament it extrudes, and thus it's more 'leaky' than long distance travels on smaller models.

Well, I guess I'm forced to using larger test models. I don't really like my testing iterations to last an hour or so. Takes much longer to resolve problems that way.

My small test model was 1.5" long. The original with a noticeable 'thread'/hair problem was 3" long. I'm going to try 2.3" and see if I can still get the same affect.

Just printed the 2.3" model. Apparently not big enough to replicate the problem (Has hairs, but few of them and barely visible), now trying the full 3". It's a 30 min print, not too bad. Not like my ideal of 7-minute prints, but it'll work.

Just finished. Weird, it doesn't seem to be showing up again. The only difference I notice is that the original had a slightly thicker wall, which seems to enable greater lengths of infill before the long distance travel. I guess more infill laid down right before traveling = more 'filament momentum'/pressure = greater ooze -> threading/hairs. Will try the unmodified original, which is a 40 minute print. Will be weird if this issue spontaneously resolved itself. I'm thinking why my makerbots didn't have this threads/hairiness issue while my m2 did was that my makerbots have a .4mm nozzle while my m2 has a .35 mm nozzle, meaning that the same model prints without the infill on the makerbot, but prints with substantial infill on the m2, thanks to the extra .1mm difference on the interior between the two .35mm walls.

So, to generate this phenomenon, it appears there needs to be long continuous lines of infill laid before the long distance travel.

Well, shoot. I have an idea for the ideal test model, now. lol

Something is making me think the extruder naturally overheats after it's been used for too long. It's not that the temp sensor's area overheats, per se, it's that more heat gathers in the extruder nozzle area over time so the nozzle is getting hotter (or chamber before the nozzle), even though the temp sensor's temp remains stable. I'm noticing on a few of my hairy prints, that the hair doesn't start until the 7th layer or so, suggestive of heat/temperature buildup in the nozzle with use that isn't compensated for by the temp sensor or S3D software. I.e., perhaps the temp limit should be decreased after the 7th layer? From 225 initially, to 220 or 215 after 7 layers?

My print is now on the 10th/11th layer, and I'm starting to see a few hairs. (I'm starting to think I didn't see any hairs on my smaller test models because it didn't have enough time to overheat the nozzle. The hairs didn't start until roughly 22 minutes of continuous printing.)

Well, interesting, S3D allows the setting of temperature setpoints for I'm guessing this reason. I'll start off with 225, and then 215 after the 7th layer to see if it reduces the hair.

I'm going to guess that the reason why my M2 seems to have this issue is because of the greater disconnect between its temp sensor and the nozzle temperature than with my makerbot (My makerbot runs at 220 degrees because the temp sensor is closer to the nozzle.). However, it seems like it can be compensated with temp setpoints - Will test and report back.

Reduced the temp to 205 after layer 7. Seems that it's reduced the amount of plastic during threading (Making thinner threads), but it hasn't eliminated it. I've watched it a few times during threading, and it appears that /right/ before it travels, it lets out a little blob that then gets stretched across the gap it travels across, creating the thread. This deposited blob than gets deposited on the other side, creating those tumor-like curls on the surface of the object. How to eliminate the blob that gets emitted right before it travels? Will wipe suffice?

Okay, tried wipe with 5mm wipe and checked the box "perform wipe during retraction". The threads are still there, though they are a /little/ better (Almost as good as 205 degrees at layer 7).

I noticed that there was more substantial blobs at 205 than 225, leading to tumors, so maybe the low temperature is causing blobs somehow? I'd try upping to the temp to 235 after layer 4 to see how that works, but tomorrow.

Here's the two points, in my test, that seem to produce, by far, the most noticeable threading.


Interesting that one is at the beginning of the layer, and the other appears to be be coming in and then immediately taking off. I haven't seen the G-code, but I'd imagine that both of these points involving "unretracting" and then immediately "retracting" right before lifting off - the extruder unretracts right at the start of the layer, and it also unretracts as soon as it lands right before taking off again. Maybe the extruder motor can't work fast enough to retract effectively under those circumstances? (Or perhaps the nozzle design doesn't allow for effective retraction under those circumstances? Unretracting at 40mm/s and then immediately retracting at 40mm/s before traveling.)

It's clear it's pushing out 1mm of filament during "unretraction" immediately before retracting 1mm of filament and traveling. This 1mm "unretraction" appears responsible for the blob I'm seeing that gets dragged across creating the prominent thread. All the other travel points lack the "1mm unretraction", going straight from laying down infill to retracting 1mm and then traveling across. All the other travel points also appear to coast 2.8mm right before retracting 1mm and traveling, whereas these trouble points don't have coasting present right before retracting 1mm and traveling.

After thinking about it, I'm surmising that the 1mm of filament has the forward momentum from the 40mm/s unretraction that can't be effectively reversed during the 40mm/s retraction, so it ends up creating a blob that gets threaded across. One way to reduce the momentum, reduce retraction speed - less speed - less momentum. I'll try 10mm/s. (My makerbot has a retraction speed of 20mm/s, so that's noteworthy.)

Edit: I'm starting to think the default retraction speed of 40mm/s might be too high. The S3D software does say "Use the highest speed your extruder motor can support", and it's possible my extruder motor can't truly support 40mm/s.

Well, it printed. The hairs were a bit thinner but they were still present. I'm going to try a retraction speed of 20mm/s in addition to a restart length of -.2mm. If it retracts -1mm and unretracts .8 mm, then maybe a "blob" won't be pushed out right before traveling? Let's try.

Well, it seems that it's still leaving hairs BUT the hairs are definitely a lot less substantial (More wispy), so maybe a restart of -.3mm would resolve this? Also, the hairs are so thin, they aren't even making it to the other side. The hairs spawning from the travel point in the second image no longer exist at this point.

I'm starting to see the print with the -.3mm restart. It appears to be doing the trick, not seeing a sign of hairs anywhere. Time will tell if this extends to the rest of the print that's yet unprinted. (And here I was thinking based on other posts that something like -.01mm or -.1mm restart was sufficient, lol.)

Based on this experience, it seems that the difference between the makergear M2 and the makerbot is that the M2 has a "leaky nozzle"(Or maybe the extruder motor isn't that precise?), but it can be compensated with a high enough restart length.

Edit: Just got done printing at 20mm/s retraction speed and -.3mm restart. Not one hair anywhere and no visible imperfections anywhere, yeehah! (I've disabled wipe)

Now I'm trying to use the -.3mm for my multi-piece print and the strands are still there. They appear thinner than I remember, however, so it seems like it's partially effective. I upped it to -.4mm and there was still a little there, so I'm now trying out -.5mm. I'm hoping I can resolve this - it appears that the hairs are worse on some models and better on others. Not sure what's happening on those models, whether it's doing a triple jump instead of a double jump in my original test case, or what. It seems like the jumps between pieces are the worst in terms of hairiness, for some reason.

Edit: It looks like -.5mm restart is doing the trick for all the hairs/threads but one, which is now very thin.

2nd edit: Just got done with another multipiece print. Looks like the hairiness is back to its regular hairy self even with -.5mm restart. Have no idea what's going on, will try limiting the retraction speed to 5mm/s as the URL below suggests.

http://www.makergear.com/products/m-series-3d-printers

Extruder extrusion speed (manual): 300mm/min. Actual speed varies based on the object print settings

Wait, what? Does that mean the filament retraction speed should be 5mm/sec (Equal to 300mm/min) instead of S3D's default of 40mm/sec? Maybe that's why the M2 is messing up with S3D's settings and creating all these hairs whereas my other printer doesn't.

Definitely going to try out 5mm/sec on the next run and eliminate restart.

Also, puzzlingly...

We also recommend using Creator with your M2 though the open-source RepRap tool chain will also work.

What's Creator? Google tells me nothing.

Simplify3D used to be called Simplify3D Creator.

Retraction speed should be as high as possible, to relieve nozzle pressure quickly. Or so I've always been told. That extrusion speed quoted is probably sustained extrusion speed, and not an instantaneous limit.

Your model with the trouble spots is getting that dotting behavior because there's a tiny empty spot that it feels needs to be filled. Try using more than one perimeter.

Also, I highly recommend trying burning off the hairs with a lighter....

jsc wrote:Your model with the trouble spots is getting that dotting behavior because there's a tiny empty spot that it feels needs to be filled. Try using more than one perimeter.

Also, I highly recommend trying burning off the hairs with a lighter....

Oh, thanks for the insight! The "tiny area" isn't visible on the S3D screen (I'm guessing it doesn't draw tiny deposits of filament?), but I can see what's happening. Anyway, I've found a solution though it's a bit of band-aid more than anything. I changed the extrusion width of my M2 to match my other printer's nozzle, which is .4mm, which had the effect of reducing the infill channel's width and presumably helps eliminate those "tiny empty areas in the corners" (My previous testing showed that if there wasn't /enough/ infill, it doesn't create hairs...) and, yep, it's working. No hairs and tiny little tumors, now. This only works because my models are specifically designed for my other printer (The wall thickness is designed to match the nozzle's width so that it's roughly two filament lines thick.).

Doesn't really solve the situation in the case where the walls are just a /little/ thicker, but that shouldn't be a problem for me since I control the wall width for the most part.

So, situation that causes hairs ->

Infill channel is wide enough -> Tiny empty areas in the corners are created -> Extruder deposits a tiny bit of filament, but instead of getting laid down on the model, it creates a blob that then... -> Extruder retracts and drags the blob across model creating hair/threads and tiny tumors.

Eliminate the tiny empty areas by reducing the infill channel width (By increasing extrusion width) resolved this for me. Be nice, however, if there was a more direct way to control this in S3D - as in, "Don't lay down anything below .9 mm^2."; i.e., don't try to uselessly fill the tiny empty areas which you don't fill, anyway.

Time to mark this problem as...

100% understood.
50% solved.