Welcome! Log In Create A New Profile

Advanced

Calibration

Posted by Axis 
Calibration
February 19, 2013 10:03PM
Hi,

I'm new to calibrating 3D printers.

I have a self-assembled Prusa Mendel with Wade's extruder, 1.75mm filament, .3mm nozzle, and Slic3r settings set accordingly (with layer height at .175mm).

Printing a test file '20mm-box.stl', the first layer seems to print okay. But after that, there doesn't seem to be enough plastic coming out.

Anybody know what I might be doing wrong?

Thanks.

Edited 1 time(s). Last edit at 02/19/2013 10:29PM by Axis.
Attachments:
open | download - Cube-Top-View-01.jpg (592.5 KB)
open | download - Cube-Bottom-View-01.jpg (570.3 KB)
Re: Calibration
February 19, 2013 10:58PM
Nevermind. In Slic3R, I turned the printing speeds down to 10, 10, 20, 20, and 20. And now it's starting to look more like a cube.
Re: Calibration
February 20, 2013 06:19PM
Not nearly enough plastic in that cube. Is the extruder steps/mm calibrated to your filament in your extruder?
Re: Calibration
February 20, 2013 07:05PM
Thanks for the reply, Dale.

In 'Configuration.h', I picked one of the defaults:

// X, Y, Z, E steps per unit - Metric Prusa Mendel with Wade extruder:
float axis_steps_per_unit[] = {80, 80, 3200/1.25,700};

I was thinking of increasing the E steps-per-unit, but was scared I might break something. But now that you mention it, 700 is probably meant for 3mm filament? Digging around the forum, I see that
Quote
nophead
The area of a 1.75mm circle is 2.4053mm, so steps per mm is 2.94 times higher...
So should I change it from 700 to 2058?
Re: Calibration
February 21, 2013 04:45AM
No actually, steps per mm should be rougly the same for 1.75mm and 3mm filament. You tell the slicer the diameter and it should feed 1.75mm faster by usind bigger E values.

To calibrate put a mark on the filament 120mm above the extruder, tell it to feed 100mm and measure how far the mark moves. If the mark is say 30mm from the extruder you have fed 90mm so E steps should be multipled by 100 / 90.


[www.hydraraptor.blogspot.com]
Re: Calibration
February 21, 2013 05:32AM
+1 the E-steps is a vital thing to change to get reasonable prints.

In sprinter and marlin it's possible to find optimal values by typing M codes which can be found on gcode list

Once you have found the value, Marlin can save it to eeprom using M500, sprinter needs a recompile.
Re: Calibration
February 28, 2013 07:17PM
@nophead: If I mark it at 120mm, how come I'm telling it to feed 100mm instead of 120mm?
Re: Calibration
February 28, 2013 09:21PM
Axis Wrote:
-------------------------------------------------------
> @nophead: If I mark it at 120mm, how come I'm
> telling it to feed 100mm instead of 120mm?


He was just using a random example. The longer length you are accurate at, the more accurate you will be calibrated. 100mm is a good calibration test.

The point is, telling your extruder to extrude 100mm of filament should pull exactly 100mm of filament.

You can even make a mark, extrude 10mm at a time, ten times, and see if your mark is moved 100mm at the end.
Re: Calibration
March 01, 2013 06:27AM
Axis Wrote:
-------------------------------------------------------
> @nophead: If I mark it at 120mm, how come I'm
> telling it to feed 100mm instead of 120mm?


If you feed 120 and it feeds 125 mm the mark will be inside the extruder so it will be difficult to measure.
Re: Calibration
March 01, 2013 07:39AM
If I tell it to extrude 100mm and the mark is 30mm from the extruder, wouldn't that mean it fed 70mm? Shouldn't it be 100/70?
Re: Calibration
March 02, 2013 01:29AM
Axis Wrote:
-------------------------------------------------------
> If I tell it to extrude 100mm and the mark is 30mm
> from the extruder, wouldn't that mean it fed 70mm?
> Shouldn't it be 100/70?

Depends on where you started. If you marked it at 100 from the extruder, told the extruder to move 100 then measured it at 30 from the extruder, then d_Ordered/d_Extruded is 100/70 and you should increase your steps per unit for E from 700 to 700*100/70=1000.

If you marked it at 120mm from the extruder, ordered 100mm of extrusion, then measured the mark at 30 from the extruder, you'd increase it by 100/90.

Instead of marking, I tape one jaw of my wide-open caliper to my filament, the other jaw to the top of my extruder, reset it and watch the motion live.
Re: Calibration
March 02, 2013 05:16AM
Oh I get it. Fuck, sorry, I'm bad at math. 1+1=2, right?

Thanks, all.
Re: Calibration
March 02, 2013 06:50AM
I just realised the filament isn't getting pushed through consistently.

The extruding motor works fine, but every couple of seconds the filament will stop-and-go, grinding on the hobbed bolt. I tried tightening the idler block, but the filament would fold or curl up inside the block. It seems the plastic isn't melting fast enough. I tried increasing the temp to 240, but no difference. I tried increasing the temp to 250, but it would automatically stop heating. And the hole through the PTFE is okay; the filament slides easily through it.

The top-end of the brass barrel is halfway up the PTFE. Should I make it go farther up? Am I getting a false reading on the temp?

Thanks.

Edited 1 time(s). Last edit at 03/02/2013 06:58AM by Axis.
Attachments:
open | download - Hot-End.jpg (288.9 KB)
Re: Calibration
March 02, 2013 10:15AM
I assume this is when you are manually feeding during calibration?
Check you r extruder speed and slow it down. Also extrude in 5-10mm amounts until you reach the desired 100mm.
Re: Calibration
March 02, 2013 11:36AM
Spk64 Wrote:
-------------------------------------------------------
> I assume this is when you are manually feeding
> during calibration?
> Check you r extruder speed and slow it down. Also
> extrude in 5-10mm amounts until you reach the
> desired 100mm.


Exactly. You have to consider, unless you manually extrude at a very slow feed rate, it will be going too fast for an extruder to keep up. So you break down the moves into smaller pieces so you don't overwork your extruder, or alternatively slow down the Esteps/min on your manual feed.
Re: Calibration
March 04, 2013 10:27AM
Okay. After doing the calibration, my E_steps_per_unit is now 583 and it works fine manually at 30mm/min.

But when I tried a print, the filament jammed. In Slic3r I decreased the parameters under 'Print speed' to 15, 15, 30, 30, and 30 (half the default), but it still jammed. I decreased the print speed again to 10, 10, 20, 20, and 20, and now it seems to print okay.

Is it typical for a printer with 1.75mm filament and a .3mm nozzle to have to print at this speed?

Thanks.
Re: Calibration
March 07, 2013 09:42AM
Is it typical for a printer with 1.75mm filament and a .3mm nozzle to have to print at this speed?

With my Ordbot with QUBD extruder at 1.75mm, I print cleanly at 150mm/sec and have achieved 180mm/sec I felt like it was going to explode at that speed. For regular printing however, I print at 120+ mm/sec.
Re: Calibration
March 22, 2013 10:35PM
Eric Fisher Wrote:
-------------------------------------------------------
> Is it typical for a printer with 1.75mm filament
> and a .3mm nozzle to have to print at this speed?
>
>
> With my Ordbot with QUBD extruder at 1.75mm, I
> print cleanly at 150mm/sec and have achieved
> 180mm/sec I felt like it was going to explode at
> that speed. For regular printing however, I print
> at 120+ mm/sec.

Post a video Eric ...i have to see printing at 180
At 100 mine threatens to jump out the window
Sorry, only registered users may post in this forum.

Click here to login