Variable print quality within a print - where did I mess up?

[alaris2 1]

So I'd really appreciate some (constructive) critique of my ultimaking so far. I've levelled beds, finally managed to get the extruder knob to the right tightness, fought with firmware, removed repg 026, installed cura and marlin, worked around some bugs in cura, calibrated the E, played with many temperatures and finally have something printing.

but it's not exactly what I was hoping for.

http://i177.photobucket.com/albums/w228 ... mtrial.png

#1 was my first attempt - the extruder knob wasn't tight enough so it wouldn't feed properly, temperature was probably too low (200 - how does anyone print this low? I can't get the PLA to flow until I reach almost 230?). it's a vase3 made almost entirely from strings and blobs. I only left it running so long because it amused me to watch and I wondered how long before it fell to pieces.

#2 is a more serious print at 210C, but I stopped it because I found the diameter of the silver PLA supplied with my UM seems to vary from 2.72 to 2.94mm - I'd calibrated it for 2.90 and it was making a mess. is this normal? I then recalibrated for 2.85mm..

#3 almost looks like a real object. the temperature is now 230C. when the print starts, the extruder 'pees' all over the place, again I'm not sure if this is normal or not. at any other temperature nothing comes out the extruder during this part..

but the print starts really well, good clean lines for maybe the first 10 layers. then it's OK for quite a bit longer before reaching the purple ringed area. I don't understand what happens here - the print turns into strings and blobs again, but sorts itself out after a few more layers. any suggestions?

and then there's the weird cyan ringed area - the strange diagonal deformations appear to be caused by part of the interior wall in some unexplained way..

then it went mad and made a birds nest on top for me. I left the room for a moment and stopped watching it so it threw a hissy fit. jealous machine.

oh, and how are you supposed to remove the print from the blue tape? it always seems to stick so well that my efforts to remove the object actually snapped it off near the base (that's why there's a fracture line a few layers up). I guess I'm missing a trick.

sorry if these questions are all answered elsewhere but I couldn't find anything other than suggestions the silver PLA might not be fit for purpose.

all the best, nik.

[Daid 305]

Oho... I see the silver PLA. Which seems to vary in thickness a lot. That might be your problem causer.

I suggest to print slow, at a high enough temperature (230C). This might help against jams and stuff. But still, with varying thickness you will have problems. Especially on those kinds of prints. Try an octopus or yoda to see if that turns out better. Organic shapes are a bit easier.

[gavinmetzler 0]

Well I'm not too experienced myself, but I'd say "peeing" before printing is normal with PLA - in fact, if it isn't oozing, I'd say your temperature is too low, and thats why you've got such a sparse print on #1. If you can spare a few dollars, get a multimeter with a thermocouple probe (about $30 here in Australia), and hold it directly in the flow of the PLA coming out of the nozzle while extruding manually and see what is - I set my program temp to 210 to get 190 at the nozzle.

I don't know about the other problems - maybe a communications problem - what baud rate are you using? Also maybe try Slic3r - its not based on Skeinforge so gives different results - I've noticed Slic3r produces smaller gcode - e.g. fewer decimal places and no Z moves when there are none etc.

[destroyer2012 0]

Do you happen to know exactly what steps/mm you used to produce #1? I'm curious because I think that consistancy would work as a support structure.

[alaris2 1]

thanks all. you reckon the silver is no good either Daid? I'll try a different print and see what happens..

my temperature probe is really loose in the metalwork, so it won't be getting good contact and probably doesn't register the right temperature, that would make a lot of sense. maybe some thermal grease or something might help here.. presumably temperature regulation won't be great if I can't monitor the temperature properly and variations might cause those strange defects?

or is the general consensus that it's definitely feed rate vs filament diameter related?

I'll give slic3r a go and see what happens too, I've heard some quite good things about it.

[alaris2 1]

the steps/mm was set to default 50 - that print is the consistency of a spiders web, I was amazed it held together and it would make excellent support, but it was caused by the thumbwheel on the extrude feeder being set too loose, which caused me to have to set the E number in cura to about 1100 or so before I got 100mm filament fed through. the temperature was set to 210 or 200C (from memory, ie a bit too low really to properly melt the PLA).

i'm not sure it's reproducible enough to be useful unfortunately..

[Daid 305]

Well, it's not that silver is no good. It's just that the last few people reporting problems with printing all had silver filament from Ultimaker, all with varying dimensions. Which can indicate a bad batch of filament from Ultimaker. (just guessing here)

I'm not an expert on temperature readings, mine are spot on. It's quite simple to check how good your temperature is. Around 160C you should be able to remove filament, around 190C you can push it trough by hand, and print with it. And if you print at 230C you get oozing problems before you get feedrate issues. Marlin (from Cura or ErikZalm) has a much more stable temperature then the default firmwares, so it's recommended to upgrade to those.

[antiklesys 3]

Doesn't look like a variation in thickness to me. I think it was more something like the plastic not being extruded properly.

I would double check the feeder screw if I were you and make sure it's clean.

[alaris2 1]

a happy ending to this topic - I bought some new PLA which varies by less than 0.05mm across the filament. this stuff works properly and the before and after can be seen at

http://s177.photobucket.com/albums/w228 ... ultimaker/

still a few very minor things to try and understand to get my prints looking as good as Daid's tho..

nik