Solved Lifting edge and dragging nozzle across bed

Hello,

I have two issues with my new Robo 3D that I was hoping someone could help me with.

Notes:

• Robo 3D R1 (Purchased Dec 2014).
• Running MatterControl v1.2.1 (build 1.2.1.3479) on a MacBook Air, 10.9.5.
• Printing with high quality ABS filament.
• 2 years printing experience on a Solidoodle 2 with RH using ABS. New to the Robo3D/MatterControl arena.

Issue 1:
After the printer does the last level check (right before printing) on the right-back corner, it drags the extruder nozzle across the surface. See drag.jpg. Some details/thoughts:

• I have leveled the bed using the configuration control panel twice with the Kapton tape before printing.
• It seems to me that it might be some sort of firmware issue where it doesn't raise the nozzle before moving into the print position.

Issue 2:
The left edge of my print looses adhesion. See lift1.jpg. Some details/thoughts:

• I am printing a simple calibration angle that has been reliably printed dozens of times on my other printer on Kapton tape.
• Filament temp = 235, Bed temp 95.
• Printer is in a box with door to contain temperature/smell.
• I have leveled the bed using the configuration control panel twice with the Kapton tape before printing.
• I verified there were no bubbles in the kapton right under the printed part.
• I tried the same print with no Kapton, and used gluestick instead. It did the same thing.
• I rotated the model before printing to see if it was just that edge. It held a bit better, but lifted on the left edge near the end of the print. See lift2.jpg (hard to see, but is raised off the surface).
• It seems to me that maybe the plate isn't heating evenly in that left area. Perhaps I need a replacement heater?
• I could try to print with the bed at 100 C, though I doubt it will work since 95 fails early in the print.
• On my other printer, I set the initially nozzle height so it "squishes" the first layer into the bed a bit using a combination of led leveling and RH settings. I cannot figure out how to do something similar on the Robo3D/MatterControl.

Any advise to get me printing reliably would be greatly appreciated.

Jeff

 

Maybe Matter Control is the problem. You don't need to use it and most people don't like it. I use Cura and so far its helped me fix most of my problems.

 

Kapton tape Kapton tape.
Use hair spray. Lower your bed temp before you start a fire (80 is fine and almost at the limit of what it will hold) , If you're peeling on Kapton tape.. not surprising, its just smooth tape. On glue? That's odd, on hair spray.. impossible.

If its CURLING, its not an adhesion problem... your fan is on. fan off for ABS. Don't trust the check box. check the fan with a flashlight. If its still peeling.. suspect you're printing with nylon not abs.. or its the crappiest ABS ever.

Matter control IS Repetier. Its fine.. has all the same functions. CURA is ... blah... I don't like it. When you switch slicers and it fixes your problems its because your settings are wrong. (sometimes they get corrupted it seems and you have to clean up and start again)

 

I too struggled yesterday with my first attempts to print (a replacement the Robo3D carriage at 230deg/90deg) with ABS. One end always lifted regardless of how the piece was oriented. Results improved when I used a 5mm Brim.

 

1) Not a firmware issue but a G-code. Add a

Code:

G1 Z5.0

After G29 in the startup gcode

2) If you modified the firmware and you have a connector clip on the bottom, be sure to cut that off and jumper the wires together. Also ensure that the ramps board has sufficient cooling.

I never had great luck with ABS on kapton. An additional material, like ABS slurry or hairspray works well.

To lower the first layer you need to adjust your Z offset from either the slicer settings or the M565 Z0 (more negative = further from bed). Do not use the automatic print leveling/paper method.

Though your first layer doesn't look too high IMO

 

recommended starting gcode:
G28; home X Y Z axis
G1 Z5 F5000; will lift Z axis by 5 mm
M565 Z-1; sets the offset for the auto-leveling mechanism by 1 mm
G29; runs auto-level
G1 Z5.0 F1800

recommended ending gcode:
M104 S0 ; turn off extruder
M140 S0 ; turn off bed
G28 X0 ; home X axis
M84 ; disable motors

 

Thank you for all the pro-tips everyone.

Mike's suggestion of using the G1 code solved my "dragging" issue.

As for the curling issue, it was the fan (thanks defendermd). I disabled it by setting the speed to 0 in all settings/layers. Tossed the Kapton and used gluestick...worked great!

Then I printed a taller part (4 cm) and it was doing a perfect job into the last 3mm (see attached). It looks like it skipped a layer? I actually had this happen once before on a 5mm tall part I was printing (the last mm or so skipped) but didn't worry about it since I was still struggling with curling/dragging issues.

Thoughts?

Jeff

 

some shapes in some materials just love to curl up.

missing layer is almost certainly underextrusion. Looks like a couple layers are nearly there.
If you're extruding say.. .1899999 and your layer height is .2 eventually it catches up with you. You get a crappy layer, then stick again.

There are other scenarios but you get the idea. same with over extrusion. you'll get a fat layer, usually rhythmically thorough the print. thin layers seem to be a bit less rhythmic. If it doesn't grab to the layer below, it cuts the corner and you get a string on the inside instead of a layer on the outside.

I think I can see a couple other near misses in there. I personally would rather see a fat layer here and there than a thin one.

I would print a test part, adjusting your layer extrusion by a tiny bit.. 1% at a time.
http://www.thingiverse.com/thing:137424

You're almost there. To get really good prints you have to balance a lot of settings... Otherwise just go fat and sloppy. Overextrusion can clog your nozzle and skew your print height though..

 

I finally got around to poking at the printer again. Very interesting results...

(1) I did an extruder calibration using this technique. Found out I was under-extruding by 14mm. I entered in the new steps, ran the same print again and produced a different result (see fixedSteps.jpg).

(2) In frustration, I decided to install MatterControl on my Ubuntu box, used the same default configuration (EEPROM did not permanently update from (1) so it was back to the default steps.), and ran the same print. It came out perfect (see perfect.jpg). So, perhaps it was the MatterControl/Mac 10.9 mix?

(3) I then proceeded to print a larger print. It got about 20% through, then my extruder jammed (see jamOfDoom.jpg). I managed to to clear the jam and clean up the extruder gears. I blamed the filament so switched to a new roll that seems to extrude nicely (I did update the filament diameter in the settings.)

(4) I decided to print a more intricate part with the new filament. It produced a very strange result (see strange.jpg). I inspected the extruder for signs of jamming, but the filament flows and there isn't any abs dust in the gears indicating it couldn't push it through. Upon closer inspection of the defective part, it looks like it printed fine until the extruder had to jump between the two geometries (Notice how it started fine, then sporadically fell apart after the base was printed, then printed fine when it just had to finish the peg.) Here's what it looks like if you break apart a layer to look at the filament pattern (see crossSection.jpg)

My thoughts:

• Perhaps it's some sort of retraction/flow setting? Try printing tower test as defendermd suggested as a benchmark.
• Might reprint after a fresh boot to make sure it wasn't Ubuntu going nuts?
• Look at gcode as Mark instructed to look for odd layers?
• Try Cura as k1e1v1i1n suggested?

Any guidance would be appreciated.

Jeff

 

I had to bump up the temp other 10C (220c, pla) with the new filament I was trying out in order to get it to print consistently. Also I adjusted the speed down just a bit.

 

PROBABLY not the filament.
Your e steps are updated in the firmware under config.h if you can't find it (its just a set of numbers defining the axis steps) you can search your current ESTEPs and find the line. Esteps are the last number.

It sure looks like its skipping layers/not extruding properly. Ensure you're temperature is good, and that your hobbed bolt isn't clogged.
Make sure your filament isn't getting snagged on the way TO the extruder. Make sure your extruder "filament clamp" is tight enough to push the material.

I would bet you a large glazed donut your slicer has zero to do with it.. your settings could.

Ensure you haven't changed any of the other steps in the firmware/eeprom.. z axis has to be 2560.. or was it 2650.. whatever.. ensure your print layers are even multiples of z steps (its easy.. 10ths of a mm fits) .1 .2 .3. Start with 0,3mm just for ease.

I'm not an electrical engineer, but I worry that your potentiometers or other hardware might be wonky... but I still think you're underextruding... and since it's not periodic, its probably mechanical unless the Z axis is acting improperly at the hardware level.

 

actually looking at your "perfect"... its quite a bit from. You have periodic underextrusion. Looks periodic anyway. see... 1 2 3 4- 5 6- 1 2 3 4 - 5 6...
the "5 6" seem to be under extruded. That probably is a matter of the z axis steps being set wrong (wrong firmware you need 8mm rod firmware) or your layers adding up shy of full. that causes a gap eventually. The z axis keeps moving but your extrusion isn't keeping up.

 

of course.. many variables.. but 220 for PLA? that's hotter than I can go without making a mess. It should extrude nicely between 190 and 210 depending on color and brand at almost any speed. Ymmv

 

Operative word being should. But it doesn't. Prints like a dream at 220C on a e3d v6. All my other plas print at 208C

 

maybe your fan is much more powerful... maybe your temp is inaccurate. all mine call 210 the top end... i rarely approach it. color can change that temp quite a bit.

 

Spent an hour at matterhackers discussing all those points. Their recommendation bump up the heat and it worked it went from inconsistent to printing reliably and beautifully 7.5" tall with an nice raft.

Fyi it was their pro line filament. It was not fusing together at 210 so I kept bumping it till it did.

 

Look to me like you need to run an oiler. You're getting jamming and inconsistent flow which is usually indicative.

Higher temp would be good.

 

OK, thanks everyone...much appreciated! I'd like to poke at the firmware first. Here's what I know:

• This unit was purchased 12/2014, new.
• I am not using the 5/16" threaded rods so I "assume" the rods are M8 (I don't have M8 bolts to verify).
• MatterControl says firmware version is: RoBo3DR1AUTOV1 and an update is available.

I researched the firmware update and saw updating was a hit/miss situation so I hesitate to hit the "update" button.

Proceed to hit the "update" button?

Jeff

 

An easy way to check which rods you have is to just count the threads over a 1 inch span. I do not have the 5/16 rods but I believe they are 18 TPI (thread/inch). The M8 are going to be right around 20 TPI (1.25mm/revolution).

The original firmware is set for the old standard rods and the update is for the metric.

Another place to check is your steps/mm in your EEPROM.