Lack of extrusion in certain part... of a part

[vincentp 0]

I wished I could have written a followup to my former topic but I can't find it on the new forum. My PM seem to have disappeared as well.

never mind. I initially got a lot of extrusion issues, with the help of a couple of members, I added active cooling of the alu plate + PEEK isolator, which removed the plug issue a lot but not totally. Still, it worked good enough to finally get reliable prints which have overall continuous extrusion (this required however to install the V2 extruder driver and the V3 knurled screw).

very happy...

BUT...

after printing filled parts (from 30 to 60% filled, or even 100%, 1mm top and bottom sides) with CURA 12.12A

I wanted to finally print MY own part for which I've been struggling for so long. Problem is that it STILL has the initial issue than before the extrusion problem (plugs, mainly).

Here's the part : a half sphere with a small shoulder at the bottom (requires support) and about 32 holes in it (the holes must be "printed", can't drill them after).

Here's the result of the print. I get pretty much the same results with or without retraction.

This is red PLA just acquired from ultimaker. PLA OD was measured properly, extruder step constant was adjusted (just a bit), I print at 213° and this works just perfect for all other parts I've printed.

The print starts perfect. There's only one hole on the path, which leads to an almost complete circle (holes are 6mm diameter). It goes well until it meets more holes.

Like another member told me, the holes have the extruder stopping (of course) and it's when there are a lot of holes that the issues are there. Print will catch up and start to be "okay" again further in the print, when it reaches layers where you have less extrusion interruptions.

I've just ordered the hot end V2 to get rid totally of the plug issue but I was wondering if my problem is something else. I don't have fiction anymore on the extruder which works very well.

I've noted that the support density has an effect on the lack of material of the sphere when it fails. It's like if the support is printed with a very small amount of material, the extrusion has a hard time coming back to the proper extrusion amount when it moves to printing the sphere. As it alternates all the time between support and object... well... you see the result.

For this print, I selected 50% of support material filling compared to 25% only before. There's less lack of material but it's clearly there.

1) are there some settings in cura to fight this ?

2) does somebody understand what the exact issue is ? Ideas for a fix ? (this is not belt tension or this sort of thing, all this has been corrected, and again other objects are printing nicely)

3) can somebody give it a try on his own printer (that works well) so that I can see if the problem can be reproduced on a said-to-work printer ??

here's the link to the STL file

http://www.plecterlabs.com/Media/Ultimaker/Sphere_top_trous-ORDRE-7.STL

thanks a bunch

 

[snowygrouch 1]

I cannot help you with CURA, but I am very surprised you are still having plugs even with an active cooled

hotend. Ive been printing for nearly 5 months on V1 hotend and cooling without a plug forming.

1) What temperatures are you printing at ?

2) Could be your retraction is set so that its pulling the filament too far back, which might also

explain the slow response. Really high retraction will also lead to plugs even if your hotend is

perfectly designed. Perhaps you should compare your retraction settings with some experienced

CURA people.

Can you post a picture of your hotend cooling setup?

Finally I would leave the outline print (where it does 3 or 4 circles outside the part), but would advice to ditch the raft. This wont help your printing problems but will be generally better !

Regards

C.

 

[vincentp 0]

thanks a lot for replying ! you're the one who actually helped me last time (thanks!!). The active cooling setup is yours, with your part (40mm fan), you directed me to your website where I got the STL.

I suspect the bowden tube junction with the PEEK was imperfect from the beginning. I could try to assemble it but I decided to move to the V2. That's a printer I use at work, we can afford that upgrade if, coupled with your active cooling, it makes sure there are no plug anymore.

To be clear, plugs haven't been the issue recently. They did happen but nothing like what I was getting in the past. As posted in post #1, I print in PLA at 213°. Lower and it really looks like the bond between layers is too weak. I believe that temp is ok for PLA, overall but I could be wrong (me = noob)

with the part above (half sphere + holes) :

Issue occurs with or without retraction. Retraction uses Cura defaults. 4mm, 40mm/s.

noted about the raft. I left it as initially I had issue with making sure the first layers would stick properly to the painter tape. Starting straight with the sphere is tricky as it's only a contour of 2mm thich and warping tended to have one extremity to lift up, reason why I tested with the raft and got some success.

can somebody give a try to the part and print it ?

 

[snowygrouch 1]

I can do, but if I print it - I would be using Kisslicer....so I dont know how much that would help you.

If I get a good result, other than looking at the pretty picture I dont think I could pass on any recommendations since you are using CURA.

Post it in the "please print this for me" section of the forum.

 

 

[vincentp 0]

thanks a lot. I'm new to the forums, I didn't know that section (which is awesome). Will post there.

 

[illuminarti 18]

What speed and layer thickness are you printing at?

 

[illuminarti 18]

And do I read it right that your wall thickness setting is 1mm?

 

[vincentp 0]

thanks !

printing at 50mm/s, wall is 0.8mm, layers thickness is 0.2mm. Bottom / top thickness is 1mm. Filament OD = 2.85mm (really round, measured with a digital caliper, I get 2.83 to 2.86mm), stock packing density in cura (1.00).

3 lines of skirt. Minimal time on layer, I tried 3, 5 and 10 seconds, no real change.

infill pattern : line. Overlap is 15%. Fan is at 100%. Bridge speed: 100%. Filling style : loops => perimeter => infill.

let me know if you need other info !

 

[illuminarti 18]

Settings don't look too bad... I don't have time to test print it at the moment, but I think that it is indeed probably related to the support structure leading to under pressure in the nozzle which doesn't correct in time. Especially as Cura/Skeinforge can do some odd things on layers where it finds overhangs/bridges - such as over the top of that first layer of holes.

Have you tried setting the support density to 100? Or, better yet, try Kisslicer, I feel like it does a saner job with support.

Also, for the warping problem, try swabbing fresh blue tape with isopropyl alcohol before printing. I never use a raft, and although I haven't tried it, I wouldn't expect a piece like this to lift off the bed, when printing on alcohol-treated tape.

 

[snowygrouch 1]

I tried it in KS last night, printed out "ok", but just some strings under the top dome face where the support

wasnt touching close enough. I think I had entered 1mm gap between support and part. Might need to make it

0.2 or something for this kind of thing.

However the result was a fair bit better that the photo above.

 

[vincentp 0]

thank you to both of you.

@Snowy, would you mind sending me the Gcode so that I can print it ? My UM has the LCD screen + SD card slot, so I can just take the GCODE you generated and see if it works better, before I start incriminating the slicer ?

if it's big, please send it via transferbigfiles (or eq) to contact@plecterlabs.com

thanks !

 

[Daid 306]

I've seen this happen quite often, and so far I could relate most cases to a badly folded fan duct where the airflow of the fan cools the nozzle tip (which causes extrusion problems)

 

[cora111 0]

Try printing at a higher hotend temp (10-15 °C) - even if that means you exceed the official temperature tolerance limit of your filament. That's not a real problem. But it gives much better flow and adhesion between layers. So give it a try.

 

[gr5 2,270]

The layers just above your lowest set of holes look like typical under extrusion. This can be caused by a feed mechanism that needs upgrading or tightening. It can be caused by too cool extrusion temperature. It can be caused by printing too fast (50mm/sec is definitely not too fast!). Temporary under extrusion can be caused by poor gcode where it is printing slowly and suddenly speeds up.

I love Daid's suggestion that maybe too much air on the extruder tip. You could test this by turning off the fan.

You seem to have stringing which might be fixed by lower temp but everything else looks like you would be better off with higher temp.

You never published your travel speed. Is it the default? I think it's 250mm by default? It's good to have a travel speed much higher than your print speed.

Your issues could also be caused by retraction issues.

You need to experiment more.

 

[vincentp 0]

Extruder drive is fine, now upgraded to the V2 with the bearing and spring. Prints are going very well for filled parts, even the fan brackers that had poles (printed with retraction enabled, all good).

Over temp seems to create plugs (even with the active cooling) but I'll five it a try, maybe it's flow issue, limited by low temp.

stringing in the print is due to no retraction at all, as I wanted to maximize the flow. I'm going to use Snowy's Gcode this morning to see if it's in the code or else.

thanks !

 

[vincentp 0]

So I gave a try to the provided Gcode. Issue remains the same, but it wasn't that bad compared to early prints. I tried at different temp, no change.

I still have plugs though :mad: I'm going to pull my hair with that. I disassembled the bowden tube and clip and horseshoe to reinstall it (after cutting a few mm to remove the clamp mark). It works for a while, but then stops. Removing the filament shows clearly a gap between the tube and the PEEK.

I suppose I have to wait for my extruder V2 parts and start over. When Printing the sphere, I could clearly see a moment when there's not enough extrusion (little material to print, normal) to get the filament going thru and it ends plugged. Always. That's at 215 and 220°. At 213° I had less problems but it's true the bond is less quality.

 

[illuminarti 18]

To take slight issue with one thing that George said above:

"[under extrusion] can be caused by printing too fast (50mm/sec is definitely not too fast!)."

I think that in this case it's probably true that 50mm/s isn't too fast... but there are times when it can be. Ultimately it's often not the linear speed per se that is the problem but the volume throughput. In my experience, a standard nozzle can start to run into difficulties if you push about 9 cubic mm per second of plastic through it. Back pressure increases, and that can lead to grinding and/or head blocks.

And in some cases you CAN do that at 50mm/s...

For instance, if you set Cura to a 0.6mm wall width, with a 0.4mm nozzle size, it will basically behave as if the nozzle width is 0.6mm. Couple that to a 0.3mm layer height - and at 50mm/s you're extruding 50 x 0.6 x 0.3 = 9mm^3/s...

This is why I always want to know about layer height and speed when people are having extruder problems... in my experience with the contemporary hot and cold end hardware, any plugging and grinding problems I get are pretty much always related to exceeding the volume limits (it's easy to do in KS, for instance, if you do infill every two or three layers, although KS's own flow rate limiter is a godsend for preventing it, once you get it tuned right).

 

[vincentp 0]

interesting about the flow limit and back pressure. I have the feeling that it could be the reason (?). I tried to print the ultimaker robot (scaled x2) and it stalled at 1/3 with a plug. As a result, I get indeed filament grinding and it's not pushing anymore.

What I mean is : could it be that the plug is collateral damage of back pressure ? I could see it this way : too much pressure pushes the melted plastic up / backward => a plug forms, slowly => after a while it's stuck, filament gets harder to push => finally the effort is too important and filament is ground / munched => nothing is extruded.

Could it be the fact that my "minimum time over a layer" is too short ? I tried 5 or 10 seconds. I had more issues at 3 seconds (of course). Not sure that that time works along with extruded volume, it's obviously something else but it might have some impact (?)

OR

can I my filament diameter / packing density be way off ? In other words, I extrude too much in quantity,

OR the volume issues is more about volume / time

 

[illuminarti 18]

Yes... plugs can form from the pressure... melted plastic under pressure hanging around... trying to force its way back up towards the bowden.

Having a longer minimum layer time is going to slow down the print, and so reduce any throughput-related issues... but is mostly intended to give your print longer to cool between layers, to help keep overhangs from drooping as soft plastic gets more and more hot plastic piled on top.

What layer height and nozzle width are you using in Cura? Also what filament size and packing density? Do you have an Ulticontroller?

If you want to send me the gcode file that goes with the picture you submitted at the start of the thread, I'd be happy to take a look - I have some tools set up to analyze the gcode.

 

[msurunner 2]

Something else to consider when getting plugs and directly related to back pressure would be if grinding is a result of a plug, or a plug the result of grinding. When you examine the bite surface for the hobbed bolt, it's not particularly large being that it is a straight bolt, but the teeth are quite sharp. When you are experiencing excessive pressure, it can exceed the capabilities of the bolt pushing the filament, whereas it is quite capable of cutting a path in the filament. Do this once or twice, maybe not an issue, but it certainly doesn't take long to create a grind spot. Then you have idle filament and your heat begins to flow up the hotend towards the PEEK. The fact that you are getting Bowden tube movement signifies to me that you do have excessive pressure. There are two ways to combat that and illuminati stated one, slow it down a bit. The other is contrary to common thought processes, but to up the heat. I have one color of my PLA filaments (not a UM one, but PLA nonetheless) that I have to run at 250 to print at 60 mm/s. As long as you are shoving filament through it, it shouldn't be major issue running a higher temp. You start to have issues in printing with thin layers and slowly with larger retractions.

SO, in summation, bump it up to 220/230 (temps I printed at with UM filament) and give it a go. What's the worst that could happen? You get a plug, like you are often getting, and you disconnect the bowden tube to remove it. Also, I'd be interested in a picture of your hotend, particularly one showing your aluminum plate...

 

[vincentp 0]

thanks a lot. I tried printing at 215 and 220 in the past. I also tried printing a x2 ulti robot at 40 mm/s only yesterday, filled at 30%. Extrusion stalled somewhere in the middle and I still don't know why, but options discussed above (back pressure, grinding spot, filament ground and extrusion stops) seem to be the possibility. I removed the filament this morning and there was no real plug, nothing that should prevent the filament from being extruded, so it's more like not being able to push the filament thru.

What I don't understand is that I'm just using the stock settings of cura for everything. About your question, my tests happen with layers of 0.2 mm and walls of 0.8 mm, nozzle of 0.4mm. I'm trying to stay with those until something works in a reliable way.

my filament (at the moment) is red PLA from UM. 2.85 mm average, properly measured. I left 1.00 for the packing density. Should I leave that value ? can I calibrate it ?

I'm currently very frustrated as I can't get any consistent results. Prints start well but won't finish most of the time. I believe the extruder drive is now reliable (upgraded from the latest parts from UM, including V3 knurled screw). I didn't overtight the spring, it's extruding and denting the filament very well over all.

Thanks !

my sphere GCODE from cura (printing at 213°, modify if you need)

https://dl.dropbox.com/u/2198333/Sphere_GCODE.zip

x2 ultimaker robot printed yesteday (215° I think). Cura 1212A as well

https://dl.dropbox.com/u/2198333/UltimakerRobot_support.zip

hot end picture

 

[vincentp 0]

double post. I started thinking about flow and over extrusion. Remembered my "adjusted" E step value then started thinking about my recent upgrade. Didn't recalibrate at that time (drive V2, knurled v3).

then started reading about people saying that new value is more like 833. Yeah, where I was supposed read to find that by myself ? product page ? nope, for sure. Mine was set to 869, so I changed it and printed again the robot at 215°.

still stalls. Now I'm looking in the settings of the ulti controller (sorry, forgot to mention I have it). There's a FLOW settings somewhere in the "tune" menu. Do I have to calibrate this or is it related to the filament settings and E steps and what I see (+176.64) is the result of the calculation by cura, of is this a setting of the UM firmware ?

ideas ? It's really like I'm discovering a new troll everyday with that machine :-|

 

[owen 19]

Have you tried without the top cooling fan? Worth a try in case it's cooling too much or the wrong spot.

 

[vincentp 0]

I tried both.

I was able to print a UM robot to the end, at 225°C. However :

- it keeps going way too fast on the top "horns", just like on other small details. Or it's too hot. Regardless, those small details aren't printed well (blobs / melted plastic)

temperature seems to improve the feed question but it's the wrong temp for printing.

any hint about adjusting the flow and how ?

thanks !

edit : I'm seing now that the flow rate is something in skeinforge but cura doesn't seem to allow access to that important setting, WTH ?

 

[illuminarti 18]

The steps per e setting that you are seeing in the ulticontroller is bug in the version of Marlin that you are using. Go grab the latest version of Cura (released yesterday) and install the stock firmware from that. That will fix the steps per e display, and also give you a better flow control option in percentages:

http://software.ultimaker.com

That will make it much easier to diagnose things.