Why have the tops of my prints started to slump ?

I've been running my MendleMax for a several months, but suddenly the tops of prints have started to become concave

I have printed some 20mm test cubes and the top is very concave and if I stop the print half way though I can clearly see that the honeycomb structure is falling in at the middle.

I've double checked the heat bed temperature using 2 different thermometers, I've reloaded the firmware in case I'd messed something up, as I calibrated the extruder steps to increase flow a bit (a week ago)

I've tried a different slicer (Cura instead of Slic3r), but I get approximately the same result.

Does this sound like a bed temperature issue, or possibly a flow rate issue.

I've not unscrewed the extruder nozzle (ever), so I'm wondering if the issue is that contamination has got into the hot end, from the wooden cabinet I built for the printer (albeit its painted inside and outside, particularly as I feed the filament in through a whole in the cabinet, which is not painted, i.e the hole was drilled after I painted the cabinet, so the filament comes into contact with MDF (perhaps its getting particulates stuck to it over a period of time)

Or perhaps its something else.

Should a try a different batch of filament, I'm using black recently but prior to that I was using white.

Help ;-)

Edit. Attached a photo. Sorry its partially out of focus, but I think you can see the honeycomb structure seems to be stretched near the edge and tight in the middle where its slumped down

Edited 1 time(s). Last edit at 11/11/2013 05:13AM by rogerclark.

i had the same problem with one of the batches of black abs i got...... the manufacturer had added something to it to make it flow more easily but because of that it would sag a lot...

try reducing the temperature and also reduce bridge flow rate....

Thanks ekaggrat

What temperature worked for you?

I have taken the hot end apart to clean it, and put a tube around the filament entry point into the cabinet, as I noticed some fluff in the extruder, but not as far as I can tell in the hot end.

I will reassemble it tomorrow and try a known batch of working blue abs to confirm its the black that has the problem, and if the blue works ok, I will try the black at a lower temperature

Thanks

Roger

Are you trying to fill every other layer? If so your flow rate might be a bit low. I'd bet on things being to hot for your matterial. I'd drop the heated bed temp 10C and try again. If that does not fix it, drop the hot end in 5C steps untill either it's fixed or the hot end jams.

I doubt it's crud that's doing it. Crud tends to plug up the hole and stop the extruder. I can send you some ABS with a bunch of crud in it if you want to see what a jam is like . If you enjoy unplugging a clogged sink drain, pulling globs of black crud back out of a hot end will be every bit as entertaining.

Hi uncle-Bob

Every layer is filled, so it looks like this is probably a material issue.

I took the nozzle off the hot end, and have got it 99% clean and can see right though the .4mm hole at the end, with no sign of anything clogging it up.
I used a soldering iron to heat the nozzle and make the abs liquid.

I will reassemble the hot end and try another reel of abs that I thought was ok

Thanks

Roger

Gee wiz, sorting through random rolls of ABS to find one that works. Why does that sound *very* familiar ? .... back to sorting ....

umm

I bought from who I thought was a reputable supplier, here in Australian, i.e not from China etc, but I've not established yet whether this is a filament issue (though it seems likely)

If so I'll contact the supplier to let them know there are issues.

I won't have time to test this until later today, but I'll post when I get to the bottom of the issue and also take some better pictures, as I think it would be handy if more causes for concave and slumping were listed on the reprap wiki, as it only mentions flow rate and temperature of the bed being causes of this.

Well I had several issues. One being that the bottom end of the heated end had come a bit unscrewed. After putting everything back together and re-leveling the bed I'm printing again with new filament. Some day I may go back and try the other stuff again.

Thanks guys,

It looks like the issue is the black ABS filament not the printer.

I've now re-assembled the printer and printed with my white ABS, and my 20mm test cube looks fine.

So the other changes I made, including reloading the firmware, replacing the heat bed with my old one, and cleaning the nozzle were a complete waste of time.
Well, not a complete waste, I found that the thermistor on my new heat bed, actually had a dry joint as the wire fell of as I was removing the kapton tape from the back of the PCB.
And I've added a small piece of plastic tube to act as a grommet where the filament goes into the cabinet in which the printer is housed.

I still need to upload the firmware again, as I'd been calibrating flow rates last week, as based on filament feed length, the stock firmware that came with my kit MendelMax (from Blomker.com) appears to have a feed rate about 5% low.
I have a E_STEPS value, so I'll need to re-instate it.

But even without getting the calibration spot on, its printing fine as long as I don't use my black ABS.

I've emailed the company who supplied it, asking for revised settings. (I'm not sure if they have this data, but its worth asking as they are local i.e not in China)

Thanks again

Roger

You can't simply use the same settings for different types of plastic. If the black ABS is less viscous it will have lower die swell. That means for a given nozzle and layer height, etc, it will come out smaller and longer and so be stretched less and hence more likely to slump. You may need to use lower layer heights.

For example I used to be able to print 0.4mm layers with ABS from a 0.4mm nozzle. Now, after changing supplier, I either have to do 0.35mm layers or use a 0.45mm nozzle to get the same quality.

An easy measurement to make is the diameter of the filament extruded into free space. You will probably find the black is coming out smaller than the previous one. Then if you compare the cross sectional area of that with the cross section of your filament paths you will probably find the "good" ABS is being stretched to a smaller cross section but the "bad" ABS isn't.

Lower temperature may help by increasing the viscosity and hence the die swell but ABS has quite a shallow viscosity temperature slope so you might find it gets too weak bonds before it gets viscous enough.

Edited 1 time(s). Last edit at 11/12/2013 08:04AM by nophead.

---

[www.hydraraptor.blogspot.com]

Nophead, how does this generally apply to different colors of ABS (other than white) when supplied by the same vendor? And how would one handle this in the software, in the instance that you are printing with more than one, or more than two colors for that matter when this is a consideration?

It is mainly a function of viscosity so if the pigments change that they will have an effect. I have noticed it with coloured PLA from Faberdashery but not really with ABS that was from the same factory but different in colour. I think the base material has the most effect and pigments affect PLA more because it is not very viscous to start with.

You don't need to handle it in software because it doesn't affect any flow maths. It simply sets the top limit on layer height so if you were using multi colours you would have to go with the lowest one. I.e. the one with the least die swell would set the maximum filament cross section you could use.

---

[www.hydraraptor.blogspot.com]

Thanks, Nophead. That helps quite a bit.><

If you buy a bunch of filament, it's amazing how different it all is. Some is pretty opaque (and loaded with something to make that be true). Some is quite stiff (but flows at a wide range of temperatures). There are no standards on this stuff. The only control over the mix is what ever the vendor decides makes sense. If their "standard" hot end is different than yours, that control may or may not be good enough. All of these materials are blended from various feed stocks. That's the way most chemical "stuff" works.