SFACT The New and Improved Skeinforge !!

Re: SFACT The New and Improved Skeinforge !!

ahmetcemturan — Wed, 01 Feb 2012 03:30:26 -0500

The packing density or e-steps corrector as I call it now is just correcting the flow.

About acceleration ratio:
please have a look here: [forums.reprap.org]

It only affects E acceleration and setting a low value l20-30) will make it dominant and give you less pressure in the nozzle and more accurate parts.

Re: SFACT The New and Improved Skeinforge !!

zsunsun — Tue, 31 Jan 2012 17:20:54 -0500

I don't have a picture to post right now, but empty area is actually little more than one filament wide, I'm guessing its because I haven't done packing factor calibration?

Also, what does the 'Add Acceleration Rate' check box do? The only thing I can relate to is 'Main Acceleration Rate for Extruder (mm/s2)' but it doesn't say what values are effected by it. Am I correct in assuming it adds acceleration in feed rate instead of instantaneous full speed?

Thanks

Re: SFACT The New and Improved Skeinforge !!

nophead — Tue, 31 Jan 2012 04:26:17 -0500

If a feature is between 2 and 3 times the filament width it can''t be filled as the gap between the outlines is less than one filament. You need to make it exactly two filaments wide or slightly more than 3 to avoid a gap.

Re: SFACT The New and Improved Skeinforge !!

zsunsun — Mon, 30 Jan 2012 23:12:18 -0500

This is amazing. I just followed instruction except for the part where I need to measure thin wall feature (I misplaced my digital caliper) and just went with printing with minor adjustments (cool, inset etc).

And print came out to be excellent all around except for it does not fill 'fingers' in this design,
[www.thingiverse.com]

I just have perimeter with no infill for 'fingers' similar to the this picture,
[www.thingiverse.com]

Also, I noticed my TOM vibrated a whole lot when printing infill, would turning on acceleration help in that case?

Thanks a ton for making SFACT

Re: SFACT The New and Improved Skeinforge !!

rhmorrison — Tue, 10 Jan 2012 10:30:59 -0500

You don't need to activate homing, put the gcode in your start.gcode file.

Re: SFACT The New and Improved Skeinforge !!

oli491 — Tue, 10 Jan 2012 09:41:36 -0500

RobertB Wrote:
-------------------------------------------------------
> I have done several successful prints with Slic3r.
> However, I am having troubles getting the top of
> a polyhedron to print well (slope just slightly
> off horizontal, and small), and so looked into
> alternatives. It seemed "cool" and "skin" in
> SFACT/skeinforge could help me.
>
> So, I got the latest SFACT package from github,
> followed the README (I think), got it to make
> gcode, and:
>
> 1. No home. After this happened the first time,
> I saw there was a "home" tab, and sure enough, the
> box was unchecked. Fixed that, and still no
> home.
>
> 2. No activation of the extruder motor
> whatsoever.
>
>
> It's not some new malfunction...doing a Slic3r
> print now.
>
> This is a vanilla Prusa, using Sprinter/Printrun.

The home thing is that you have to write your own home script. That script can added in -> Users-> .Skeinforge -> alteration

Some programs, like Mach3, are having an own G-code for homing (in Mach 3 it is G28)

If you want to get your motor running you should check the Dimension tab (if you are using the 4th axis for the extruder)

Re: SFACT The New and Improved Skeinforge !!

RobertB — Mon, 09 Jan 2012 21:28:13 -0500

I have done several successful prints with Slic3r. However, I am having troubles getting the top of a polyhedron to print well (slope just slightly off horizontal, and small), and so looked into alternatives. It seemed "cool" and "skin" in SFACT/skeinforge could help me.

So, I got the latest SFACT package from github, followed the README (I think), got it to make gcode, and:

1. No home. After this happened the first time, I saw there was a "home" tab, and sure enough, the box was unchecked. Fixed that, and still no home.

2. No activation of the extruder motor whatsoever.

It's not some new malfunction...doing a Slic3r print now.

This is a vanilla Prusa, using Sprinter/Printrun.

Re: SFACT The New and Improved Skeinforge !!

galatorg — Sat, 31 Dec 2011 05:31:32 -0500

well, still trying to figure it out, but realized the gcode i attached....kinda worked anyways. i had added the G28 gcode option to it....something is still a little whacked, but i'm getting closer anyways. thanks in advance for any help.

Re: SFACT The New and Improved Skeinforge !!

galatorg — Sat, 31 Dec 2011 04:58:37 -0500

This looks awesome, thanks! However, I am having some issues, and I'm really not too sure what to change. I had done a bunch of printing with slic3r, but I'm looking for something better, and just hadn't gotten to config SF. I just came across this and plan on using it. I have everything all tied in correctly (it was pretty good right off the bat). I am using a 0.5mm nozzle, which is what it's pretty much set for already. I could not find the filament packing density box mentioned in the text file, I know it's usually in the DIMENSION plugin, but I don't see it there. Either way, I don't see that related to my problem. I only made it as far as trying to print the initial _40x10 block. I'll include the gcode, but it jumps way up to like 15mm on the Z axis and starts printing that high. in the preview box it just looks like it's only trying to do the walls in a few places. since i didn't see this problem mentioned I assume I am doing something stupid? maybe it's time for bed...

Re: SFACT The New and Improved Skeinforge !!

droftarts — Tue, 27 Dec 2011 11:20:15 -0500

ahmetcemturan Wrote:
-------------------------------------------------------
> You should always delete old profiles with a new version..

Thanks for you quick replies, ahmetcemturan. Each time I update SFACT, I update Printrun at the same time. I delete the whole Printrun folder, which has SFACT inside it. I download the latest master of each from github. Unless SFACT or Printrun hide settings somewhere else as well - which, in the case of Printrun, I think it does, but I don't know where - I set up fresh profiles each time. I notice the problem of the Quick Settings is reported by other people in the Printrun github: [github.com]

Can you answer my other question? Does SFACT only connect inner loops to outer loops if the outer loop is printed first?

Re: SFACT The New and Improved Skeinforge !!

ahmetcemturan — Tue, 27 Dec 2011 10:30:25 -0500

You should always delete old profiles with a new version..

Re: SFACT The New and Improved Skeinforge !!

droftarts — Tue, 27 Dec 2011 05:54:51 -0500

ahmetcemturan Wrote:
-------------------------------------------------------
> You should increase the minimum move distance after retraction.

Thanks, ahmetcemturan, but I have already tried this, and it is currently set to 5.0 in 'Minimum Travelmove after Retraction (millimeters)'
Might it be because I have the print order as 'Loops > Perimeter > Infill'? I find I get neater overhangs printing this way.
I thought connecting perimeter loops was controlled by 'Clip', 'Threshold for connecting inner loops (ratio)'. I currently have this set to '10'. Does it only connect inner loops to outer loops if the outer loop is printed first?

Another small issue - I have updated Printrun and SFACT to the latest versions, but the SFACT Quick Settings option seems to have broken. Some settings are missing, eg there is nothing listed under 'Raft', and changing other settings results in the following message in the Terminal window:

Saving ... Layer Height = Extrusion Thickness (mm): = 0.3
Traceback (most recent call last):
  File "/Applications/REPRAP/kliment-Printrun-a4b3751/SkeinforgeQuickEditDialog.py", line 134, in OnSave
    settings.saveRepository(repo)
  File "/Applications/REPRAP/kliment-Printrun-a4b3751/skeinforge/fabmetheus_utilities/settings.py", line 484, in saveRepository
    setting.setToDisplay()
  File "/Applications/REPRAP/kliment-Printrun-a4b3751/skeinforge/fabmetheus_utilities/settings.py", line 1754, in setToDisplay
    self.value = ( self.latentStringVar.getString() == self.radiobutton['value'] )
  File "/Applications/REPRAP/kliment-Printrun-a4b3751/skeinforge/fabmetheus_utilities/settings.py", line 1395, in getString
    return self.getVar().get()
  File "/Applications/REPRAP/kliment-Printrun-a4b3751/skeinforge/fabmetheus_utilities/settings.py", line 1400, in getVar
    self.stringVar = Tkinter.StringVar()
  File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/lib-tk/Tkinter.py", line 251, in __init__
  File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/lib-tk/Tkinter.py", line 182, in __init__
AttributeError: 'NoneType' object has no attribute 'tk'

This is on Mac OS X 10.6.8. Are there some hidden settings hung over from an earlier version of Printrun/SFACT that need to be deleted?

Re: SFACT The New and Improved Skeinforge !!

ahmetcemturan — Tue, 27 Dec 2011 04:17:18 -0500

You should increase the minimum move distance after retraction.

Re: SFACT The New and Improved Skeinforge !!

droftarts — Mon, 26 Dec 2011 14:09:24 -0500

I'm having a small issue with the current version of SFACT. When doing perimeters, it does the inner shell (assuming just 1 extra shell), then a small retraction/extension while it moves the tiny distance to the perimeter/outer shell, before doing the outer shell. This retraction leaves a small blob on the perimeter, on the outside of an object. I'm sure in previous versions there was no retraction between inner shell and outermost shell/perimeter. Any setting I could check? It's one of those annoying things you don't notice in the gcode until it starts to print.

Re: SFACT The New and Improved Skeinforge !!

Andrew Diehl — Thu, 22 Dec 2011 11:14:53 -0500

I retract 1.75mm filament by .5mm at 20mm/s.

The reverse amount has a lot to do with the distance is between nozzle tip and the pinch wheel.

Longer distance, longer (and faster) retract necessary.

Re: SFACT The New and Improved Skeinforge !!

nophead — Thu, 22 Dec 2011 07:12:51 -0500

I rewind between 0.75 and 1.5mm.

1.75mm filament will need to rewind about 3 times further (and three times faster).

I don't have any experience of Bowden extruders, they presumably need a bit more rewind.

Re: SFACT The New and Improved Skeinforge !!

Andke — Thu, 22 Dec 2011 05:49:51 -0500

ok, that sounded nasty, please suggest sane defaults :)

Re: SFACT The New and Improved Skeinforge !!

nophead — Thu, 22 Dec 2011 04:32:42 -0500

You don't want to reverse more than about 2mm! Otherwise you will suck air into the melt chamber which allows the plastic to oxidise and go brown. It will form a hard layer on the inside that eventually blocks the flow because it no longer melts and isn't dissolved by solvents..

Re: SFACT The New and Improved Skeinforge !!

Andke — Thu, 22 Dec 2011 02:50:29 -0500

thanks - and one minor default suggestion; I noticed that the default value for reversing the extruder was 100mm , -most extruders will end up with stuck hot filament is reversed that far - for me, a reliable max reverse length is more like 20mm.

Re: SFACT The New and Improved Skeinforge !!

martinprice2004 — Wed, 21 Dec 2011 18:40:15 -0500

Thanks for depeloping sfact, it works much beter than skeinforge, which seems to constantly change and put you back to square one all the time!

I just want to report an error in the "multiply" field. Setting the value to say 20,20 moves the part 20mm toward the centre of the build platform, but putting unequal values in scales the X and Y of the gcode, so a square part becomes rectangular, however this isn't shown in the visual output. This doesn't seem right somehow.

Thanks

Re: SFACT The New and Improved Skeinforge !!

ahmetcemturan — Sat, 17 Dec 2011 03:51:01 -0500

Pls download current daily version of SFACT from Git.
The default settings should work fine with your nozzle..

Read the SFACT FAQs there is a lot of information.....

Re: SFACT The New and Improved Skeinforge !!

Andke — Fri, 16 Dec 2011 06:08:16 -0500

Can a prusa mendel owner with wades extruder and .5mm nozzle please post a good PLA profile for SFACT 2.1. 11.08.11 ?

and/or confirm that sprinter software is using this setting?:
// X, Y, Z, E steps per unit - Metric Prusa Mendel with Wade extruder:
float axis_steps_per_unit[] = {80, 80, 3200/1.25,700};

Re: SFACT The New and Improved Skeinforge !!

brnrd — Thu, 15 Dec 2011 22:25:58 -0500

Andke Wrote:
-------------------------------------------------------

> When I ask pronterface to extrude 5mm, then
> exactly 5mm filament disappears into the extruder
> (measured)
> - still, if my packing ratio is 1.0 the single
> wall test is over 1mm thick.
> is there anything else that can be "wrong" ?
>

5 mm is too small to measure accurately. You need to extrude longer than that. I use between 25 and 100 mm for calibration. One quick test that I like to do is to mark the filament flush with the extruder top. Then retract by 25 mm and measure how much came up. Then extrude the same amount to put it back to the mark, change the steps per mm and repeat the procedure again.

Re: SFACT The New and Improved Skeinforge !!

rhmorrison — Thu, 15 Dec 2011 13:49:00 -0500

O.K., sometimes I forget I'm not in the German Language forum... 8-)

Re: SFACT The New and Improved Skeinforge !!

Andke — Thu, 15 Dec 2011 13:28:50 -0500

"Luke, use the source!"
Bob .. let's use English ! :)

Re: SFACT The New and Improved Skeinforge !!

rhmorrison — Thu, 15 Dec 2011 13:27:22 -0500

You cant measure the hot end temperature reliably with an IR thermometer you need to use a thermocouple.

Re: SFACT The New and Improved Skeinforge !!

Andke — Thu, 15 Dec 2011 13:14:49 -0500

I don't get it.
My filament diameter is set (correctly) to 2.8mm
measured width of extrusion (single wall object) IS 0.5mm
I did configure the Sprinter as:
// X, Y, Z, E steps per unit - Metric Prusa Mendel with Wade extruder:
float axis_steps_per_unit[] = {80, 80, 3200/1.25,700};
When I ask pronterface to extrude 5mm, then exactly 5mm filament disappears into the extruder (measured)
- still, if my packing ratio is 1.0 the single wall test is over 1mm thick.
is there anything else that can be "wrong" ?

BTW: forgot to mention; I use an industrial IR thermometer with -35..+250degree range with max 2% error., not an "ear-thermometer"

Re: SFACT The New and Improved Skeinforge !!

brnrd — Thu, 15 Dec 2011 11:15:18 -0500

Note that you have defeated the convenience of the volumetric method that SF and SFACT use by playing around with the filament packing density and the flow rate. If you have set the firmware steps per mm for the extruder correctly, then there's no reason for the filament packing density to to be set to any other value but 1 and for the flow rate to equal the feed rate. With those set correctly, there's no need to play around with other parameters to get the infill correct and the width to come out exactly as entered. The sw does the work for you.

Re: SFACT The New and Improved Skeinforge !!

Andke — Thu, 15 Dec 2011 10:49:33 -0500

Thanks for the advise, I'll comment in between lines.

> First of all, start by calibrating the steps per
> mm for your extruder in the firmware by measuring
> the actual filament pulled and adjusting the
> firmware value so that it matches what you asked

-That's done it's very precise.

> for. Measure maximum and minimum diameter around
> your feed and take the average. Set SFACT or SF to
> this value. And set filament packing density to
> 1.

the 0.5mm thin wall prints very fine, but with filament packing at 2.3 , 1.0 would give over 1mm thick wall.
I think my nozzle is 0.5mm.

> Set the extrusion width to at least 1.5 times the
> extrusion height. Note that the range of extrusion
> height is limited by your extruder nozzle

let's say the ex.width is today 0.5 and height is 0.4 - - increasing the width would give gaps, should I lower the height then ? - sorry - I don't get it :)

> diameter. The cross-sectional area of the extruded
> filament based on the height and thickness must be
> less than the cross sectional area of your nozzle
> so that the filament stretches a little bit or it
> will sag when going through bridges. At 0.38 mm
> height, you should be using a 0.5 mm nozzle.

I am using a 0.5 mm nozzle.

> Set infill overlap over perimeter to 0.15-0.2.

Yes, this is interesting, and inconsistent with Wiki - wiki says "Infill Perimeter Overlap (*ratio*)" and says "1 gives an overlap of one complete extrusion width. A value of slightly more then 0 is recommended for a good bonding of the two."

one complete overlap seems good ?
SFACT settings calls it "Infill Perimeter Overlap (*scaler*)" - and my is 1.0 - because I thought I were supposed to get one complete overlapping - but is seems getting a true overlapping is at low value of 0.15 ? (that would make sense, as that's what I miss.)

> Also, how did you calibrate the nozzle temperature
> with an IR thermometer? The sensitive zone for
> most of those thermometers are too loarge for a
> nozzle. It also measures surface temperature which
> is not the same as the melt zone where the
> filament is.

Re: SFACT The New and Improved Skeinforge !!

brnrd — Thu, 15 Dec 2011 09:27:12 -0500

First of all, start by calibrating the steps per mm for your extruder in the firmware by measuring the actual filament pulled and adjusting the firmware value so that it matches what you asked for. Measure maximum and minimum diameter around your feed and take the average. Set SFACT or SF to this value. And set filament packing density to 1.

Then in SFACT or Skeinforge, set the feed rate and the flow rate equal to each other. your choice of starting with 20 mm/s is good. You can set the perimeter feed rate to 20 also and the perimeter flow rate scaler to 1.

Set the extrusion width to at least 1.5 times the extrusion height. Note that the range of extrusion height is limited by your extruder nozzle diameter. The cross-sectional area of the extruded filament based on the height and thickness must be less than the cross sectional area of your nozzle so that the filament stretches a little bit or it will sag when going through bridges. At 0.38 mm height, you should be using a 0.5 mm nozzle.

Set infill overlap over perimeter to 0.15-0.2.

Also, how did you calibrate the nozzle temperature with an IR thermometer? The sensitive zone for most of those thermometers are too loarge for a nozzle. It also measures surface temperature which is not the same as the melt zone where the filament is.