1

Topic: Problem with layer adhesion (I think)

I've been trying to calibrate this Solidoodle 2 for a while now, and singe switching to an M3 threaded rod I got rid of my banding problem. However now other problems are more obvious, chief among them seems to be a problem with layer adhesion. See the picture below:
https://dl.dropboxusercontent.com/u/8410101/IMG_6763.JPG

I'm not sure if anybody else experiences this, or what might cause it. Could it be the filament I'm using? I set the temps from 180 - 210 and still the same effect so I'm not sure what's up. Should I try an even higher temp? What's the max Solidoodle temp?

Another interesting problem seems like the infil doesn't always reach the parimeter, or doesn't like to stick to it? Here's a picture:

https://dl.dropboxusercontent.com/u/8410101/IMG_6763.JPG


And lastly (sorry for the image dump) there seems to be something going on where the corners aren't nicely done.

https://dl.dropboxusercontent.com/u/8410101/IMG_6767.JPG

The sides seem like they were printed nice and straight, but the edges often have a bit of extra filament oozing out. Any tips would be appreciated smile

2

Re: Problem with layer adhesion (I think)

Maybe the extrusion isn't consistent.  Try extruding something like 50mm straight down a bunch of times and see if the thread is the same length every time.  Also turn off randomize starting points if it isn't already.  Is the temperature steady?  Does the hotend or extruder wiggle?

3

Re: Problem with layer adhesion (I think)

Do you have retract on? I tend to prefer the results with retract for when I'm using PLA, because PLA oozes more.

4

Re: Problem with layer adhesion (I think)

You are printing way too hot for that filament. back your temps off by 10C or so. That should help with the blobbing. What temps are your using? You can see the accessory mounting posts are charred. You might be dangerously close to a clogged hotend with those temps.

SD3 w/ mods:
Glass bed with QU-BD heat pad upgrade, threadless ballscrew w/ 8mm smooth rod, spectra line belt replacement, lawsy MK5 extruder, Lawsy replacement carriage, E3D hotend, Ramps 1.4 w/ reprap discount controller, DRV8825 drivers, 12v 30A PS, Acrylic case, Overkill Y-idlers, Filament alarm, Extruder fan + more.

5 (edited by gettingbored 2013-09-14 00:47:40)

Re: Problem with layer adhesion (I think)

IanJohnson wrote:

Maybe the extrusion isn't consistent.  Try extruding something like 50mm straight down a bunch of times and see if the thread is the same length every time.  Also turn off randomize starting points if it isn't already.  Is the temperature steady?  Does the hotend or extruder wiggle?

I'm not sure what is considered wiggle so here's a video of my extruder. There's definitely some play there.

I calibrated the extruder, as far as I can tell it pulls in exactly 100mm of filament when I tell it to pull in 100mm. I'll turn off randomizing starting points and see if that helps.

Tomek wrote:

Do you have retract on? I tend to prefer the results with retract for when I'm using PLA, because PLA oozes more.

Yes retract is on.
Retract length: 2.5
Speed: 50
Extra length on restart: .3
Minimum travel after retraction: 1
Wipe before retract: ON

2n2r5 wrote:

You are printing way too hot for that filament. back your temps off by 10C or so. That should help with the blobbing. What temps are your using? You can see the accessory mounting posts are charred. You might be dangerously close to a clogged hotend with those temps.

That makes sense, I was running it at 205 to start and 200 the rest of the layers, I'll give it a shot at... 190 for first layer and 180 the rest?

I usually have the metal enclosure on to keep things warm inside, just took it off for the quick video.

Any ideas on the little separations between the layers? I thought it wasn't hot enough, therefore not adhering together...

6

Re: Problem with layer adhesion (I think)

I get that from high humidity either in the filament (if it's been out in humid weather too long), or the air.

(I keep feeling like someone has to point out that there's a difference between adhesion and cohesion...)

7

Re: Problem with layer adhesion (I think)

Hunter Green wrote:

I get that from high humidity either in the filament (if it's been out in humid weather too long), or the air.

(I keep feeling like someone has to point out that there's a difference between adhesion and cohesion...)

Humm, adhesion is bonding two things together.
Cohesion is making two things into one single thing.
If I'm not mistaken.



As a little update, I lowered the temp to 190 on first layer and 180 on the rest. Then I started to print the same piece as in the original picture but there were obvious problems.

https://dl.dropboxusercontent.com/u/8410101/IMG_6772.JPG

The filament must have gotten caught at some point as the bottom few layers are off set from the rest.
Other than that though, it seems like it just isn't extruding enough material? If I turn the flow rate up any more I won't have accurate line widths. i.e. I've already calibrated for .42 extrusion width by following solidoodles instructions. (http://wiki.solidoodle.com/flow-rate)

I then tried to print a 50mm cylinder, which seemed to print fine but when I went to pull it off the board it split along one of the layers...
https://dl.dropboxusercontent.com/u/8410101/IMG_6776.JPG

Here are some more pictures of the cylinder.
https://dl.dropboxusercontent.com/u/8410101/IMG_6777.JPG

https://dl.dropboxusercontent.com/u/8410101/IMG_6786.JPG

The top of the cylinder was poorly done as you can see. I easly took my finger nail and pulled the top couple of layers off. In fact I could basically peel apart individual layers with my fingernails. I'm uploading some video to YouTube and I'll add it here in a few minutes.


This is a video of the print at the top of the post.


Here's a video of me pulling apart the layers of the cylinder.

8

Re: Problem with layer adhesion (I think)

Your first layer is too flat. It is pushing filament out the sides. It also looks like under-extrusion between layers.

OK. The first answer you neglected: Is you Extruder Steps/mm set correctly? By this, if you manually command the extruder to move 100mm, does 100mm of filament get moved? The SD gets shipped with a default value of 138, and is usuallly wrong. You need to measure 100mm on the filament, and then command 100mm movement and measure the difference. Then you can calculate what the steps/mm needs to be and save it to EEPROM using Repetier.

This is done first, before you can adjust your flow rate.

Second, level your bed. I only say this because it is required before the next step, and considering your extrusion in the videos, your level is probably getting knocked out every time your nozzle hits a high bump.

Third, fix your Z height screw so that the first layer is about .3 to .25 thick. If you squish too much, you will have problems.

Fourth, now you can check your flow rate with the single wall cube. If the first three calibrations are not done correctly, this calibration will not be right.

Fifth, the X and Y belt tension calibration will help with the corners.

Lastly, I wonder more about the filament you are using. It helps greatly to do all of these calibrations using really high quality filament you can trust. Especially since all of the calibrations do not need to be altered when changing filament types/brands except for flow rate. I save filament configurations based upon manufacturer.

Chuck Bittner is a quadriplegic gamer who is petitioning the major console developers to include internal button remapping in all console games. You can help.
Sign Chuck Bittners petition

9

Re: Problem with layer adhesion (I think)

IanJohnson wrote:

Maybe the extrusion isn't consistent.  Try extruding something like 50mm straight down a bunch of times and see if the thread is the same length every time.  Also turn off randomize starting points if it isn't already.  Is the temperature steady?  Does the hotend or extruder wiggle?


I tend to agree with you on the extrusion.I have everything calibrated for .3 layer height,extrusion,flow rate,etc.

I recently printed out the same MK5 at .3 layer height and 50% infill and ended up with some fine looking prints.

I thought I'd experiment with .1 layer height to see if I could improve the result,I started with the tension arm.The only thing I did to start with was go from .3 to .1 in slic3r and hit "print".

My results were less than perfect.The tension arm printed at .3 looked as good as you could want.The arm printed at .1 had the exact same blobs on the corners as gettingbored is getting.That leads me to believe it's an extrusion problem.

I wasn't about to re-calibrate for .1 when I already had a good set on MK5 prints at .3 so I stopped there.

I've also experimented with turning of "randomize starting points" on various prints.It might not make a difference when printing out a MK5 but if you do so on a taller print such as a vase,you will end up with what looks like a seam that runs the height of the print.

10

Re: Problem with layer adhesion (I think)

What is the extrusion width you are using?  Did you set it manually or let Slic3r do it automatically?  You should go thnner than the nozzle diameter, so if you have an older Solidoodle with a .35mm nozzle, don't use less that .35 as extrusion width, and if you have a newer one with the .4 nozzle, don't go thinner than .4.    If the extrusion width is narrow than the nozzle diameter, there is less control over where the plastic lands.

11 (edited by 2n2r5 2013-09-14 14:42:07)

Re: Problem with layer adhesion (I think)

@gettingbored: Do you have another filament you can try? Did you have the same layer seperation issues at 205C? 180 might be a bit on the low side. 190-195 always seem to be the sweet spot for me.


The picture that shows the small layer shift and the extrusion problems. -> Did you notice the extruder skipping? or see if the filament was having trouble feeding? That also might be from the temps being a little low.


on a side note: Looks like you don't have any banding issues! Nice job.

SD3 w/ mods:
Glass bed with QU-BD heat pad upgrade, threadless ballscrew w/ 8mm smooth rod, spectra line belt replacement, lawsy MK5 extruder, Lawsy replacement carriage, E3D hotend, Ramps 1.4 w/ reprap discount controller, DRV8825 drivers, 12v 30A PS, Acrylic case, Overkill Y-idlers, Filament alarm, Extruder fan + more.

12

Re: Problem with layer adhesion (I think)

gettingbored wrote:

I've been trying to calibrate this Solidoodle 2 for a while now, and singe switching to an M3 threaded rod I got rid of my banding problem.

Um, stupid question: What did you change your steps/mm for your Z axis to after changing the rod?

Chuck Bittner is a quadriplegic gamer who is petitioning the major console developers to include internal button remapping in all console games. You can help.
Sign Chuck Bittners petition

13

Re: Problem with layer adhesion (I think)

Hazer wrote:

Your first layer is too flat. It is pushing filament out the sides. It also looks like under-extrusion between layers.

OK. The first answer you neglected: Is you Extruder Steps/mm set correctly? By this, if you manually command the extruder to move 100mm, does 100mm of filament get moved? The SD gets shipped with a default value of 138, and is usuallly wrong. You need to measure 100mm on the filament, and then command 100mm movement and measure the difference. Then you can calculate what the steps/mm needs to be and save it to EEPROM using Repetier.

This is done first, before you can adjust your flow rate.

Second, level your bed. I only say this because it is required before the next step, and considering your extrusion in the videos, your level is probably getting knocked out every time your nozzle hits a high bump.

Third, fix your Z height screw so that the first layer is about .3 to .25 thick. If you squish too much, you will have problems.

Fourth, now you can check your flow rate with the single wall cube. If the first three calibrations are not done correctly, this calibration will not be right.

Fifth, the X and Y belt tension calibration will help with the corners.

Lastly, I wonder more about the filament you are using. It helps greatly to do all of these calibrations using really high quality filament you can trust. Especially since all of the calibrations do not need to be altered when changing filament types/brands except for flow rate. I save filament configurations based upon manufacturer.

Must have missed it, but I answered the extruder calibration question in post #5, I tell it to pull in 100mm and it pulls in exactly 100mm. I followed the instructions from Solidoodle.

Yes the first layer is pretty pressed into the bed, but it tends to not stick if I don't extrude that low. Maybe it's worth looking into a glass bed and some hair spray or something. I'm not exactly sure how the first layer could effect every other layer above it in the way my print is effected, but I'll try backing off on the first layer a bit.

After I calibrated my extruder, I did print the single walled cube to calibrate line thickness, but the only way I could get accurate line thickness was by bringing the flow rate down to .8 which basically resulted in the gaps appearing between layers.

IanJohnson wrote:

What is the extrusion width you are using?  Did you set it manually or let Slic3r do it automatically?  You should go thnner than the nozzle diameter, so if you have an older Solidoodle with a .35mm nozzle, don't use less that .35 as extrusion width, and if you have a newer one with the .4 nozzle, don't go thinner than .4.    If the extrusion width is narrow than the nozzle diameter, there is less control over where the plastic lands.

I have the line width set to .42 and I believe I have a .35mm nozzle. Is there a reliable way to tell?


2n2r5 wrote:

@gettingbored: Do you have another filament you can try? Did you have the same layer seperation issues at 205C? 180 might be a bit on the low side. 190-195 always seem to be the sweet spot for me.

The picture that shows the small layer shift and the extrusion problems. -> Did you notice the extruder skipping? or see if the filament was having trouble feeding? That also might be from the temps being a little low.

on a side note: Looks like you don't have any banding issues! Nice job.

Yeah there are still some separation issues with the higher temps, the pictures in the first post are with higher temps and you can definitely see those little gaps all over the place.

I tried to print some more with the lower temps and the extruder started skipping, which made me think the temps were too low at 180. I'll try going to 195 first layer and 190 for the rest to see how that works.

Do you have a recommended place to buy filament from, or one that you know ships consistent stuff? If I know a reliable distributor I'd like to get something that would be more "standard" so to speak, that way I can eliminate that variable.

Yeah getting rid of those banding issues was a huge relief, was probably the most annoying problem with the printer ever since nothing but swapping the threaded rod fixed it. Now I have 2.7 meters of M3 threaded rod I'll have to find an interesting use for.

14

Re: Problem with layer adhesion (I think)

It still seems like a partial clog or some kind of jamming.  Even though you get a consistent 100mm going in, measure the strand coming out a bunch of times and see if the lengths are consistent, or if there is some variance.  You can also run a finger along them and feel if they are smooth, or if there is any bulging along the way.  Try a couple of difference speeds since extruding into air isn't the same as printing.

You can also try cooling the PEEK with a fan and see if the problem is too much heat creeping too high up.

15

Re: Problem with layer adhesion (I think)

IanJohnson wrote:

It still seems like a partial clog or some kind of jamming.  Even though you get a consistent 100mm going in, measure the strand coming out a bunch of times and see if the lengths are consistent, or if there is some variance.  You can also run a finger along them and feel if they are smooth, or if there is any bulging along the way.  Try a couple of difference speeds since extruding into air isn't the same as printing.

You can also try cooling the PEEK with a fan and see if the problem is too much heat creeping too high up.

I did have a major problem with clogging a while back, and ended up taking apart the entire hot end and cleaning it out by soaking it in acetone and basically scraping it out. I thought I got it pretty much fully cleaned out though but maybe there's some remnant left in there that's causing problems.

Do you have any tips on extruding the single strands? My filament tends to curl up in a ball when it's extruded.

16

Re: Problem with layer adhesion (I think)

Did you take the barrel out of the PEEK when you took it apart?  If any gap gets left between the top of the barrel and the PTFE liner inside the PEEK it can cause problems if melted plastic is getting pushed into the gap.  Cooling the PEEK can help there by getting the filament to be more solid when it passes that area.

17

Re: Problem with layer adhesion (I think)

gettingbored wrote:

and I believe I have a .35mm nozzle.

That. Right there. You have .42 for width manually set (a good thing) and if you set your nozzle for .35 when its actually .4, you will get exactly what you have been seeing.

Also, I would still like to know what you set your Z steps/mm after switching your rod out. The wrong setting would mean the Z stepper could be dropping more than 0.3mm each layer (which would also possibly be the problem).

Chuck Bittner is a quadriplegic gamer who is petitioning the major console developers to include internal button remapping in all console games. You can help.
Sign Chuck Bittners petition

18 (edited by gettingbored 2013-09-15 20:22:10)

Re: Problem with layer adhesion (I think)

Hazer wrote:
gettingbored wrote:

and I believe I have a .35mm nozzle.

That. Right there. You have .42 for width manually set (a good thing) and if you set your nozzle for .35 when its actually .4, you will get exactly what you have been seeing.

Also, I would still like to know what you set your Z steps/mm after switching your rod out. The wrong setting would mean the Z stepper could be dropping more than 0.3mm each layer (which would also possibly be the problem).

I set the z-steps per/mm to 6400. Based on 3200 steps per rotation and a .5 pitch on the M3 screw.

Also, I believe my nozzle setting is set to .35, and my extrusion width to .42 but I'll have to double check tomorrow. Is there a way to calibrate the extrusion width, or is there a specific with determined by the nozzle diameter and the flow rate?

IanJohnson wrote:

Did you take the barrel out of the PEEK when you took it apart?  If any gap gets left between the top of the barrel and the PTFE liner inside the PEEK it can cause problems if melted plastic is getting pushed into the gap.  Cooling the PEEK can help there by getting the filament to be more solid when it passes that area.

I did take the barrel out of the PEEK when I took it apart but I twisted it back in pretty well, fairly sure I had contact with the PTFE liner inside. Are there any recommended setups for cooling the PEEK?

19

Re: Problem with layer adhesion (I think)

gettingbored wrote:

My filament tends to curl up in a ball when it's extruded.

Assuming the nozzle starts clean (i.e. there isn't a bit of leftover extrusion sticking to the nozzle or something like that) curling extrusion like you describe tends to indicate a clog in the nozzle. I went through fairly similar issues with my stock SD2 hot end until I finally gave up and bought an E3D to replace it, because I just couldn't print anything with my stock hot end anymore. Luckily yours doesn't seem to be in as bad of shape as mine, though.

One thing, looking at the difference between the MK5 print and the cylinder print, I think at least part of your issue has to do with retraction settings. Try printing the 0.5mm-thin-wall.stl from this calibration set: http://www.thingiverse.com/thing:5573/.

With a single-wall print like that you'll be able to see exactly where the layers change and, with that, see the effects of your retraction settings. If you end up with little gaps in the wall after each layer height change you know your retraction setting is a bit high.

20

Re: Problem with layer adhesion (I think)

The SD now ships with a 0.40 nozzle. You have an updated aluminum bed and heater on yours, so there is a good chance you may have a 0.40 nozzle. If you set the nozzle for 0.35 in slicer, you will have problems.

I have no idea how to confirm what nozzle size you have, it depends on when you bought it.

Chuck Bittner is a quadriplegic gamer who is petitioning the major console developers to include internal button remapping in all console games. You can help.
Sign Chuck Bittners petition

21

Re: Problem with layer adhesion (I think)

Hazer wrote:

The SD now ships with a 0.40 nozzle. You have an updated aluminum bed and heater on yours, so there is a good chance you may have a 0.40 nozzle. If you set the nozzle for 0.35 in slicer, you will have problems.

I have no idea how to confirm what nozzle size you have, it depends on when you bought it.

Is there a serial number or version number on the system somewhere that might correspond to the time it was made? Not sure if the time we got it in the office it was new or not.







Also,
Did anybody see the video of the extruder play? Does it seem normal to rock a little bit like that or is that more play than there should be?

22

Re: Problem with layer adhesion (I think)

gettingbored wrote:

Did anybody see the video of the extruder play? Does it seem normal to rock a little bit like that or is that more play than there should be?

Solidoodle gantries tend to have a fair bit of play in them, so rocking like that is pretty common... it doesn't look like the looseness is between the carriage and the rods, though, so that's good. You can try some combination of shoving a shim between the motor and the carriage and further tightening that cable tie to reduce the amount of play but it doesn't look significant enough to be the cause of all of your problems.

Rocking like that may give you a bad surface finish on prints, especially at higher speeds, but shouldn't be enough to cause what you're seeing.

23

Re: Problem with layer adhesion (I think)

beavertank wrote:
gettingbored wrote:

Did anybody see the video of the extruder play? Does it seem normal to rock a little bit like that or is that more play than there should be?

Solidoodle gantries tend to have a fair bit of play in them, so rocking like that is pretty common... it doesn't look like the looseness is between the carriage and the rods, though, so that's good. You can try some combination of shoving a shim between the motor and the carriage and further tightening that cable tie to reduce the amount of play but it doesn't look significant enough to be the cause of all of your problems.

Rocking like that may give you a bad surface finish on prints, especially at higher speeds, but shouldn't be enough to cause what you're seeing.

Actually the play does seem to come from between the carriage and the rods. Like there's a .1mm gap between those white plastic bushings(?) and the rod. Not sure how much of a problem that would cause, but I'm not sure how I would fix it. Is there a way to snug up the bushings?



Hazer wrote:

The SD now ships with a 0.40 nozzle. You have an updated aluminum bed and heater on yours, so there is a good chance you may have a 0.40 nozzle. If you set the nozzle for 0.35 in slicer, you will have problems.
I have no idea how to confirm what nozzle size you have, it depends on when you bought it.

I looked on the back of the machine and saw on the board it reads "Sanguinololu Rev 1.3a" but on the chip on the board it reads "Solidoodle Rev 1.3c" so I'm not sure which nozzle my printer would have... By default it was set to .35 but maybe it's a .4 nozzle. Either way my extrusion width is set to .42 so maybe I could just try changing it to .4 and see what happens...

24

Re: Problem with layer adhesion (I think)

John from Solidoodle said they started shipping .4 nozzles 6 months ago.  Even though .42 extrusion width is ok, if Slic3r thinks you have a .35 nozzle, it might not be extruding the right amount of plastic.  On the other hand if your single wall cubes are correct, that might not be the problem.  Either way set the nozzle diameter to .4 and see if that helps.

25

Re: Problem with layer adhesion (I think)

Wish there was some marker letting people know when they had .35 and .4 nozzles. Is the SD timestamped, or just that serial number on the back?

Regardless, the change in slicer will take one second to test.

If that is not the case (and like Ian said, if that was the problem, then your wall thickness would not be any good) I have one lat thing to offer: Double check your EEPROM settings in Repetier for steps/mm. There is something in the firmware that reloads defaults, so maybe your steps/mm on the Z got switched back? That would also take one second to rule it out as a possibility.

Chuck Bittner is a quadriplegic gamer who is petitioning the major console developers to include internal button remapping in all console games. You can help.
Sign Chuck Bittners petition