Is there a repository of Skienforge/RepG(17,18) settings?

[Martin Bogomolni]

There are lots of MakerBot operators that have great luck printing
using relatively stock makerbots, and even more using makerbots
upgraded with new components (HBP, MK5 feed pulleys, aluminum
idlerwheels, paxtruders..)

We also use a dizzying variety of extruder PID settings, extruder
reversal settings, and even firmwares...

Is there a central repository of working settings that people have
found produce good prints on their hardware? If not, should we start
one on the wiki? For people that use RepG, it's possible to pack up
the settings in the skeinforge/prefs/<profile> directories into
zipfiles and the like, and post them on the wiki...

-Martin

[Sonny Mounicou]

I'm not aware of any, but it would be great if there was. I'll
happily share mine, although I'm much more intrigued to see those of
the well tuned Makerbots. Even with the different configurations, it
would be immensely valuable.

> --
> You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
> To post to this group, send email to make...@googlegroups.com.
> To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.
>
>

[Nick]

I WANT THAT !!!
;-)

I asked for that couple of month ago but didn't had time to push more.
My bot is printing but it need so much calibration. I think that
trying with settings from a bot that work well can be so helpful

Nick
--
http://blog.tekamoi.com

[Nathan Hilderman]

I totally agree, I'm always trying to calibrate - it is such a waste of time. If this could be made easier I'm all for it.

Nathan

[ddurant]

There's always going to be some differences between machines and
filament stock. While a collection of skeinforge profiles isn't a bad
idea, it probably doesn't mean the end of having to tweak yours. Buy
stock in Calibration Cubes because they're going to be around for a
while...

If y'all are having a hard time getting decent prints, don't be shy -
ask! (pictures, if you can, are always good)

On Jul 21, 9:48 pm, Martin Bogomolni <martinb...@gmail.com> wrote:

[Bill Culverhouse]

An exhaustive database of all the settings everyone is using might yield some interesting data. Related to the wisdom of crowds idea descovered by Galton at the turn of the century. 

 

Given a stock extruder the average of the settings used by 1000+ Makerbots are these...

 

Would be interesting to see how various mods change those settings. HBP or none, stepper extruder or standard. .5mm head or a .35mm.

 

Just my $0.02

 

-b 

[Martin Bogomolni]

Okay, when I get home tonight, I'll take the time to create a table in
the wiki with all of them. #1549 doesn't print well, yet .. but I
have settings that, at least, produce good calibration cubes and thin
wall objects. (It does not, however, produce good 'holes' .. holes
and towers are blobby)

-Martin

[Luis E. Rodriguez]

Would posting skeinforge preference files on thingiverse help? My
worry is people with and without calibrated thermistors. Rob from
CCCKC calibrated several bots in the space and they had like a 10-30
degree offset that was corrected with the thermocouple calibration
procedure from the wiki. Just a nice first step to eliminate the out
of control temperature variable.

Luis E. Rodriguez
Creative Services Coordinator

Kauffman Foundation
4801 Rockhill Road
Kansas City MO 64110-2046

816-932-1254
816-751-6873 Fax

*sent from iPhone 4

[ddurant]

A useful column on such a table would be some rating from the person
who created the config on how well they think it works. Not
necessarily how well what they're trying to do will work - more like
how well the config does at getting to that goal. I know I've got
profiles that are dead-on, profiles that are pretty close and profiles
that just need to be tossed and started over from scratch.

This effort is made harder because people use a variety of at least
somewhat incompatible vesions of skeinforge. Some people use 0006,
some use 0007, some use the version built into RepG (0006+ ?) and some
use whatever the latest version is. If people do zip & post their
configs somewhere public, please make it clear which version of
skeinforge it's for.

On Jul 22, 11:31 am, Martin Bogomolni <martinb...@gmail.com> wrote:
> Okay, when I get home tonight, I'll take the time to create a table in
> the wiki with all of them.    #1549 doesn't print well, yet .. but I
> have settings that, at least, produce good calibration cubes and thin
> wall objects.  (It does not, however, produce good 'holes' .. holes
> and towers are blobby)
>
> -Martin
>
> On Thu, Jul 22, 2010 at 10:17 AM, Bill Culverhouse
>
>
>
> <bill.culverho...@gmail.com> wrote:
> > An exhaustive database of all the settings everyone is using might yield
> > some interesting data. Related to the wisdom of crowds idea descovered by
> > Galton at the turn of the century.
>
> > Given a stock extruder the average of the settings used by 1000+ Makerbots
> > are these...
>
> > Would be interesting to see how various mods change those settings. HBP or
> > none, stepper extruder or standard. .5mm head or a .35mm.
>
> > Just my $0.02
>
> > -b
>

> >http://groups.google.com/group/makerbot?hl=en.- Hide quoted text -
>
> - Show quoted text -

[Casey H]

I'm having difficulty with Skeinforge settings as well.  I can print gcode that someone else has created, but when I try to create the gcode myself, it doesn't work.  Knowing what the settings in skeinforge were for the print would be tremendously useful! 

Casey

[Nick]

I agree !

Too much time is needed to make it working well..
I wrote some idea on my blog about that : What should be done to help
us is...

http://blog.tekamoi.com

Nick

> > > >> makerbot+u...@googlegroups.com<makerbot%2Bunsubscribe@googlegroups.c om>
> > .

> > > >> For more options, visit this group at
> > > >>http://groups.google.com/group/makerbot?hl=en.
>
> > > > --
> > > > You received this message because you are subscribed to the Google
> > Groups
> > > > "MakerBot Operators" group.
> > > > To post to this group, send email to make...@googlegroups.com.
> > > > To unsubscribe from this group, send email to

> > > > makerbot+u...@googlegroups.com<makerbot%2Bunsubscribe@googlegroups.c om>
> > .

> > > > For more options, visit this group at

> > > >http://groups.google.com/group/makerbot?hl=en.-Hide quoted text -

>
> > > - Show quoted text -
>
> > --
> > You received this message because you are subscribed to the Google Groups
> > "MakerBot Operators" group.
> > To post to this group, send email to make...@googlegroups.com.
> > To unsubscribe from this group, send email to

> > makerbot+u...@googlegroups.com<makerbot%2Bunsubscribe@googlegroups.c om>
> > .

[Sonny Mounicou]

So how do we propose to solve it? Let's do something about it.

Thinking through solutions we have two paths:

A) Upload them to a wiki/fileshare
B) Put them in a structured database that allows us to search to look
for trends.

There is no reason we can't do A today, but I'm not sure it gives us a
way to discover the most effective settings. We can do B, but it
takes much more time/energy. I propose we start with A tonight. Who
wants to create a landing spot that we can start uploading our
settings to?

> To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.

[ddurant]

Are you guys using the latest ReplicatorG that has the built-in
skeinforge? If not, that's your best bet for getting started.

> >> > > >> makerbot+u...@googlegroups.com<makerbot%2Bunsubscr...@googlegroups.c om>

> >> > .
> >> > > >> For more options, visit this group at
> >> > > >>http://groups.google.com/group/makerbot?hl=en.
>
> >> > > > --
> >> > > > You received this message because you are subscribed to the Google
> >> > Groups
> >> > > > "MakerBot Operators" group.
> >> > > > To post to this group, send email to make...@googlegroups.com.
> >> > > > To unsubscribe from this group, send email to

> >> > > > makerbot+u...@googlegroups.com<makerbot%2Bunsubscr...@googlegroups.c om>

> >> > .
> >> > > > For more options, visit this group at

> >> > > >http://groups.google.com/group/makerbot?hl=en.-Hidequoted text -

>
> >> > > - Show quoted text -
>
> >> > --
> >> > You received this message because you are subscribed to the Google Groups
> >> > "MakerBot Operators" group.
> >> > To post to this group, send email to make...@googlegroups.com.
> >> > To unsubscribe from this group, send email to

> >> > makerbot+u...@googlegroups.com<makerbot%2Bunsubscr...@googlegroups.c om>

> >> > .
> >> > For more options, visit this group at
> >> >http://groups.google.com/group/makerbot?hl=en.
>
> > --
> > You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
> > To post to this group, send email to make...@googlegroups.com.
> > To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.

> > For more options, visit this group athttp://groups.google.com/group/makerbot?hl=en.- Hide quoted text -

[Sonny Mounicou]

I am using 17 that has it built in and I'm pretty happy with my
progress thus far. The most complex item that I've printed is the
whistle. My walls are solid, everything sticks to the HBP. What
plagues me right now is too much ooze which causes strings and some
sloppiness. I've attached a picture to this email for review.

I think the point of a repository was still valid. You would assume
that the default makerbot skeinforge settings were the best choice
across the board, but I don't think that is really the case. It seems
like everyone has to do some major tweaking to the settings to get it
right. Maybe a repository would help speed up that process.

[Martin Bogomolni]

http://wiki.makerbot.com/repg_skienforge_settings

I uploaded my settings, and made a link to the zipfile of the current
cupcake-abs settings. This configuration doesn't work well, but is
part of my attempt to get the makerbot to print with an 0.28mm Z axis
accuracy.

-Martin

[Bot#1559]

Do you think locality might have any partial role in calibration?
Sometimes baking instructions are different for people who live at
different altitudes. If someone in Colorado shipped their machine to
Florida, would it behave differently?

[ddurant]

> I've attached a picture to this email for review.

Looks like it's maybe either too hot or your feed rate is too slow.
Does it seem like it should be moving fast enough to span that gap
across the whistle?

I don't know if I'm for or against a repository of configs..
Skeinforge tormented me mercilessly at the start but once I sorta
figured it out, my build quality doubled almost instantly. In the
month or so since then, it's probably doubled again. If I want to
print an object, I've gotten pretty good at eyeing what sort of
profile it needs (mostly strength vs detail then, secondary, how long
I want it to take) and if I don't already have a profile for that, I
just put one together. Unless we're talking about layer heights under
0.2mm or over 0.5mm, that doesn't take more than an hour.

So, I definitely understand the desire to get a repository together
but such a thing doesn't really teach people what the important bits
of the software are..

</soapbox>

> >> >> > > >http://groups.google.com/group/makerbot?hl=en.-Hidequotedtext -

>
> >> >> > > - Show quoted text -
>
> >> >> > --
> >> >> > You received this message because you are subscribed to the Google Groups
> >> >> > "MakerBot Operators" group.
> >> >> > To post to this group, send email to make...@googlegroups.com.
> >> >> > To unsubscribe from this group, send email to
> >> >> > makerbot+u...@googlegroups.com<makerbot%2Bunsubscr...@googlegroups.c om>
> >> >> > .
> >> >> > For more options, visit this group at
> >> >> >http://groups.google.com/group/makerbot?hl=en.
>
> >> > --
> >> > You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
> >> > To post to this group, send email to make...@googlegroups.com.
> >> > To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.

> >> > For more options, visit this group athttp://groups.google.com/group/makerbot?hl=en.-Hide quoted text -

>
> >> - Show quoted text -
>
> > --
> > You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
> > To post to this group, send email to make...@googlegroups.com.
> > To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
> > For more options, visit this group athttp://groups.google.com/group/makerbot?hl=en.
>
>
>

>  IMG_0007.jpg
> 40KViewDownload- Hide quoted text -

[ddurant]

> Do you think locality might have any partial role in calibration?
> Sometimes baking instructions are different for people who live at
> different altitudes. If someone in Colorado shipped their machine to
> Florida, would it behave differently?

I don't think it has a big impact on the hardware but things like big
differences in air pressure might make the plastic behave
differently..

> > > -Martin- Hide quoted text -

[Martin Bogomolni]

ddurant,

The point is, that operators shouldn't //have// to figure out all the
fiddly bits, when an expert system can predict it for you. Getting
data is the first step in getting a solid set of configurations,
better algorithms, and better control of our MakerBots.

-=Martin

[ddurant]

> The point is, that operators shouldn't //have// to figure out all the
> fiddly bits,

Believe me, I know. This stuff had me pulling my hair out for weeks..
That's why I said I wasn't sure if I was for or against it..

That said, shouldn't operators be able to look at their print and say
"oh.. that's not right. I need to change the (one of the 5-6 things in
skeinforge that really matter) setting some" and have the next one
print come out better?

> >> >> > For more options, visit this group athttp://groups.google.com/group/makerbot?hl=en.-Hidequoted text -

>
> >> >> - Show quoted text -
>
> >> > --
> >> > You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
> >> > To post to this group, send email to make...@googlegroups.com.
> >> > To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
> >> > For more options, visit this group athttp://groups.google.com/group/makerbot?hl=en.
>
> >>  IMG_0007.jpg
> >> 40KViewDownload- Hide quoted text -
>
> >> - Show quoted text -
>
> > --
> > You received this message because
>

> ...
>
> read more »- Hide quoted text -

[Nathan Hilderman]

A couple of other thoughts:
1. An option for simplified skeinfox menu for makerbots with only key fields and example pictures of what they change (e.g. temp, layer hight, speed,..).
2. A configuration wizard to help you configure. Sets up as defaults, and prints an object. Asks you which picture it matches best and adjusts settings, then prints another object...

[Mark Cohen]

repg needs to have a wizard script that asks the right questions and runs a default profile that can be adjjusted based on the answers. the result can become the default skeinforge settings for your machine.

On Jul 23, 2010 9:37 AM, "Nathan Hilderman" <nathan.h...@gmail.com> wrote:

A couple of other thoughts:
1. An option for simplified skeinfox menu for makerbots with only key fields and example pictures of what they change (e.g. temp, layer hight, speed,..).
2. A configuration wizard to help you configure. Sets up as defaults, and prints an object. Asks you which picture it matches best and adjusts settings, then prints another object...

On Thu, Jul 22, 2010 at 11:25 PM, ddurant <ddur...@gmail.com> wrote:
>

> > The point is, that op...

[ddurant]

> 2. A configuration wizard to help you configure.

I've been thinking about this for the last week or so.. Well, I've
been thinking about it for months but a lot more over the last couple
weeks, now that I think I know what I'm doing.

What would you have such a wizard ask about? Things like material type/
size, machine nozzle size and base temperature are (sorta) one-time
hardware setup questions but what questions would you ask the user
about what they want in the print? Would you ask about layer height
and width/thickness ratios or about minimum detail size, model
strength and print speed? What questions would you ask? I'm
interested in what people things about this - I may give writing such
a wizard a shot...


On Jul 23, 9:37 am, Nathan Hilderman <nathan.hilder...@gmail.com>
wrote:

> > > >> >> >> > > >> makerbot+u...@googlegroups.com<makerbot%2Bunsubscribe@googlegroups.c­om>

> > <makerbot%2Bunsubscr...@googlegroups.c om>
> > > >> >> >> > .
> > > >> >> >> > > >> For more options, visit this group at
> > > >> >> >> > > >>http://groups.google.com/group/makerbot?hl=en.
>
> > > >> >> >> > > > --
> > > >> >> >> > > > You received this message because you are subscribed to
> > the Google
> > > >> >> >> > Groups
> > > >> >> >> > > > "MakerBot Operators" group.
> > > >> >> >> > > > To post to this group, send email to
> > make...@googlegroups.com.
> > > >> >> >> > > > To unsubscribe from this group, send email to
>

[Adam Sheesley]

It could work in the same way some 2D printer calibrations work.  Print something, display to the user what it should look like then ask: does it look like this? No? Is it too big? Too stringy? Not sticking? etc. and adjust the settings according to their responses, then print again. Rinse and repeat until satisfied.

To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.

[ddurant]

Yes, the post-print tweak is important.

I was wondering more about pre-print questions to ask..

I've been thinking that the 3 big questions that effect a build are:
- speed (build slow <-> build fast)
- resolution (more detail <-> less detail)
- strength (fragile <-> bulletproof)

Those really impact the important skeinforge bits of layer height,
feed rate and width/thickness. You can have a layer height of 0.4mm
with a w/t of 1.0 and it will print fairly quickly but it won't be
very strong and you won't get any fine details. Or you can do 0.4mm
height with a 1.8 w/t and it will print slower and have even less
horizontal detail but you'll need a hammer to break it. You can juggle
these around any which way (within reason) to change how the object
comes out..

Plus there are other secondary bits to think about
- solidity (hollow <-> solid) (Fill: extra shells, solid surface
layers)
- cleanup time (print faster, more cleanup <-> print slower, less
cleanup) (Comb)
- more?

These don't really impact the 'big 3' settings but are still
important. They're also more standalone - they don't have as much of
an effect on other settings.

I guess I'm just thinking out loud.. I like the idea of a wizard but
am also sticking to my previous statement that people should be able
to look at a bad print and have some idea of what they need to do to
make it better.

[ddurant]

I went back this weekend and updated my normal profiles to go from
adjusting PWM based on the diameter of my filament stock to using 255
for them. It's probably "5.9 vs a half dozen" but bumping up the feed
rate instead of dropping flow rate means things print a bit faster.
Sorta like people told me, when I talked about dropping PWM.. :)

These are all for 0.5mm nozzles + 3.0mm ABS in skeinforge 27, dated
2010-06-28. If your filament is smaller than 3.0mm, drop the feed rate
a bit. I have No idea how to convert these numbers into incantations
that previous versions of skeinforge will be happy with..

Not all of these are perfect, some even less than others, but I
wouldn't call any of them worse than 'not bad'. As layer height
decreases and/or feed rate increases, it's going to be happier with a
machine that's been oiled/tightened/tuned - if you're starting out,
the slow & heavy-set ones are probably easier to catch.

layer height feed rate width/thickness
0.25, 49.50, 1.8
0.25, 60.00, 1.6

0.28, 44.00, 1.8
0.28, 49.75, 1.6
0.28, 54.25, 1.4
0.28, 61.50, 1.2

0.32, 34.25, 1.6
0.32, 42.25, 1.4
0.32, 51.75, 1.2
0.32, 60.50, 1.0

0.35, 34.25, 1.4
0.35, 46.25, 1.0

[Spacexula]

It would be an intesting way for Makerbot to gather data if they had a
volentary check box where you could send them your settings
automatically. That way they could gather statistics from all the
different copies of RepG, if the person decided to allow it. They
could then make that data public.

On Jul 23, 11:10 am, ddurant <dduran...@gmail.com> wrote:

> ...
>
> read more »

[ddurant]

I'm not sure that'd be needed - I think we're getting close to being
able to do a wizard where you can dial in a few preferences and have
it set skeinforge up for you. Especially if we get stepper extruders.

It's a certain diameter filament going through a certain diameter
pinch wheel coming out a certain diameter nozzle. If the rate that all
that gets to be more predictable (aka: stepper), it's not that much of
a leap to then say what shape you want the threads to be (layer height
& w/t) and have it figure out the speed. Or pick speed and layer
height and get w/t. Or whatever. That gets you dials for speed,
strength and resolution. Everything else in skeinforge is (I think)
just a modifer for those things.

If I wasn't so horribly helpless when it comes to math, I'd take the
numbers I posted above and try to write the program myself. Maybe,
once I get a little less sick of printing 20mm cubes, I'll try to
clean up those settings a bit more and see if a formula jumps out at
me.

[Will Brown]

I've started using a repo on Github so that I can have version control of my skeinforge profiles. I'm not sure if it's useful to anyone at this stage, but my repository is here: http://github.com/haldean/makerbot-skeinforge-settings

I've only been doing this for a night, but I've already found it to be incredibly useful. Maybe if there was an established convention for naming skeinforge profile repositories on Github, we could {build off of,use} that?

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

--
Will Brown

6902 CFE2 C49B 17C2 710B  E4AD A789 89EF EA80 319B