I need a Printer Whisperer... I think I'm the underdog.

[RayaSanguin]

So I'm experiencing a few errors that to me, are very random. Granted, I've been assigned by my boss to be the Printer Master, which entails that I learn everything about it as if it were my best friend. We were just antiquated yesterday as it started off great because my boss had already done a lot of calibration of it after he built it... so it could make hollow boxes, ninja stars, and hollow cubes. However, there is occasional oozing, warping, etc and he told me to "play with it".

Tbh I hadn't even changed anything about the settings and such, all I played with was the Feedrate and Flowrate.


God damn me if I didn't totally ruin the thing.

Now, it can't even start a project...

I've restored all of the previous settings yet for some reason the extruder seems to struggle putting the beginning layer down which of course ruins the rest of the piece til it gets pulled up and off and twirls around into a goopy blue ABS plastic wad of panic.

He told me to raise the extruder temp to 240, and it ran well, really well... for one more ninja star then BAM... about 20 trials of trying to peg the first layer of any project...

I figured maybe it was adhesion, changed tape. No Cigar.
Maybe its the bed temp? Nope.
Maybe its the Z axis? Tested the paper trick... nope, thing is calibrated brilliantly.

Please Help me.

These are the settings:

Extruder - 240 C
Bed - 90 C
Fan - OFF

Filament Diameter - 1.5
Extrusion Multiplier - .9


Layer height - .3
First Layer Height - .35

Perimeter - 4

Fill Density - .6

Perimerters (Speed) - 30
Small Perim - 30
External Perim - 80%

First Layer Speed - 30%

If you need anything else, I will happily post.. Just help me XD

BTW we are using Blue ABS 1.75 mm.

[geneb]

The one thing that stands out to me is that filament diameter. Is that the ACTUAL averaged diameter over a few feet of filament? In my experience I'll see an average of about 1.67 to 1.72. Under-specifying your filament diameter will result in way too much being pushed out the hot end. The extrusion multiplier should be the default (1.0 I think).

g.

[Jimustanguitar]

The temps look right-ish. 140 is a little hot, but if you haven't calibrated your thermisor it's not accurate anyway and you should just use what works. Hot enough that it's smooth, but not so hot that it starts popping.

You could also tweak the first layer height setting. Obviously this will vary from person to person depending on how tight they calibrated it with the piece of paper, but it's worth a try.

Set your multipliers all back to normal. This could be in the repetier software or on the machine itself (you have the LCD, I assume?).

[lordbinky]

I use ABS on glass with hairspray (search forum for the proper requirements of the hairspray), I print my black ABS at 240°C (calibrated) for .3 mm layers, at 255 it looses a little detail but prints end up with a pretty gloss. I had gray abs from a different company needed 255 to stay good on tall models before this filament. I am able to keep my bed temp down to 60°C using hairspray, and if I used the glue sticks I needed about 80°C but the hairspray is more consistent with less effort for me. I skipped using tape for ABS.

[bubbasnow]

im up to 105c bed temp for black abs with hairspray. I could get away with lowering it but 105 has worked well so far.

[RayaSanguin]

The one thing that stands out to me is that filament diameter. Is that the ACTUAL averaged diameter over a few feet of filament? In my experience I'll see an average of about 1.67 to 1.72. Under-specifying your filament diameter will result in way too much being pushed out the hot end. The extrusion multiplier should be the default (1.0 I think).

g.

Right now, the multiplier is on .9, and the input diameter is 1.5... My boss said that he had measured the width in 3 different areas and that it was 1.7 - 1.75.... Apparently he's telling the machine that it's 1.5 because it accounts for how slow the filament is?

I use ABS on glass with hairspray (search forum for the proper requirements of the hairspray), I print my black ABS at 240°C (calibrated) for .3 mm layers, at 255 it looses a little detail but prints end up with a pretty gloss. I had gray abs from a different company needed 255 to stay good on tall models before this filament. I am able to keep my bed temp down to 60°C using hairspray, and if I used the glue sticks I needed about 80°C but the hairspray is more consistent with less effort for me. I skipped using tape for ABS.

Currently, we are using an aluminum plate however I'm certain glass will be used on our orions... so thank you for this!

im up to 105c bed temp for black abs with hairspray. I could get away with lowering it but 105 has worked well so far.

patience of yoda... I'll try my luck.

The temps look right-ish. 140 is a little hot, but if you haven't calibrated your thermisor it's not accurate anyway and you should just use what works. Hot enough that it's smooth, but not so hot that it starts popping.

You could also tweak the first layer height setting. Obviously this will vary from person to person depending on how tight they calibrated it with the piece of paper, but it's worth a try.

Set your multipliers all back to normal. This could be in the repetier software or on the machine itself (you have the LCD, I assume?).

The ABS has been popping, so I'll play around with that.
We have the Rostock Max.
I'm certain he has calibrated it, however I'll look around on the forum for trouble shooting on that. XD

Thank you All so much.

[RayaSanguin]

Okay, so I have restored all settings, raised the bed temp to 105 C, Lowered the Extruder temp to 220 C, brought the first layer height to .35, however looking closely, I see that there is no issue with adhesion... except for the first 2 seconds of laying down the line. The plastic is curling just right everytime to stick to the back of the extruder and get pulled along the twists and turns of the path... the plastic is not laying flat at the very beginning... which is screwing everything up. Could the extruder still be too hot? There is light pops when the machine is stagnant and as it sits, waiting for the bed to heat up (again), it makes at least 2 inches of waste string...

[geneb]

I suspect he's telling the machine the filament is 1.5 because he doesn't know what he's doing.

Pull about 2 meters of filament off the roll and spot check it in five places. Average the readings and put that result in the filament diameter. Set the extrusion multiplier back to the default.

You can print ABS on an 80C bed without problems. That's what I've been running Blue Max at - with Aqua Net hairspray on the glass.

Is the start of the job laying down a couple of loops around where the part is going to be before starting the actual part?

g.

[lordbinky]

If you want to take it a step further, I do the same thing as was just outlined by geneb the four rolls of filament I measured (so small sample size...) all have oval shaped filament, some more so than others. Taking a measurement one direction, which the filament typically rolls to make it the smallest diameter, and then holding the filament firmly and rotating it 90° then taking the measurement again will typically be larger, I average those two values and consider that one measurement point. I don't think going any further will provide any worthwhile benefit, if even that extra step is worth it.

[lordbinky]

Oh, and the initial line problem is easily solved by putting an initial outline around the print, this is called a raft in Kisslicer and I forget the term in other slicers. This helps with priming the hotend back up with filament and since it only prints that outline for a single layer, as long as the trace doesn't curl into the print itself, everything is fine. In Kisslicer you change how many mm away from the print it places this outline which increases the length of the trace.

[geneb]

NOT a raft. A raft is a whole nuther animal. It's a perimeter.

g.

[lordbinky]

HA! yes! That's the normal term for it. My brain has began to awaken a bit more since my initial attempt at what I call a memory. I could still be wrong, but I think in Kisslicer the perimeter is under the raft section and is the skirt selection, and then what is typically called the raft is the grid selection. I really wish people could stick to common conventions in their jargon.

[RayaSanguin]

I suspect he's telling the machine the filament is 1.5 because he doesn't know what he's doing.

Pull about 2 meters of filament off the roll and spot check it in five places. Average the readings and put that result in the filament diameter. Set the extrusion multiplier back to the default.

You can print ABS on an 80C bed without problems. That's what I've been running Blue Max at - with Aqua Net hairspray on the glass.

Is the start of the job laying down a couple of loops around where the part is going to be before starting the actual part?

g.

Haha, I'm glad that I'm not the only one that felt it didn't make sense...

I've found that 80C is pretty darn close to the sweet spot on the aluminum. I was reading around on the forum and cooling has a larger factor on the ABS than I had originally thought. 90C and above really delayed the 1st layer's gripping and it would freak out.

Yes, I have it do at least 3 perimeters, now that I've cooled it to 80C it grips on within the 1st or 2nd go round... however the end still gets caught on the extruder and if I don't intervene with my "poking stick" (destroyed paper clip) to lay it down before the whole thing gets messed up. I've just been micromanaging the 1st layer, poking it and flattening it ASAP before it cools as a lump that will get caught again.

The perimeter doesn't help me at all tbh, because it struggles at laying down the starting point... which it has to do for the project itself.

BTW thank you all for replying so quickly. I really appreciate the help!

[geneb]

I wonder if your nozzle is too high.... When it's at 0,0,0, you should be able to just barely get a sheet of paper under it and have the nozzle tip gently drag on the paper.

g.

[lordbinky]

and there he said it....

Bed calibration print time!

For a rostock:
http://www.thingiverse.com/thing:50505

If your calibration isn't spot on it'll show up with this. If your printer is perfectly calibrated on a flat print surface with a flat nozzle that is parallel with the bed, the width of the first layer will be exactly the same on the entire bed. It'll give you a starting point on whether you need to knock the Z height down .05 a couple times or if you need to go through a full alignment check and calibration process again.

[RayaSanguin]

I wonder if your nozzle is too high.... When it's at 0,0,0, you should be able to just barely get a sheet of paper under it and have the nozzle tip gently drag on the paper.

g.

I thought so before, but it went smoothly... XD
I'm going to have to take a video of this bizarre beginning curl and post it.

[RayaSanguin]

and there he said it....

Bed calibration print time!

For a rostock:
http://www.thingiverse.com/thing:50505

If your calibration isn't spot on it'll show up with this. If your printer is perfectly calibrated on a flat print surface with a flat nozzle that is parallel with the bed, the width of the first layer will be exactly the same on the entire bed. It'll give you a starting point on whether you need to knock the Z height down .05 a couple times or if you need to go through a full alignment check and calibration process again.

Oh my goodness thank you!

[RayaSanguin]

Brilliant. That .STL was an eye opener. Too cool.
*laughs*
You guys are too awesome, saved me so much stress.
Thank you!

It seems that the calibrations are were the real weak point is. In some places it was more than -.35 off...

[lordbinky]

If calibration alone does not solve your problem:

I was having onyx issues where it would warp just slightly in the center in a slightly different way every heat cycle that was forcing me to recalibrate far too often. I fixed it by doing a cutout in the snowflake spacer melamine board (I saw this after I went a little cthulu crazy and had already taken my dremel to the board).

http://forum.seemecnc.com/viewtopic.php?f=39&t=1234
Since then my bed has been fantastic with the glass plate on top.

IF you know your aluminum is milled with a tighter tolerance than the height variation you are seeing AND the 0 Z level variation is at points that are in between the towers

THEN I would suggest you consider that modification, if not at a minimum check the seating of your onyx.

ELSE Consider other factors like inconsistant endstop triggering. Simple fix is to move the endstop back a notch to make the screw hit the flat part of the lever arm or the button itself. http://forum.seemecnc.com/viewtopic.php?t=2459&p=18640http://forum.seemecnc.com/viewtopic.php?f=35&t=1170There are upgrades to this if you want to go further with it, optical/magnetic reed/ field effect endstops that have different trade-offs in exchange for possibly more accuracy (which may be completely unneccessary).



Forgive me for my formating.. I think there are 5 people in my entire building right now...

[RayaSanguin]

I re-calibrated thanks to all of the awesome advice, you were right, it was too high and now I'm printing awesome shurikens and octopi and my boss is very happy with the outcome. The real work is about to begin, and I can't thank you all enough. I also found that I get the best adhesion on 95 and that the blue ABS is smooth at 260C... on my machine anyway. I have no idea if he calibrated the heat sensors.. but now it works.. I'll have to post pics of the successful items.

[johnoly99]

Raya,

congrats! Now, be careful on those temps. Due to the nature of the beast, not all sensors are installed the same, therefore they all read differently. You can go in and edit firmware to make it read more exact, or, you can measure it, and compare and keep that number offset in your head. Our hotend is good to about 240c actual temp. Much past that, and we're always here to sell replacement ptfe liners and peek barrels


John Oly

[lordbinky]

^___ That is definitely a big reason to calibrate the thermistor, the Peek and PTFE like to follow physics and melt at their melting point, regardless of what your thermistor says.

A side note about level beds, I was talking with a friend at work and I was wanting to get some aluminum for a heat spreader and flat surface, but asking for parallel and flat within .01mm is apparently asking for something I didn't quite understand lol. I got a very blank look from a couple of machinists like I was crazy and they showed me how tight their tolerances would have to be to mill that, then they pointed me to a hydraulic shop that had shut down in town, but it had the only planer in town and no one knows who got it at the auction. Even then, if I went the patient do it by hand method, the only thing close to having a better tolerance than what I'm asking are cast iron lapping plates. I've moved away from asking for flat pieces of metal and decided to go with thicker borosilicate glass sheets, apparently glass makers don't scoff at tight tolerances or suspect you of making a missile.

[edward]

I just purchased a granite block for our shop that is flat to 0.00026" (so 0.0066mm) - well, that's what the papers say. We've occasionally needed one for quite some time so I used my desire to lap a build plate as an excuse to push the request.

As you said, lots of work, and I'm not sure when I'll get to it, but it seemed more accessible than the cast iron plates. So far I'm having awesome results with some 5/16" MIC-6 that I ran over with a rubber sanding block and some blue dye to take out the obvious high-spots and surface knicks.

[geneb]

IMHO, you guys are nuts.

The machine you're using is made out of wood. Wood expands and contracts with humidity and temperature changes.

Getting a build plate to the degree of flatness that you're after is just crazy as long as you're using it on a machine that can change shape with changes in humidity.

Also know that the machines aren't good enough to exploit a .01mm surface accuracy even if you didn't have the material changes to deal with. (there's simply too many unknown factors, such as hysteresis in the bowden tube and ongoing changes in filament diameter, etc)

Keep in mind that perfection is the mortal enemy of good enough and that FDM 3D printers are rapid *prototype* machines, not rapid *production* machines.

Build the machine that you're most comfortable spending the money on and then forget about the little things that at the end of the day, really don't have any effect on the end result.

g.

[edward]

Your point is a good one and I don't necessarily disagree, but I see it as removing one of the variables that might be causing bad print results. If I can say that my bed is *absolutely* flat, then I can move on to other things when troubleshooting.

Plus my build plate cost me ~$20 and 30 minutes of my time.