1st half of part prints poorly, 2nd half ok. Why??

[MarcoRemius]

I'm using Slic3r to generate the G-Code. I can post up the .ini file if necessary.

But essentially, the "lower" half of anything I try to print comes out sloppy, but the second half looks nearly perfect. Any idea what could cause such a phenomenon?

Sorry for the crappy pictures, but the iPhone autofocus wasn't cooperating. Any help is appreciated.

[ednisley]

the "lower" half of anything I try to print comes out sloppy, but the second half looks nearly perfect.

If the first layer doesn't adhere to the bed, subsequent layers will be misaligned and look terrible, but, surprisingly, the mess eventually levels out and starts looking pretty good.

If you haven't already done so, reduce the first layer speed to about 20 mm/s.

If it's already running at a reasonable speed and not sticking, then use hairspray to glue the first layer onto the glass. Various threads around here have suggestions, but it seems you can't go wrong with any super-duper maximum extra strength ultimate hold hairspray.

While you get all that straightened out, smaller test objects may be helpful:
viewtopic.php?f=2&t=1350&p=3917#p3917

[MarcoRemius]

Thanks! I'll give it a whirl!!

[Dale Reed]

MarcoRemius,

The bracelet on the MG sample g-code is probably a good one to test on for your issue. It's a single layer of filament thick and prints pretty quick, so you don't waste a whole lot of filament, even if you let it finish.

The first couple layers are critical. If you don't get them stuck down firmly on the glass, you'll mess up the print fairly severely. You've seen how many layers it takes to get things "evened out". The idea on the first layer is to print it slower, with a high enough extruder temperature, high enough bed temperature and with a good clean well-prepped bed to get it stuck down solid, especially at the corners. (I am one of the many using AquaNet Extra Super Hold hair spray -- the purple can -- but as Ed said, any hair spray that advertises extra monster ridiculous mondo hold is a good bet. Look for lots of ingredients that seem like polymers: polyxxxxxx and or acrylates and such.)

Other things that are critical on the first layer are:

* Bed leveling: The idea is to make sure that if you hold Z at a particular height, the nozzle is the same height above the bed at all X,Y coordinates.

* Bed spacing: The idea is to extrude the first layer onto the bed so that the "squishes" it just a bit. You want adjacent threads to meld together, not be separate, but you don't want the bed to block the nozzle and clog it.

After the first layer, the trick is to get each layer to stick to the one below it. This is primarily a function of extruder nozzle temperature and cooling fan speed.

I think jimc uses a 0.006-inch feeler gauge to set the gap between the nozzle and bed when Z is at home. That's probably a great way to do it. I run my gap a tad higher, then set up a small negative Z offset in S3D when I slice. For now, with printing test bracelets, I'd suggest this:

1. Get the bed level (consistent gap from the nozzle at all X,Y coordinates, whatever that gap may be.

2. Get the Z stop adjusted so don't clog the nozzle. A little too much gap is better than to little at this point.

3. Clean the glass, then give it a fine misting of hairsparay.

4. If you're using MG PLA, set the bed temp to 70 C and the extruder temp to about 215.

5. Print the bracelet a few times. WATCH CLOSELY as the first layer prints. If it's not sticking down, click the Z knob once clockwise to nudge the extruder closer to the bed and see if that helps moosh it down on the be a bit. If the nozzle seems to be dragging, click the knob counterclockwise to bump the extruder away from the bed a tad.

Once you get it printing well, adjust the Z offset in S3D and/or tweak the adjustment of the Z limit screw to get it to print the first layer at that height consistently.

Hope this helps!
Dale

[MarcoRemius]

Dale,

Thank you for the writeup. I'll be sure to try exactly that once my frustration cools off a little. I'm pretty annoyed at this thing right now.

Any good tips on how to clean excess plastic off the extruder nozzle? The remnant plastic grabs the filament stream and drags the model with it.

[jimc]

heat it up and wipe the bulk off with a rag. a good tool to have is a small wire brush. avail in the tool section from home depot near the welding stuff. they have small wood handled soft stainless bristle ones. heat the nozzle then wire brush it a little to clean off excess and burnt on bits.

[MarcoRemius]

heat it up and wipe the bulk off with a rag. a good tool to have is a small wire brush. avail in the tool section from home depot near the welding stuff. they have small wood handled soft stainless bristle ones. heat the nozzle then wire brush it a little to clean off excess and burnt on bits.

Ah, thank you.

I think the remnant plastic was heating up on the nozzle... and when the head would traverse, it would stick to the printed plastic and drag it around. I'll clean that off today and see where I end up. Thanks all, I'll keep you posted on my progress. I really need this machine to run well for my business and it's been a fun yet frustrating start

[Dale Reed]

MarcoRemius,

What jimc said. I use some old heavy cotton t-shirt material, reserved for the purpose and not used for cleaning the glass. Just use enough thicknesses so you don't toast your fingertips. When I prime the nozzle with filament before starting to print, I wipe it off, test the extrude to make sure it comes out straight and doesn't curl back up on the nozzle. When it extrudes straight down, you're ready to go.

Dale