Maybe Belt/Motor Skipping?

[analog_banana]

After finding out that I had a defective PSU with the kit I have replaced it with a high quality, high power PSU.
I have also beefed up the cabling and my bed / hotend are heating up quickly and reliably.

The print problems I have been experiencing persist - possibly worse - but they have moved sides..
The blobbing / nibbing is more on the opposite right side now.

I am no longer using a print cooling fan except on the PEEK as recommended for ABS. (This was my way of controlling the blobbing)

To me the problem still looks like the head mashing over a noodly, squishy unset layer, causing errors as it
squishes its way over the boggy previous layer.

Now some of the more serious power defects have been looked at in detail I am thinking about the motors and belts.

Occasionally I get an incosistent sound when the machine is doing a G28 home manuever.
The sound / skip may be present at aother times, but it is most apparent on thelong continuous travel of homing.
It's probably only one time in 30.
You can view, and more importantly hear, the problem on this movie:

[youtube]http://www.youtube.com/watch?v=IjgRf8SZw-s[/youtube]

photos showing the current problems:
Owl from last week = LEFT / Latest owl = RIGHT

[cope413]

When you need help troubleshooting, it really helps if you post your print settings...

If you don't post your settings (namely, Material, temp, speed, etc), then we have to add extra posts to your thread to ask you for them. Kinda like this post right here.

[analog_banana]

Er yeah, I was just doing that, honest...
Transcribing multiple instances of about 50 floating point values and integers is my favourite repetitive task
see below:

[geneb]

You know that the configuration values for the print job are shown in the g-code window of rep-host, right?

Put the extrusion multiplier back to 1 and drop the extrusion temp by 15 degrees.

Try a cube and see how it goes.

g.

[analog_banana]

thanks, that's a much easier way to share settings! - see below..
Not having much luck with the output though.
I still suspect that something is wrong with the machine.
I may go back to the exact pdf manual settings and explore slow variations from there.

I guess I should probably remeasure all the temperatures,
recalibrate the extruder and redo the pid tuning - post PSU revamp????

; layer_height = .2
; perimeters = 3
; top_solid_layers = 3
; bottom_solid_layers = 3
; fill_density = .2
; perimeter_speed = 30
; infill_speed = 30
; travel_speed = 130
; nozzle_diameter = 0.5
; filament_diameter = 1.7
; extrusion_multiplier = 1.0
; perimeters extrusion width = 0.50mm
; infill extrusion width = 1.02mm
; solid infill extrusion width = 0.85mm
; top infill extrusion width = 0.85mm

[Batteau62]

One problem could be your extrusion widths. I would set them all to .55mm (10% more than nozzle size). I think anything over that is causing over extrusion. The perimeter you have set at .50mm should work? Have you tried printing a single wall cube?

[analog_banana]

Thanks - I was using .55 on th owl etc - I just took a step back with this last cube.

I will try a return to .55 extrusion for my next cube.

I have tried many many many single wall cubes!
They mainly show the corner abberations that you see on sharp corners here.
Sometimes sagging at higher temps - as you might expect.
They do not suffer from the blobbibg nibbing - as the simplicity of the geometry seems to prevent it.
I won't bore you with the photos.

[geneb]

Try backing it off another 5 degrees and see what you get. It LOOKS like over-temp filament.

g.

[cope413]

I agree with Gene, looks like over temp, and perhaps over extrusion.

Have you measured your filament? Is it really 1.70mm?

Also, have you tried Cura? Because it's got less options, it's often easier for isolating issues.

[analog_banana]

ANOTHER TEST CUBE BASED ON COMMENTS:

; layer_height = .2
; perimeters = 3
; top_solid_layers = 3
; bottom_solid_layers = 3
; fill_density = .2
; perimeter_speed = 30
; infill_speed = 30
; travel_speed = 130
; nozzle_diameter = 0.5
; filament_diameter = 1.7
; extrusion_multiplier = 1.0
; perimeters extrusion width = 0.55mm
; infill extrusion width = 0.55mm
; solid infill extrusion width = 0.55mm
; top infill extrusion width = 0.55mm
; first layer extrusion width = 0.55mm
G21 ; set units to millimeters
M190 S75 ; wait for bed temperature to be reached
M104 S210 ; set temperature

No real difference with the lower temperature and .55 extrusion width.
Print looks pretty much the same.
I do measure my filament in multiple places / 90 rotational degree offsets and average the results.

The fact I can't get a decent print may make it look like I'm just a bit dim, or not very practical.
But even if this is the case, after 10 months, I reckon *If the machine works*
the dilligence and iterative repetition I have put in would get me better prints.

I suspect that endlessly varying the settings, or switching to Cura, are hidings to nothing.
You can't fix hardware with software.
You can improve hardware performance with software.
I think I need a better mechanical baseline before I can benefit from software-based enhancement

I will keep playing with the settings though..
That's been my SeemeCNC Rostock Max experience for 10 months.

I'm going to recalibrate the temperatures and start focusing on the belts bearings and motors.
In Review: Here are the machine's current strengths and weaknesses:

PSU - replaced, new PSU tested OK, delivering plenty of power.
Ramps: OK as far as I can tell.. Maybe Pololu trim pots need adjusting to stop skipping?
Cabling - replaced with fat copper - tested and seems low resistance.
Bed - OK, nice and hot, temperature accurate. **
Motors - skipping maybe?
Pulley Gears - old obsolete type - seem ok - maybe skipping?
Timing belts - OK? Tension OK?? skipping?? Maybe test more?
Delta Arms/Bearings - Seem Ok - no play, using lithium grease
Extruder - replaced with EZstruder, calibrated, delivers accurate feed.**
Hotend - added approved peek fan and aproved shroud - temp accurate **
Bed levelling - radius set - flat all over - adhesion is now good.

** May need recalibration based on PSU extra power??

[DavidF]

Would it be possible to take a video of the machine in action? It may help with any mechanical issues there may be. I'm new to 3d printing myself but have not had any mechanical issues with my rostock max. Any time I have had a problem its been with that damn slic3r doing something I don't understand.

[analog_banana]

not the full experience - but see the text of the OP
The link below is a video of the very occasional, but noticeable, skipping i get during homing
one clue is I *think" it affects all motors simultaneously.
If one motor or another failed intermittently I think it would sound different, less regular.
edited to update link:
[youtube]http://www.youtube.com/watch?v=IjgRf8SZw-s[/youtube]

[analog_banana]

Yeah, gearing myself up for the full slic3r testing experience again...

[DavidF]

For some reason I can't get the video to play...ill try again when I get to my home computer.
Have you tried the cura for slicing yet? Works wonders for me. I swear that slic3r is crazy and does what it wants to at times. I'll have to look at its latest blunder when I get home to make sure it wasn't something I did before I say more, but it printed my part solid when I had infill set at .2

[analog_banana]

I guess cura's on the list after the next temperature check and calibrate.
I do know that slic3r has a threshold setting for solid infill.

Look for "Solid Infill Threshold Area" under the Infill tab.

If the thing you were printing had thin walls or parts, they may be automatically printed as solid -
e.g thinner than the default to 70mm squared, which is approx 8mm x 8mm - quite big really!

Here's my vid on youtube


[youtube]http://www.youtube.com/watch?v=IjgRf8SZw-s[/youtube]

[DavidF]

Sounds more like a pause than a skip?
Down load cura, slice a part save the gcode, then open and print it in repetier.
Do this as your next step so your not chaseing a ghost in the machine.
You'll have to enter printer size and height in the first run wizzard, but you can be printing with cura in about 30 minutes after you read this, have a drink, and take a leak.

[bubbasnow]

copy this gcode to script 5 and run it with video going


G28
g0 z10 x0 y129 f3500
g0 z10 x0 y129 f3500
G4 P500
g0 z10
G0 Z10 X111.72 Y-64.5 F3500
G0 Z10 X111.72 Y-64.5 F3500
G4 P500
g0 z10
G0 Z10 X-111.72 Y-64.5 F3500
G0 Z10 X-111.72 Y-64.5 F3500
G4 P500
g0 z10
G0 Z10 X0 Y0 F3500
G0 Z10 X0 Y0 F3500
g4 p500

;drag out
g0 z10 x0 y129 f1000
G4 P500
G28
G0 Z10 X64.5 Y111.72 F1000
G4 P500
G28
G0 Z10 X129 Y0 F1000
G4 P500
G28
G0 Z10 X64.5 Y-111.72 F1000
G4 P500
G28
G0 Z10 X-64.5 Y-111.72 F1000
G4 P500
G28
G0 Z0 X-129 Y0 F1000
G4 P500
G28
G0 Z10 X-64.5 Y111.72 F1000
G4 P500
G28

[analog_banana]

thanks bubbasnow, here's that test:
[youtube]http://www.youtube.com/watch?v=BAxq9wV170I[/youtube]

[analog_banana]

I tried with cura - same temperatures, layer heght, infill and speed as in slic3r.
Different slice tool, Same defects.

Settings:
M140 S70.000000
M109 T0 S210.000000
T0
M190 S70.000000
;Sliced at: Wed 29-01-2014 09:42:47
;Basic settings: Layer height: 0.2 Walls: 1.5 Fill: 20

[geneb]

That is just WEIRD. Are you absolutely positive the material you're printing is ABS? Just for giggles, try printing at 190C with a 55C bed temp and see what you get. I'm wondering if someone mis-labeled the material and you got sent PLA instead of ABS.

g.

[analog_banana]

I will try that at the next opportunity.
I certainly does not smell like PLA.
Smells like hot plastic, - not the pleasant smell of pancakes!

I might go back to some PLA tests with authenticated PLA now I have moved forward on other fronts.
Any opinions on th eintermittent skipping?

[analog_banana]

Tried it - It won't extrude properly at lower than 200 - 210
the ezStruder skips and what plastic does come out is a bit fat and struggles to get out...
So it is ABS - maybe just BAD ABS.

I don't know where to point the finger of blame next...

[lordbinky]

The corners of the last picture remind me of when I go too fast or too thin and the filament is stretching alot. I doesn't look like that's the issue when I look at your settings though.

Have you tried drying your filament out yet? That improved some crappy ABS I've had before. Well, it went from really shitty to kinda crappy, but still an improvement.

[analog_banana]

Another crazy scheme...
Luckily I own a powerful electric dehumidifier and 6 one-pound bags of silica gel.
I store my filament in bags with the big silica gel pouches.

I will be trying some PLA again, fresh from the de-humidor,
The ABS will be electrically dehumidified and silica-gel stored while it's on the bench.

[cope413]

I don't think it's the material... I've had bad filament, but the thing with bad filament is that it's typically inconsistent. This is doing the same thing consistently. Also, if it's PLA, it likely wouldn't stick to the bed at 85C.

How long does that cube take you to print? It really looks like the hot end is just sitting on each of the pillars and melting the crap out of the filament, leaving you with that mess.