Introducing Sailfish (faster than a Marlin)

[Jetty]

Today, we're excited to announce the pre-release of Sailfish for
Replicator 1 / ThingOMatic for tinkerers.


** Why the name change?

Sailfish is the next release of Jetty Firmware for the Replicator 1
and ThingOMatic. Jetty firmware received it's name from the
users at the beginning, this was when Jetty was the main
developer working on it so it fitted. But shortly after it's
release,
Dan joined the team and it represents mainly the work of two
developers.

Sailfish provides a major (5-10 times) speed improvement to
acceleration over Marlin and Jetty firmware. With the name
Sailfish, we're paying homage to the origin of the acceleration
part of the project (Marlin). Both Sailfish and Marlin are ocean
fish, and a Sailfish is faster than a Marlin.


** What's the difference in acceleration between firmwares?

Sailfish represents the state of art in an 8 bit processor. In
our opinion, it's better than Marlin (due to the increase in
planning speed, bug fixes, JKN Advance, Yet Another Jerk).
On Replicator, it's also faster, produces higher quality prints
compared to
5.5 and 6.0. On ToM, it's the next step up in print quality
compared to Jetty 3.4/3.5.

The accelerated capabilities of this firmware can be considered
to be higher performing than the other printers out there,
including Ultimaker.


** What's the resolution and speed?

First off, we haven't pushed the limits to the ridiculous (you
can likely do better than us with a little work), but we've
tested the following:

Replicator, down to 0.01 Micron layer heights in ABS using
PrintOMatic and minimal tuning. Travel Speeds up to 340mm/s
on the X axis. Printing up to 150mm/s. We routinely print at
120mm/s and 200mm/s Travel (limited by the Replicator's
mechanics )

ToM, down to 100 Micron layer heights using PrintOMatic
and minimal tuning. We routinely print at 120mm/s and
150mm/s Travel.


** Volumetric 5D and SF50?

On both ThingOMatic and Replicator 1, all our testing has
been with SF50 and Volumetric 5D. On ToM, when using
Sailfish, we now recommend and that you use SF50 and 5D,
not legacy SF35.



** Other features

Replicator 1:
- fast menus with top/bottom wrap around
- jogging menu improvements
- long scrolling filenames in build menu
- Filament usage odometer and trip counter
- override gcode temperature
- 4 custom material profiles (e.g. ABS, PLA etc.)
- Pause @ ZPos and Pause
- Switch off heaters during pause
- Build Z Position in build stats
- Home offsets adjustment
- Ditto printing (Duplicate printing of a
small object to 2 nozzles via firmware)
- Reminder Beep on filament load

ToM:
- everything that was in 3.4 / 3.5
- Ditto printing and dual extruder support (untested)
- No more MCode configuration for non LCD-users
- New Sailfish driver in RepG to replace makerbot4g

Both:
- Machine settings automatically sent to bot from the xml
- Acceleration configuration via Onboard Preferences in RepG
- Fine and Draft quality settings
- Eeprom menu to save / read and erase Flash eeprom
- Part of the acceleration calculations are pre-calculated in RepG


** Is it Open Source?

Of course, it always has been. Source code can be found here:
https://github.com/jetty840/Sailfish-MightyBoardFirmware
https://github.com/jetty840/Sailfish-G3Firmware


** Where can I get support?

Help / Support can be found over on the Jetty Forum:
http://groups.google.com/group/jetty-firmware


Is this firmware supported by Makerbot Industries?

No, this is not an MBI official release and we don't work for MBI.
Although they will likely interoperate to some degree, we recommend
writing down your bot's settings before upgrading for input into
Sailfish
after installation. Once in Sailfish, you can save your firmware's
settings
to SDCard.


** Is Makerware supported?

Currently, not all of Makerware is Open Source, so no, but we hope
to have support in the future. You can still "plate" in Makerware,
and
export to stl, and then load it RepG.


** ReplicatorG

Sailfish requires use of a custom ReplicatorG 0039 which is called
ReplicatorG 0039 - Sailfish to generate the new acceleration
commands.


** Where can I get it?

See http://www.thingiverse.com/thing:32084 for details

[Shawn]

I had the incredible luck to be able to test out this code ahead of
time. This is the main reason I've been resisting switching to the 6.0
firmware - the beta code I received made a world of difference for me,
and seeing the bugs being reported regarding 6.0 I decided to stay with
what was working.

But I'll be upgrading to Sailfish this weekend, I'm sure.

Thanks to Jetty, Dan, and anyone else who contributed to this code!!!

Shawn

[Jetty]

Checkout the picture in the Thing:
http://www.thingiverse.com/thing:32084
120mm/s accelerated at 100Micron.
Using Printomatic and everything at defaults, with
deprime = 0 and starting the layer height 0.05mm above normal.

Text at high resolution and 120mm/s accelerated is one of the
hardest things to get printing well on firmwares. That picture
is as it prints on a Replicator 1 with Sailfish, no post processing.

Also, here's a 10Micron (0.01mm) print I did a few weeks ago, also
Printomatic and settings same as above.

http://www.flickr.com/photos/75199291@N03/8004474571/in/photostream

Also, DItto Printing is pretty useful. If the object your printing
can fit between the
nozzles, you can print 2 copies of it at the same time simply by
switching the
"Ditto Print" setting on in the firmware menu (General Settings).
You can print the original to either extruder, and Ditto Print will
figure out which extruder
isn't being used, and print the 2nd copy to that one.
Very useful if you need to print lots of small objects in half the
time.

[delsydsoftware]

Sweet, I've gotta check this out. I was getting pretty close to 10 micron prints with the bowden tube setup. But, that was at 25m/s. Hopefully, this will push it over the top.  :)

[Cymon]

Makerbot releases firmware 6.0 and I think "Eh, I'll wait until the official announcement in case there are any bugs." Sure enough there are bugs and I'm feeling totally justified. Jetty released a new experimental firmware and I've hit download before I finish reading what it can do. 

[Doogiekr]

Im running test prints with the Jetty firmware now on my Rep1... so far incredible...

[Cymon]

Before I really commit I'd like to hear someone comment on dual extrusion printing. Does it work and are the print heads settings correct?

[Shane Graber]

Will Cupcakes be supported?

Shane

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/Xxo8gjw4oYQJ.
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.

[Doogiekr]

The example image was on the thingiverse page was printed by Jetty on a Rep1 using dual extrusion at 120mm/s .1mm layers... so yeah I think it works

[Doogiekr]

Shane - according to the Sailfish firmware wiki here http://wiki.makerbot.com/sailfish-firmware-installation-guide

 

 

Machine Requirements

The Sailfish firmware described here runs on MightyBoard, Gen 4 and Gen 3 electronics. For Replicators, single and dual extrusion is supported.

• Replicator 1s: Makerbot MightyBoard revE or later electronics with one or two extruders.
• Thing-o-Matics: a Makerbot v2.X motherboard, a Makerbot v3.6 Extruder Controller, Makerbot v3.3 Stepper Drivers, a stepper-based extruder, and an Arduino Mega or Arduino Mega 2560. Other than the extruder, these are all "stock" Thing-o-Matic electronics; i.e., Gen 4 electronics. Stepper-based extruders include the Mk6+, Mk7, and Mk8. Other electronics and extruders may or may not work with the Sailfish firmware.
• Cupcakes with Gen 4 electronics: As per the above.
• Cupcakes with Gen 3 electronics: A 3G5D Shield or the "Ugly Cable Hack" is required in order to control a stepper-based extruder. Additionally, the remaining complement of Gen 3 electronics: RepRap Motherboard v1.2, Plastruder Extruder Controller, and v2.3 or later stepper drivers.

[Martin Galese]

I've been testing betas of this firmware and I have to say it has been rock solid and incredible.  The quality, as you can see, is obviously stellar, even at ridiculously high speeds.  

One point I haven't seen made is that this also makes your Replicator far more quiet, particularly at more normal speeds.  I love that, since my Replicator is in a home office I share with my bemused-by-my-obsession wife!

Great work Jetty and Dan!  I think the name change will help Dan get the credit he also deserves!

Thanks again,

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/Kjkc6aoGXh8J.

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.

--
Martin Galese

[Dan Newman]

The issue with Cupcakes is that neither Jetty or I have one on which to test.
Consequently, this release has seen essentially no testing on Gen3 hardware.
So, if you are a Cupcake + 3G5D owner, then please use with the willingness
to be a beta tester -- we welcome the feedback and working with you to ensure
that the code is working on Cupcakes.

Thanks & Regards,
Dan

P.S. I passed my Cupcake CNC on to someone else who rebuilt it and installed
RAMPS + Pololu electronics.

On 13 Oct 2012 , at 6:01 AM, Doogiekr wrote:

> Shane - according to the Sailfish firmware wiki here
> http://wiki.makerbot.com/sailfish-firmware-installation-guide
>
> Machine Requirements
>
> The Sailfish firmware described here runs on MightyBoard, Gen 4 and Gen 3
> electronics. For Replicators, single and dual extrusion is supported.
>

> - *Replicator 1s:* Makerbot MightyBoard revE or later electronics with
> one or two extruders.
> - *Thing-o-Matics:* a Makerbot v2.X motherboard, a Makerbot v3.6

> Extruder Controller, Makerbot v3.3 Stepper Drivers, a stepper-based
> extruder, and an Arduino Mega or Arduino Mega 2560. Other than the
> extruder, these are all "stock" Thing-o-Matic electronics; i.e., Gen 4
> electronics. Stepper-based extruders include the Mk6+, Mk7, and Mk8. Other
> electronics and extruders may or may not work with the Sailfish firmware.

> - *Cupcakes with Gen 4 electronics:* As per the above.
> - *Cupcakes with Gen 3 electronics:* A 3G5D Shield or the "Ugly Cable

> Hack" is required in order to control a stepper-based extruder.
> Additionally, the remaining complement of Gen 3 electronics: RepRap
> Motherboard v1.2, Plastruder Extruder Controller, and v2.3 or later stepper
> drivers.
>
>
>
> On Saturday, October 13, 2012 9:44:46 PM UTC+9, Shane Graber wrote:
>
>> Will Cupcakes be supported?
>>
>> Shane

>> On Oct 13, 2012 8:40 AM, "Cymon" <joeal...@gmail.com <javascript:>> wrote:
>>
>>> Before I really commit I'd like to hear someone comment on dual extrusion
>>> printing. Does it work and are the print heads settings correct?
>>>
>>> On Saturday, October 13, 2012 6:37:45 AM UTC-6, Doogiekr wrote:
>>>>
>>>> Im running test prints with the Jetty firmware now on my Rep1... so far
>>>> incredible...
>>>>
>>>>
>>>> On Saturday, October 13, 2012 9:30:41 PM UTC+9, Cymon wrote:
>>>>
>>>>> Makerbot releases firmware 6.0 and I think "Eh, I'll wait until the
>>>>> official announcement in case there are any bugs." Sure enough there are
>>>>> bugs and I'm feeling totally justified. Jetty released a new experimental
>>>>> firmware and I've hit download before I finish reading what it can do.
>>>>>
>>>>> On Saturday, October 13, 2012 2:04:11 AM UTC-6, delsydsoftware wrote:
>>>>>>
>>>>>> Sweet, I've gotta check this out. I was getting pretty close to 10
>>>>>> micron prints with the bowden tube setup. But, that was at 25m/s.
>>>>>> Hopefully, this will push it over the top. :)
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Saturday, 13 October 2012 03:42:50 UTC-4, Jetty wrote:
>>>>>>>
>>>>>>> Checkout the picture in the Thing:

>>>>>>> http://www.thingiverse.com/**thing:32084
>>>>>>> 120mm/s <http://www.thingiverse.com/thing:32084120mm/s> accelerated

>>>>>>> at 100Micron.
>>>>>>> Using Printomatic and everything at defaults, with
>>>>>>> deprime = 0 and starting the layer height 0.05mm above normal.
>>>>>>>
>>>>>>> Text at high resolution and 120mm/s accelerated is one of the
>>>>>>> hardest things to get printing well on firmwares. That picture
>>>>>>> is as it prints on a Replicator 1 with Sailfish, no post processing.
>>>>>>>
>>>>>>> Also, here's a 10Micron (0.01mm) print I did a few weeks ago, also
>>>>>>> Printomatic and settings same as above.
>>>>>>>

>>>>>>> http://www.flickr.com/photos/**75199291@N03/8004474571/in/**
>>>>>>> photostream<http://www.flickr.com/photos/75199291@N03/8004474571/in/photostream>

>>>>>>>
>>>>>>> Also, DItto Printing is pretty useful. If the object your printing
>>>>>>> can fit between the
>>>>>>> nozzles, you can print 2 copies of it at the same time simply by
>>>>>>> switching the
>>>>>>> "Ditto Print" setting on in the firmware menu (General Settings).
>>>>>>> You can print the original to either extruder, and Ditto Print will
>>>>>>> figure out which extruder
>>>>>>> isn't being used, and print the 2nd copy to that one.
>>>>>>> Very useful if you need to print lots of small objects in half the
>>>>>>> time.
>>>>>>>
>>>>>> --
>>> You received this message because you are subscribed to the Google Groups
>>> "MakerBot Operators" group.
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msg/makerbot/-/Xxo8gjw4oYQJ.

>>> To post to this group, send email to make...@googlegroups.com<javascript:>
>>> .

>>> To unsubscribe from this group, send email to

>>> makerbot+u...@googlegroups.com <javascript:>.

>>> 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 view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/jNBj_BX5k8MJ.

[Jetty]

We asked for some Beta testers for Cupcake on the Jetty Forum a few
days ago.
If you want to see it on cupcake, we need you to test, please check
out the
message and help us out. As Dan said, we don't have any cupcake
equipment to
test on. I've created a cupcake simulator which I can run on my ToM,
but there's
nothing like having it tested on the real hardware. So we're looking
to the community
for cupcake testing.

[Jetty]

The demo print on the thing was with Dual Extrusion and dual extrusion
has been tested.

The dual extrusion in Sailfish was based on the way it's done in 5.5,
(not 6.0). So you
should be okay. If not, just erase your eeprom (using the new
Eeprom Erase function in Jetty), and everything will be set to
defaults. Configure your number of
Tool Heads, run through the First Run Experience and you should be
good to go. Then you run
through the nozzle calibration in Utilties.

Then, using the Eeprom menu, save your Eeprom settings to SD Card.
Then if you play around,
change things, you can always restore them.

The Eeprom menu is available on Replicator 1 and ToM (with LCD)

[Joseph Chiu]

Wow.  That print is amazing.  The top surface of the tag is so flat!

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

[Jetty]

This one is an interesting technical point. One way we found out that
we could measure
performance / quality of the acceleration in Sailfish was by how it
sounded, which makes sense
when you think about it.

If interrupts are firing when they should, you'll hear "1 Note" from
the steppers. If they're
not, you'll hear multiple notes and harmonics.

Also, vibration is cut down.

So you'll notice overall in Sailfish that it sounds quieter, and will
be one tone.

We've had the X axis up to 340mm/s, before the bearings in the X axis
couldn't keep up.
Now the extruder can't handle 340mm/s (I've tested the extruder to
150mm/s, before
you start getting slippage). But the kickstarter extruder project with
more grip we think will help
break that 150mm/s barrier with the stock extruder. Also if you have
the minimalistic
extruder replacement, we think that should help break 150mm/s too, so
we're interested
to see what you find. (we tend to keep to stock setup's during
development for testing purposes)

And the firmware didn't miss a beat, even at 340mm/s. I'll post a
video of it from our testing
shortly.

[Jetty]

Here's a video of 340mm/s on the X axis testing speed limits of the
firmware
(the hardware gave up first by the way):

http://www.youtube.com/watch?v=0F9TjBnNvkE&feature=youtu.be

(recommend watching it in HD)

Bearings started skipping at 340mm/s. The scary thing is if there
was a mechanical failure at those speeds, watch out any breakables
nearby,
that's a fairly large amount of mass moving around at high speed. :-)

[ddurant]

> ** What's the resolution and speed?
>

> ...Replicator, down to 0.01 Micron layer heights in ABS

 

That's pretty impressive! I'd like to see a print done at that layer height..

 

:P

[Dan Newman]

On 13 Oct 2012 , at 12:17 PM, ddurant wrote:

>> ** What's the resolution and speed?
>>
>> ...Replicator, down to 0.01 Micron layer heights in ABS
>
> That's pretty impressive! I'd like to see a print done at that layer
> height..

Yes, that is ;) Jetty meant 0.01 mm layer heights (10 microns).

You'll see that in a couple of postings later, Jetty sent a link to
a Flickr photo of a squirrel (13 October 2012 12:42:47 AM PDT was
the date on the e-mail I received).

That was back a few weeks ago before we had finished tuning the
software and was using defaults except for turning all flavours
of reversal/deprime off. At those layer heights and with ABS,
so little plastic is being extruded per mm of x/y/z distance, that
reversal/deprime doesn't do much good.

Also at such small layer heights, you start seeing the downside
of ABS's higher viscosity relative to PLA: it starts behaving
like chunky peanut butter. Some folks I know for whom ABS
extrusion/molding is their stock-and-trade were actually a bit
surprised that ABS worked at all at those layer heights. At any
rate, I'd put my money on PLA working better overall at

Dan

[Jetty]

And I think the point is, that it was very easy to do. It was
my first foray into high resolution printing, and I was
totally surprised when it worked with PrintOMatic, I had
expected to spend hours tuning SF50.

The issue is that at those small flow rates, everything has to
work perfectly from a firmware perspective, otherwise the
extrusion will break up. That's where JKN Advance and YAJ
come in, to solve that issue that other firmwares generally don't.

[ddurant]

> I had expected to spend hours tuning SF50.

 

If you spend even one hour 'tuning' skeinforge with volumetric 5d, you're doing it wrong..

 

It really is a better solution than printomatic. Even the guy who came up with the original idea and code behind printomatic thinks that. :)

 

On Saturday, October 13, 2012 3:35:54 PM UTC-4, Jetty wrote:

And I think the point is, that it was very easy to do.  It was
my first foray into high resolution printing, and I was
totally surprised when it worked with PrintOMatic, I had
expected to spend hours tuning SF50.

The issue is that at those small flow rates, everything has to
work perfectly from a firmware perspective, otherwise the
extrusion will break up.  That's where JKN Advance and YAJ
come in, to solve that issue that other firmwares generally don't.

On Oct 13, 1:26 pm, Dan Newman wrote:

[Dan Newman]

On 13 Oct 2012 , at 1:27 PM, ddurant wrote:

>> I had expected to spend hours tuning SF50.
>
> If you spend even one hour 'tuning' skeinforge with volumetric 5d, you're
> doing it wrong..
>
> It really is a better solution than printomatic. Even the guy who came up
> with the original idea and code behind printomatic thinks that. :)

I wonder who that might be….

Dan

[Z LeHericy]

Is slic3r supported with Sailfish?

Ever since i switched to Slic3r, i've vowed not to go back to skeinforge...

Slic3r is several orders of magnitude faster at slicing up parts, and is easier (in my opinion) to configure right on the first try.

-Zeno LeHericy

//((=:Z:=))\\
INVENTIONS
Technologies
zinventions.com

Dan

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

[Martin Galese]

Zeno,

I'm  playing around with Slic3r + Sailfish now and for the same reasons.  From talking with Jetty, it seems like there are two different things that the Sailfish ReplicatorG does that Slic3r doesn't.  The first is that there's a new instruction in the s3g that helps the firmware with movement planning.  But that seems to work fine if you load the Slic3r gcode into the new RepG and build your s3g as normal.

The second thing is that there is a new M code to turn acceleration on and off.  The altshell plugin uses this in order to turn off acceleration for a nicer surface finish.  Currently, Slic3r isn't going to generate those.  However, I've had very good luck just using Scli3r's normal support to reduce the print speed for exterior perimeters.

I'm still testing with Slic3r, but so far so good!

(FYI, I'm using the current dynamic-flow branch out of git for my testing, since that what I had been playing with earlier.  YMMV with the various actually released versions.)

--
Martin Galese

[Jetty]

We've have a volunteer looking at that.

[Jetty]

I said I expected to, didn't say I did.

Spent zero time on Skeinforge, did it all from PrintOMatic.

And, this was at 120mm/s and in ABS, not 40 mm/s :-)

I plan to give PLA and 5 Micron, and possibly 2.5 Micron a try
sometime, should be
possible with Sailfish. 10 Micron is probably the limit for ABS.

[Dan Newman]

On 13 Oct 2012 , at 2:13 PM, Martin Galese wrote:

> Zeno,
>
> I'm playing around with Slic3r + Sailfish now and for the same reasons.
> From talking with Jetty, it seems like there are two different things that
> the Sailfish ReplicatorG does that Slic3r doesn't. The first is that
> there's a new instruction in the s3g that helps the firmware with movement
> planning. But that seems to work fine if you load the Slic3r gcode into
> the new RepG and build your s3g as normal.
>
> The second thing is that there is a new M code to turn acceleration on and
> off. The altshell plugin uses this in order to turn off acceleration for a
> nicer surface finish. Currently, Slic3r isn't going to generate those.
> However, I've had very good luck just using Scli3r's normal support to
> reduce the print speed for exterior perimeters.

FWIW, I did a lot of testing on the Replicator comparing altshell to perimeter
slowdown and found them to be, in general, comparable in finish quality on
approximately smooth surfaces: surfaces without lots of discontinuities in
their gradients (discontinuities in the derivatives of lines if you want the
2d equivalent). On surfaces such as cushwa's owl, I think altshell was
slightly better on the wings and chest but the perimeter slowdown samples were
quite good as well.

Dan

[Dan Newman]

For perimeter slowdown testing, I was using a multiplicative factor of 0.5.

Dan

[Jetty]

> Before I really commit

For those who are in any doubt about the quality difference between
5.5, 6.0 and Sailfish on the Replicator, I've made up some hi-res
shots
showing the difference with the defaults for each Firmware, printed at
120mm/s
accelerated.

There's no post processing, Yes really :-)

http://thingiverse-production.s3.amazonaws.com/renders/ac/7f/42/e3/83/IMG_0882_display_large.jpg
http://thingiverse-production.s3.amazonaws.com/renders/d1/0f/4c/83/62/IMG_0886_display_large.jpg
http://thingiverse-production.s3.amazonaws.com/renders/03/26/9c/50/8e/IMG_0883_display_large.jpg

And here's a comparison of different resolutions down to 10 Micron
from some testing
we were doing a few weeks ago. All Replicator, 120mm/s and
PrintOMatic:

http://thingiverse-production.s3.amazonaws.com/renders/3d/85/1b/eb/9c/IMG_0202_display_large.jpg

[Jetty]

As my wife mentioned, those squirrelz have fur.

[Doogiekr]

wow, very clear improvement... impressive work

[Gerald Orban]

Congrats to the folks who worked on this firmware. I'm extremely impressed with the improvements over the MBI FW! My Replicator runs smooth as silk and I love the new beep tunes.

[Matthew Marlowe]

Jetty,

General questions before I consider switching my replicator 1 over to sailfish:

a) Generally the print settings in replicatorg refer to the max
printing speed, not to the average. In nearly all my prints using
replicatorg37 with firmware 6.0 and my own reductions to the firmware
accelleration settings, I have max speed set to 60 and the skeinforge
stats module reports an avg speed of ~48 and the print usually takes
~20% longer so perhaps the true average with the firmware settings is
only 40mm/s. How much of a speed increase in avg printing speed if I
upgrade? Even if I put in 100mm/s for max speed, will the new avg
still likely be under 60?

b) Also, exactly what firmware accel speed/jerk settings have you
tested with using sailfish? Assuming that we need to manually modify
them, are they documented somewhere? Also, what are they tuned for?
General best compromise of speed/performance....are there other
recommended settings for more quality focused replicator 1 users?
While I'd love my replicator1 to be faster, I'd rather not trade print
quality or overall reliability/durability of the unit for it so I'd
like to have an idea of how conservative the default settings are.

c) It is suggested that we write down certain numbers from the
firmware settings, exactly which ones are critical for the
replicator1?

d) To use sailfish, I'm going to have to bump the version of
replicatorg distributed via gentoo linux....anything that other
packagers of replicatorg should know? Are all the replicatorg
files/install process generally the same with just updated contents
and maybe some additional settings/etc?

Thanks,
Matt

[Mark Cohen]

The only thing that confuses me a bit is fine and quick draft. I was
expecting one of them to set the x/y max back to 500 and they show
1000 on the TOM. There may be others that were not set back to
defaults but I was not paying attention to it.

> ** What's the resolution and speed?
>

[Jetty]

How much of a speed increase in avg printing speed if I
> upgrade? Even if I put in 100mm/s for max speed, will the new avg
> still likely be under 60?

It all depends on what you're printing and what your settings are.

We provide conservative settings that we've found to work well on a
Replicator.

> b) Also, exactly what firmware accel speed/jerk settings have you
> tested with using sailfish?

The defaults. There are 2 buttons in the acceleration settings. Fine
and Draft.
We suggest starting with Fine still you're familiar with everything,
then tweaking stuff
if you see the need. All the demo prints you have seen from myself
have
been printed with "Fine".

One of the limitations of the Replicator is the platform vibration and
mass up top.
The ToM suffers less because the top mass doesn't move around and
induce vibration
into the build platform.

So, the Replicator could benefit from for example the bowden drive
setup
someone was working on, or platform strengthening. With that you can
make the acceleration
settings more aggressive whilst still maintaining print quality.

> Assuming that we need to manually modify
> them, are they documented somewhere?

You don't really need to modify then, just check they were setup
correctly on install.

What they do is documented in the old Jetty Firmware docs, and what
they should be
is documented in the new Sailfishu doc.s

>  Also, what are they tuned for?
> General best compromise of speed/performance....

Yes, primarily speed versus print quality. But again, we've done the
leg work
for you. Start off with the defaults and 120mm/s feed rate, 200m/s
travel.

And remember slower doesn't always net a benefit.

>are there other
> recommended settings for more quality focused replicator 1 users?

No, those are them. We find the print quality with those settings to
be for
the most part as good or better than printing non-accelerated, and
certainly
a few magnitudes better than acceleration in 5.5 or 6.0

> While I'd love my replicator1 to be faster, I'd rather not trade print
> quality or overall reliability/durability of the unit for it so I'd
> like to have an idea of how conservative the default settings are.

Default settings are very conservative. We value reliability and
print quality first
too.

> c) It is suggested that we write down certain numbers from the
> firmware settings, exactly which ones are critical for the
> replicator1?

Everything in Motherboard Onboard Preferences. (I generally
screenshot it
myself).

> d) To use sailfish, I'm going to have to bump the version of
> replicatorg distributed via gentoo linux....anything that other
> packagers of replicatorg should know?  Are all the replicatorg
> files/install process generally the same with just updated contents
> and maybe some additional settings/etc?

Use our RepG0039-Sailfish, use that to generate your files, and
regenerate your
gcode / s3g's with this new version.

Diffs from the standard RepG can be found on our github, and there's
minimal changes from a standard repG apart from adding a couple of new
acceleration commands, but nothing you need to worry about from
a usage point of view, accept remember the RepG and firmware are
paired together.

Also, although you probably won't want to after trying it, you can
always
go back to RepG 6.0/5.5

[PropellerScience]

OK, stupid question here: I just installed the USB drivers for the
first time for the Replicator, and it's showing up in the Windows
hardware section as COM4, but RepG 39 says that port is already in
use. I did the Stop Conveyor in Makerware, although I've never used it
for anything, but still I'm stuck here.
Suggestions why Com4 is already in use?

[Jetty]

I've seen USB drivers get locked up before on all firmwares, this is
what I normally do to fix it:

1. Shutdown conveyor
2. Shutdown RepG.
3. Unplug the usb cable
4. Reset the bot
5. Connect everything up again.

Generally the problem happens a lot more to me since installing
Makerware, so it's probably Conveyer that is
stealing the connection.

[Dan Newman]

Some other application has grabbed the port. Unfortunately, there's plenty of
windows (and Mac) applications which will grab USB ports which they incorrectly
think have a device of theirs attached. (The LiveScribe application is notorious
for doing this with AVR and PIC USB hardware, for example.)

I'm not a Windows users, so I'm not very adept at figuring out what program has
allocated what I/O device.

Dan

[PropellerScience]

I'm excited too, but I'm having bad luck so far. My starting point is
I've just plugged the USB cord in for the first time. Driver install
went OK, but that's about it. Windows says the Mightyboard is on COM4,
but RepG33 says COM4 is in use. I finally got it to work by ending the
python process and restarting RepG. I could home, but not jog the
axis, and the temp seemed to be scrolling. The python process always
comes back even after I stop it. I did manage to get to the point
where I was able to upgrade the firmware to 5.5 as a stepping stone,
but now there's no LCD or backlight? The Replicator still whistles
after booting up, and the LED light strip turns on though. Now I can't
get Rep39 to see the COM4 port no matter what I do. I keep killing
python, but it keeps coming back for some reason. The conveyor thing
in Makerware was stopped even before I started all of this. What's up
with python always restarting? OK, I unplugged the USB and plugged it
back in again and Rep39 sees the port. Oops, I bet I forgot to
reconnect the ribbon cable to the Mightyboard while thinking I needed
to find the revision. Oh noooooo! It's playing some evil tune now when
it boots up, and look at what happened to my display!!!!!
http://www.flickr.com/photos/thesannio/8087355643/in/photostream/lightbox/

[Jordan Miller]

Jetty, Dan,

What would it take to get Sailfish running on a reprap board, like RAMPS, RAMBo, or the Ultimaker Board?

I'd LOVE to test this out for 3D sugar printing for science.

jordan

[Jwo Fox-Lee]

Ok, I've got Sailfish on my TOM (TOM w/ HBP and Stepstruder MK7 (Sailfish)), and everything seems to be going really well.
The only weird thing is when starting up ReplicatorG, I get an error message saying:

Tool Count Mismatch. Expecting 1 tools, reporting -1tools
Please double-check your selected machine.

It doesn't seem to affect anything, since I am able to slice and print. Just curious what this is.

Also, are there any tricks to printing at .1mm? The corners curl up slightly, and it looks almost as if it's putting down too much plastic.

- Jwo

[RocketSled]

Unplug the USB, wait for the "USB disconnecting" sound, plug it back in, try again.  On my Win7 system, sometimes when I start RepG with my Rep1 plugged in, it can't find the port.  Leaving RepG running and cycling the connection always fixes it.

[RocketSled]

Separately, I gotta say that Sailfish is a significant, huge upgrade from MB's 6.0 FW.  The difference is immediately detectable, the Rep is much quieter, both motor noise and the associated rattles it produces in the structure while printing.  

My first print is running now.  It's a mechanical part, not very complex.  6.0 FW was giving me fits, with all sorts of wavy artifacts in the shell.  I'd done a bunch of messing around to try and tune it out, but even at <40mm/s it was still messing up my exteriors.  Sailfish is running now at 60mm/s and the print is about half done and it looks friggin' perfect. 

The part has a lot of radii, it's round-ish on the outside and I have 3mm diameter screw holes going through a few places in the center.  MB 6.0 FW made a really horrible sound when extruding the curves, especially small ones like the screw holes, like something was grinding (just bad accel, though, nothing mechanically wrong).  Sailfish makes a pleasing harmonic whine. If you know anything about CNC, you know that's a good thing.

Wow.  I'm really, really impressed.  And I'm an engineer, that's hard to do.  It's like a different printer, now.  IMO, Makerbot should license this FW and scrap their own.  Nice work, Jetty!

[Jetty]

> Tool Count Mismatch. Expecting 1 tools, reporting -1tools
> Please double-check your selected machine.

Connect via RepG, go to Motherboard Onboard preferences, and change
the tool count to the number of extruders
you have, then click Commit.

> Also, are there any tricks to printing at .1mm? The corners curl up
> slightly, and it looks almost as if it's putting down too much plastic.

I normally back off the Z Home Offset a bit, so if I'm at -3.0mm, I'll
change it to -3.5mm or -3.75mm, that provides
more room. But if the corners curl up, clean the kapton with acetone,
and you may want to move it closer.

Then, make sure you have deprime set to 0, for 0.01mm (10Micron) to
0.1mm (100Micron) printing.

[Jetty]

> <40mm/s it was still messing up my exteriors.  Sailfish is running now at
> 60mm/s and the print is about half done and it looks friggin' perfect.

Glad it's going well for you. Regarding the 60mm/s, through what you
learnt
previously away, i.e. try 100mm/s or 120mm/s. I routinely print at
120mm/s and 200mm/s travel
on stock hardware. For example that demo piece with all the text in
was printed at that speed.

> FW and scrap their own.  Nice work, Jetty!

(and Dan :-) )

[Dan Newman]

>> Also, are there any tricks to printing at .1mm? The corners curl up
>> slightly, and it looks almost as if it's putting down too much plastic.
>
> I normally back off the Z Home Offset a bit, so if I'm at -3.0mm, I'll
> change it to -3.5mm or -3.75mm, that provides

Speaking from experience, if you don't do that you will find that your 0.1mm prints
stick REALLY well to the platform. I basically had to peel the kapton tape up to
get some owls removed. None of the usual tricks would work for me.

Dan

[Jetty]

> > I normally back off the Z Home Offset a bit, so if I'm at -3.0mm, I'll
> > change it to -3.5mm or -3.75mm, that provides

And, that should have been -0.3mm, -0.35mm and -0.375mm

[Charlie Prevost]

After the first print with the new sailfish firmware, all I can say is: Oh. My. God!
The improvements to surface finish, speed and machine movements are huge (and I havent even begun tweaking for my machine yet). Thanks Jetty and the sailfish crew. legends.

Cheers, Charlie

[Mark Cohen]

Everything went well with my install on Replicator 1 and TOM. I can
finally use my Replicator. I printed perfect heart gears just by
following the directions. On the TOM I noticed one thing after I
finished printing the stand for the screwless heart gears. It had a
long name and the display was right instead of left justified after
the build, I am attaching a picture. Also when I was fiddling around
on the replicator I discovered what may be the cause of people seeing
the 2 bars and the machine locking up after loading firmware. If they
are loading mbi 6.0 they may have loaded the 1.7 fw for Rep 2 instead
of the 1.5 fw for the Rep 1. This causes that. You need to power down,
unplug usb, power up and reset and plug in the usb and upload the
correct fw.

> --
> You received this message because you are subscribed to the Google Groups
> "MakerBot Operators" group.

> To view this discussion on the web visit
> https://groups.google.com/d/msg/makerbot/-/J1wx0uI0wXwJ.

[DB]

I have a pretty serious flutter in the xy motion when the extruder is printing inset loops on curved or cylindrical features.  It's enough vibration to shake the build platform.  Anyone else seeing this?  

Thanks,

Dan

[aek...@charter.net]

Here's a video demo in typical out of focus, bad lighting style.
someday I'll learn :-)
Enjoy!
http://youtu.be/eQmbltbpWTQ

[DB]

Sorry, I've got a Replicator if that helps.

Thanks,

[Dan Newman]

On 15 Oct 2012 , at 7:13 AM, DB wrote:

> I have a pretty serious flutter in the xy motion when the extruder is
> printing inset loops on curved or cylindrical features. It's enough
> vibration to shake the build platform. Anyone else seeing this?

What print speed and what bot? I'm guessing a Replicator if it's vibrating
the build platform. Try the "fine" settings. They reduce the max x & y
speed changes. On Jetty's Rep1 as well as mine, somewhere between 20 - 25 mm/s,
we start seeing lots of induced vibration in the cantilevered build platform.
Obviously it may be more or less severe on other bots. The "draft" acceleration
settings are designed for more speed, less quality and use 40 mm/s as the max
x & y speed changes.

Dan

[Jetty]

> On the TOM I noticed one thing after I
> finished printing the stand for the screwless heart gears. It had a
> long name and the display was right instead of left justified after
> the build, I am attaching a picture.

That one is a bug, it'll should be fixed in the next release.

[DB]

Thanks, Dan.  I've got a Replicator 1.  I've been printing with the fine defaults at speeds from 70mm/s up to 120mm/s.  The issue seems to be only with inset loops.  Not only are they wavy, they are printing at about half the speed of the perimeter (which is already printing at half speed).  The vibration is not present when printing perimeters. I did not have any of these issues with the standard RepG 39 and Makerbot firmware 5.5.  The vibration is minimized when I lower the max acc. rates to 500mm/s, but at 1000mm/s, it's pretty rough.  

I tried printing without print-o-matic to see if that helped, but it made no difference.  

If I can't get this worked out, how do I go back to 5.5 firmware?  It's not populating in the upgrade firmware window when I open standard RepG 39.  Do I need to use an older version to do this?

I love how quiet the bot is with Jetty, but I don't really have any speed increase on account of the slow, wavy inset loops.

Thanks!

Dan

[Jetty]

Hi Dan,

It might be time to go back to the recommended defaults for the
acceleration (Fine Quality) and the following setting
listed near the end of here:
http://wiki.makerbot.com/sailfish-firmware-setup-guide

Also, there's a variety of Skeinforge settings that could cause this
issue. Please copy the Replicator defaults for Skeinforge
, alter as per the setup guide and try again.

[DB]

Thanks for the suggestion.  I went a step further and wiped RepG from my machine and installed a fresh Sailfish version, went through the setup again, and calibration cubes are printing beautifully.  I will try something with some curves and see how it goes.  Thanks for getting back to me. 

Best,

Dan

[DB]

So, no luck with curves or cylinders.  With all settings the same as the square calibration blocks, I printed cylindrical calibration blocks and the interior extra shells, or loops (inset) printed slower than the perimeter and with some flutter (vibration) in the x axis.  Again, I do not see this behavior while printing straight lines--only curves, and speed of print seems not to matter.  

Thanks,

Dan

[Dan Newman]

DB wrote:
> So, no luck with curves or cylinders. With all settings the same as
> the square calibration blocks, I printed cylindrical calibration
> blocks and the interior extra shells, or loops (inset) printed slower
> than the perimeter and with some flutter (vibration) in the x axis.

Are you using the "altshell" plugin in SF? Do you have both Altshell
activated and Perimeter slowdown
in the "Speed" plugin? (I do not know offhand if Perimeter slowdown
impacts the loops as well. I'd have
to look at the Python code for the Speed plugin to tell.) Please ensure
that if you have Altshell activated,
then you have the perimeter factors set to 1.0 in the Speed plugin.
And, another thing you can try is to
disable Altshell and set those two factors to 0.5 in the Speed plugin.

Dan

[Dan Newman]

Isaac Bergman wrote:
> Hi Jetty,
>
> Newb here. I updated my Replicator to the Sailfish 6.2 beta and
> followed the guide you posted but there were a couple of points which
> were ambiguous.
>
> For example: "In the "Speed" plugin, set the slicing feed rate in the
> "Feed Rate (mm/s)" field."
>
> Set it to what speed exactly?
If you use print-o-matic, then you do not need to do that step:
print-o-matic will set the speed for
you. It's the feed rate you want to print at (e.g., 120 mm/s).

Dan

[DB]

Hi Dan-

I'm not using altshell in SF.  I just set perimeter feed and flow rates to .5.  

I just ran a whole bunch of test prints with print speed of 75mm/s and started with acceleration settings for rough draft.  Starting with the default x/y maximums of 40mm/s, and working my way down in increments of 5mm/s until I got the same failure of vibration in the x axis during curves or circles.  The extra shells and perimeters behaved as they should for 40, 35, 30, 25, and 20mm/s.  As soon as I set the max x/y to 15mm/s, the extra shells printed with the vibration, and at a slower rate than the perimeter.  Back up at 20mm/s, everything smoothed out again.  

Another thing I should note is that I've dropped deprime in favor of retraction in SF.  I was getting a lot of zits with deprime--something I've never seen with the Makerbot firmwares.  As soon as I turned off deprime and set SF retraction to .5mm and 0 restart, the zits disappered.  I'm also using the dwindle plug-in in SF, and very pleased with how that has improved the quality of my prints.  

Thanks for all the help.  I'll keep pushing the limits with the speed when I've got time, but for now, I'm at 80mm/s.   

Best,

Dan

[Dan Newman]

On 15 Oct 2012 , at 6:57 PM, DB wrote:

> Hi Dan-
> I'm not using altshell in SF. I just set perimeter feed and flow rates to
> .5.
>
> I just ran a whole bunch of test prints with print speed of 75mm/s and
> started with acceleration settings for rough draft. Starting with the
> default x/y maximums of 40mm/s, and working my way down in increments of
> 5mm/s until I got the same failure of vibration in the x axis during curves
> or circles. The extra shells and perimeters behaved as they should for 40,
> 35, 30, 25, and 20mm/s. As soon as I set the max x/y to 15mm/s, the extra
> shells printed with the vibration, and at a slower rate than the perimeter.
> Back up at 20mm/s, everything smoothed out again.

Jetty and I both have noticed that there's some slow
speed regimes best avoided. When we were choosing the draft and fine settings,
I found that if I had the max speed changes for x & y around 10 mm/s on my
Replicator, I got much worse quality than at 15 mm/s. Then there was another
transition somewhere between 20 mm/s and 25 mm/s. That's why we chose 15 mm/s.

So, for your bot, it sounds like there is a vibration problem when you fall
below 20 mm/s.

>
>
> Another thing I should note is that I've dropped deprime in favor of
> retraction in SF. I was getting a lot of zits with deprime--something I've
> never seen with the Makerbot firmwares. As soon as I turned off deprime
> and set SF retraction to .5mm and 0 restart, the zits disappered. I'm also
> using the dwindle plug-in in SF, and very pleased with how that has
> improved the quality of my prints.

Interesting as it's been my impression that the folks who put dwindle in
have abandoned it as not working. At any rate, you absolutely cannot mix
dwindle with deprime. We didn't even document that since, again, our take
was that dwindle was abandoned almost as soon as it was put into SF 50.
That's not to say that it doesn't work for everyone: clearly it works well
for you in which case go for it.

Also, I've had some models for which I've found reversal to work slightly
better. (Probably means I needed to tune deprime differently for that
model.) One advantage of deprime is that the firmware also can use
it when the planner is running behind and the queue empties of planned
segments. So, if you ever print something with lots of very small
segments and you start seeing little zits or blobs, then consider re-activating
deprime as that's one of the things it can combat. And, retraction cannot
help there as those sorts of pauses are not known to the slicer.

Dan

[DB]

Hey Dan-

Thanks for your response.  I'm totally open to using deprime if I can get it tuned properly.  I had the value set at the recommended default of 16, but did not play around with it at all.  Are you using the same value?  higher?  lower?  

In terms of regimes, with my bot, it seems that as I increase print speed, I also have to increase max x/y acceleration values to avoid the flutter in the extra shells.  Something about the speed/acceleration ratio.  Is this part of the math?  I'm not a software guy at all, just curious.  

Thanks,

Dan

[James McCracken]

On Cupcake 3G5D, are you anticipating being able to print from USB without issues?  I've had my work computer locked down as far as removeable storage goes, and would love to be able to "just hit print" from within RepG without having to e-mail myself the files, go on my backup computer, copy to SD, etc.

[Dan Newman]

On 16 Oct 2012 , at 6:53 AM, DB wrote:

> Hey Dan-
>
> Thanks for your response. I'm totally open to using deprime if I can get
> it tuned properly. I had the value set at the recommended default of 16,
> but did not play around with it at all. Are you using the same value?
> higher? lower?

There was a time when Jetty and I thought that 34 was a good value for
Replicator 1's, but after several refinements to this or that in the code,
and many more test prints later, we ended up at 16. However, we were doing
a lot of our judgements based upon looking at the exterior zipper you
get when you use "perimeter > loops > infill" for the ordering of pieces
in a layer. There are certainly other ways to judge whether or not the
correct deprime action is occurring (as witness g.wygonik's stringing
when travelling from one section to another of a build) [the extruder
is turned off there, but ooze can occur, even if you do a pretty severe
deprime -- getting rid of the excess pressure in the nozzle is pretty
tricky when dealing with molten, viscous plastic. But it's achievable
since I assume that g.wygonik didn't have the issue before. So it's
a matter of tuning this or that.]

> In terms of regimes, with my bot, it seems that as I increase print speed,
> I also have to increase max x/y acceleration values to avoid the flutter in
> the extra shells. Something about the speed/acceleration ratio.

It sounds more and more as though there is some speed range in which
the stepper motor frequency excites vibration in some part of the bot.
Thus, whenever the speed passes through that speed range, the problem
shows up. It may be that depending upon the max accel and the length
of the segments, that the planner has an unfortunate tendency to land
you in that speed range. Increasing the acceleration then helps the
planner to allow a higher/larger speed to be attained, thus on some
segments taking you beyond that speed range BUT now for some shorter
segments which previously fell below that range, you now end up in that
range.

You may want to ensure that the axis rods are lubed. Printing towards
the back of the cantilevered platform can help as it reduces the
amplitude of some vibrations, especially the differential vibration
between the platform an extruder nozzle.

Dan

[Dan Newman]

We've not tried it ourselves (not having cupcakes available to us). But,
yes the improvements we made in that dept. for ToM's applies equally to
Cupcakes.

Dan

[James McCracken]

Hmm if I can bring myself to finally mess with the first working firmware I've had in 4-5 months, I might be willing to be your guinea pig...

 

Who am I kidding?  If I can just hit print that's worth every bit of pain and suffering.  As soon as my baby boy gives me a night or two to work on it, I'll let you know how it goes... I assume the sailfish forum is the best place to provide that feedback?

Dan

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

[DB]

Thanks, Dan.  I really appreciate the help.  However, I would like to revert to 5.5 and compare again, but when I try to upgrade back to 6.0, it locks my bot and energizes the hpb.  Gives me two solid bars on the lcd.  Can you tell me how to get back to the old firmware?

Thanks,

Dan

[Dan Newman]

On 16 Oct 2012 , at 10:56 AM, DB wrote:

> Thanks, Dan. I really appreciate the help. However, I would like to
> revert to 5.5 and compare again, but when I try to upgrade back to 6.0, it
> locks my bot and energizes the hpb. Gives me two solid bars on the lcd.
> Can you tell me how to get back to the old firmware?

1. From the LCD display, under utilities, you can use the EEPROM menu to
save all of your EEPROM to a SD card file (eeprom_dump.bin) which you
can later restore (with Sailfish).

2. Just to be on the safe side, take screen snapshots of your Onboard Preferences. Particularly,
the offsets in the first tab. Do this from RepG.

3. Install 5.5 and then, from RepG, do a "reset to factory defaults". That should
be sufficient. Worse comes to worse, "reset eeprom". These are done from the
Machine > Onboard Preferences menu

Dan

[DB]

Awesome.  Thanks, Dan.  Much appreciated.  

[David]

Will this work with sanguinololu?

Is it as simple as changing pinouts in the code?

 

Dave

[Dan Newman]

On 16 Oct 2012 , at 1:13 PM, David wrote:

> Will this work with sanguinololu?

The requirements are Gen 3 (RepRap Motherboard v1.2), Gen 4, or Mightyboard electronics.
If it does not conform to those, then modifications to the firmware would be required.

> Is it as simple as changing pinouts in the code?

Someone else may be able to answer: I was never that familiar with the Sanguinololu.

Dan

[hellphish]

You're not going crazy. Sailfish is a pre-release, and not all the documentation has been updated to reflect the new firmware. Extruder steps/mm is now read from the machine.xml file and sent to the bot directly. As for acceleration, just make sure that "enable acceleration" is checked at the top. 

On Thu, Oct 18, 2012 at 11:42 AM, Michael Buffington <michael.b...@gmail.com> wrote:

I'm new to alternative Replicator 1 firmwares, so forgive me if I'm asking some obvious questions. I've read the Install and Setup guides multiple times and I'm actually confused.

Machine details first: a month old Dual Extrusion Replicator 1 that was on the MBI 6.0 firmware. The Board is a MightyBoard 1.5 revE board. I've installed the Sailfish firmware according to the install guide here: http://wiki.makerbot.com/sailfish-firmware-installation-guide. I've also verified that the firmware was installed by navigating, on the Replicator's LCD, to Utilities > Version Number. There, I see "Sailfish, Free SRAM 778, Thing: 32084, Firmware Version 6.2"

Several places in the setup guides refer to LCD menu items or ReplicatorG menus that don't exist are don't mirror what I'm seeing in software. Are the setup guides for previous versions and need updating?

For example, on the Setup Guide (http://wiki.makerbot.com/jetty-firmware-setup-guide):

Under Step 4a there's an image showing the Onboard Preferences menu. In that image we can see 5 possible sub sheets when the Motherboard sheet is toggled: Endstops/Axis Inversion, Homing, Acceleration, Acceleration (Advanced), and Misc.

The instructions in Step 4a say to make sure the "Acceleration planner enabled" box, under "Acceleration (Advanced)" is checked.

Problem is, my ReplicatorG (just triple checked to make sure it's the Sailfish version), doesn't even have the "Acceleration (Advanced)" sub sheet. Consequently there's no check box to check.

Is it safe to assume that maybe the guide is specific to a Thing-O-Matic? Am I doing something wrong?

Another example:

On the Jetty Firmware Manual (http://wiki.makerbot.com/jetty-firmware), there are lists of Menu items available on through the machine's LCD that aren't on my machine. One, in particular, is required as part of setup tuning: the Extrdr Steps/mm menu.

Did I miss a step? Is there a separate update for the menuing system on my Replicator?

 

Sailfish is the next release of Jetty Firmware for the Replicator 1
and ThingOMatic. Jetty firmware received it's name from the

--

You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.

To view this discussion on the web visit https://groups.google.com/d/msg/makerbot/-/vF6Rct67FpEJ.

[Dan Newman]

On 18 Oct 2012 , at 11:42 AM, Michael Buffington wrote:

> I'm new to alternative Replicator 1 firmwares, so forgive me if I'm asking
> some obvious questions. I've read the Install and Setup guides multiple
> times and I'm actually confused.

…

> For example, on the Setup Guide
> (http://wiki.makerbot.com/jetty-firmware-setup-guide):

Those are the wrong directions. Those are for the Jetty Firmware 3.5 (Thing-o-Matic
and Cupcake). You need to see the Sailfish setup guide (which is, essentialy
v4.0 of the Jetty Firmware).

http://wiki.makerbot.com/sailfish-firmware-setup-guide

Dan

[Michael Buffington]

Shame on me. I was being so careful to make sure I wasn't reading older docs.

I'll need to go learn how to tweak my machine.xml (new to me).

[Dan Newman]

On 18 Oct 2012 , at 12:11 PM, Michael Buffington wrote:

> Shame on me. I was being so careful to make sure I wasn't reading older
> docs.
> I'll need to go learn how to tweak my machine.xml (new to me).

Which tweak? I can possibly give you some pointers. (And maybe
I can update the wiki if you find any parts less than clear.)

Dan

[Michael Buffington]

I'm pretty sure I need to be extruding at a higher rate than my current rate when using PLA. While the first layer of the calibration cube prints, I don't get nice smooth lines - I get pulses. I thought the bed might be too low, but even raising the bed as close as it can be and still extrude doesn't change the pulsing.

I'm considering turning off Print-O-Matic, and testing out tweaks in the Speed plugin.

[Erwin Ried]

Wow! I was not aware of Slic3r! it really kicks RepG ass!

On 13 oct, 18:13, Martin Galese <mar...@galese.net> wrote:
> Zeno,
>
> I'm  playing around with Slic3r + Sailfish now and for the same reasons.
>  From talking with Jetty, it seems like there are two different things that
> the Sailfish ReplicatorG does that Slic3r doesn't.  The first is that
> there's a new instruction in the s3g that helps the firmware with movement
> planning.  But that seems to work fine if you load the Slic3r gcode into
> the new RepG and build your s3g as normal.
>
> The second thing is that there is a new M code to turn acceleration on and
> off.  The altshell plugin uses this in order to turn off acceleration for a
> nicer surface finish.  Currently, Slic3r isn't going to generate those.
>  However, I've had very good luck just using Scli3r's normal support to
> reduce the print speed for exterior perimeters.
>
> I'm still testing with Slic3r, but so far so good!
>
> (FYI, I'm using the current dynamic-flow branch out of git for my testing,
> since that what I had been playing with earlier.  YMMV with the various
> actually released versions.)
>
>
>
>
>
> On Sat, Oct 13, 2012 at 5:01 PM, Z LeHericy <zeno...@gmail.com> wrote:
> > Is slic3r supported with Sailfish?
>
> > Ever since i switched to Slic3r, i've vowed not to go back to skeinforge...
>
> > Slic3r is several orders of magnitude faster at slicing up parts, and is
> > easier (in my opinion) to configure right on the first try.
>
> > -Zeno LeHericy
>
> > //((=:Z:=))\\
> > INVENTIONS
> > Technologies
> > zinventions.com
>
> > On Sat, Oct 13, 2012 at 1:38 PM, Dan Newman <dan.new...@mtbaldy.us> wrote:
>
> >> On 13 Oct 2012 , at 1:27 PM, ddurant wrote:
>
> >> >> I had expected to spend hours tuning SF50.
>
> >> > If you spend even one hour 'tuning' skeinforge with volumetric 5d,
> >> you're
> >> > doing it wrong..
>
> >> > It really is a better solution than printomatic. Even the guy who came
> >> up
> >> > with the original idea and code behind printomatic thinks that. :)
>
> >> I wonder who that might be….
>
> >> Dan

>
> >> --
> >> 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.
>

> >  --
> > 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.
>

> --
> Martin Galese

[Tyler Pope]

I'm also having problems printing arcs and circles. I'm using default "fine" settings and the straight lines look great but any arc or circle shutters back and forth. The parts in the below picture should have been round. Any ideas?

On Monday, October 15, 2012 9:13:41 AM UTC-5, DB wrote:

I have a pretty serious flutter in the xy motion when the extruder is printing inset loops on curved or cylindrical features.  It's enough vibration to shake the build platform.  Anyone else seeing this?  

Thanks,

Dan

On Saturday, October 13, 2012 2:47:36 AM UTC-4, Jetty wrote:

Today, we're excited to announce the pre-release of Sailfish for
Replicator 1 / ThingOMatic for tinkerers.


** Why the name change?

Sailfish is the next release of Jetty Firmware for the Replicator 1
and ThingOMatic. Jetty firmware received it's name from the

users at the beginning, this was when Jetty was the main
developer working on it so it fitted.  But shortly after it's
release,
Dan joined the team and it represents  mainly the work of two
developers.

Sailfish provides a major (5-10 times) speed improvement to
acceleration over Marlin and Jetty firmware.  With the name
Sailfish, we're paying homage to the origin of the acceleration
part of the project (Marlin).  Both Sailfish and Marlin are ocean
fish, and a Sailfish is faster than a Marlin.


** What's the difference in acceleration between firmwares?

Sailfish represents the state of art in an 8 bit processor.  In
our opinion, it's better than Marlin (due to the increase in
planning speed, bug fixes, JKN Advance, Yet Another Jerk).
On Replicator, it's also faster, produces higher quality prints
compared to
5.5 and 6.0.   On ToM, it's the next step up in print quality
compared to Jetty 3.4/3.5.

The accelerated capabilities of this firmware can be considered
to be higher performing than the other printers out there,
including Ultimaker.


** What's the resolution and speed?

First off, we haven't pushed the limits to the ridiculous (you
can likely do better than us with  a little work), but we've
tested the following:

Replicator, down to 0.01 Micron layer heights in ABS using
PrintOMatic and minimal tuning. Travel Speeds up to 340mm/s
on the X axis.  Printing up to 150mm/s. We routinely print at
120mm/s and 200mm/s Travel (limited by the Replicator's
mechanics )

ToM, down to 100 Micron layer heights using PrintOMatic
and minimal tuning.  We routinely print at 120mm/s and
150mm/s Travel.


** Volumetric 5D and SF50?

On both ThingOMatic and Replicator 1, all our testing has
been with SF50 and Volumetric 5D. On ToM, when using
Sailfish, we now recommend and that you use SF50 and 5D,
not legacy SF35.



** Other features

Replicator 1:
        - fast menus with top/bottom wrap around
        - jogging menu improvements
        - long scrolling filenames in build menu
        - Filament usage odometer and trip counter
        - override gcode temperature
        - 4 custom material profiles (e.g. ABS, PLA etc.)
        - Pause @ ZPos and Pause
        - Switch off heaters during pause
        - Build Z Position in build stats
        - Home offsets adjustment
        - Ditto printing (Duplicate printing of a
                small object to 2 nozzles via firmware)
        - Reminder Beep on filament load

ToM:
        - everything that was in 3.4 / 3.5
        - Ditto printing and dual extruder support (untested)
        - No more MCode configuration for non LCD-users
        - New Sailfish driver in RepG to replace makerbot4g

Both:
        - Machine settings automatically sent to bot from the xml
        - Acceleration configuration via Onboard Preferences in RepG
        - Fine and Draft quality settings
        - Eeprom menu to save / read and erase Flash eeprom
        - Part of the acceleration calculations are pre-calculated in RepG


** Is it Open Source?

Of course, it always has been.  Source code can be found here:
https://github.com/jetty840/Sailfish-MightyBoardFirmware
https://github.com/jetty840/Sailfish-G3Firmware


** Where can I get support?

Help / Support can be found over on the Jetty Forum:
http://groups.google.com/group/jetty-firmware


Is this firmware supported by Makerbot Industries?

No, this is not an MBI official release and we don't work for MBI.
Although they will likely interoperate to some degree, we recommend
writing down your bot's settings before upgrading for input into
Sailfish
after installation.  Once in Sailfish, you can save your firmware's
settings
to SDCard.


** Is Makerware supported?

Currently, not all of Makerware is Open Source, so no, but we hope
to have support in the future.  You can still "plate" in Makerware,
and
export to stl, and then load it RepG.


** ReplicatorG

Sailfish requires use of a custom ReplicatorG 0039 which is called
ReplicatorG 0039 - Sailfish to generate the new acceleration
commands.


** Where can I get it?

See http://www.thingiverse.com/thing:32084 for details

[Dan Newman]

On 18 Oct 2012 , at 7:19 PM, Tyler Pope wrote:

> I'm also having problems printing arcs and circles. I'm using default
> "fine" settings and the straight lines look great but any arc or circle
> shutters back and forth. The parts in the below picture should have been
> round. Any ideas?
>

> <https://lh4.googleusercontent.com/-syyGbl5Fdio/UIC4ketiwMI/AAAAAAAADUE/XmutRxkr2ag/s1600/20121017_201927.jpg>

I'll bet that those "stutters" correspond to the edges of the facets
which make up the STL model. That is, you're seeing slightly more
plastic at just the spots where the print goes a little slower: at the
junction point between line segments.

Things you can try:

1. If you made that model in OpenSCAD then set $fn to a larger value (e.g., 100). You
can do that globally as well as locally in function calls. I.e., you can just put "$fn = 100;" at
the start of the OpenSCAD file.

2. Enable Altshell

3. Try increasing the max x & y speed change values. Try 20 or 25.

4. Play a little with K and/or K2.

Dan

[Dan Newman]

5. This may indicate that you are extruding a little too much plastic?
Have you tuned your 20 x 20 x 10 calibration cubes to have a nice, flat
surface?

Dan

[Tyler Pope]

I can physically see the head zigzag as it goes around arcs. I am also exporting my files out of Solidworks and haven't had issues with arcs until using sailfish.

I did however have issues with MBI's 6.0 firmware. I was getting a bad finish after the first 1/4". I don't know if they could be related in some way, but my bot seems to be off since I upgraded from 5.5.

[Dan Newman]

On 18 Oct 2012 , at 8:05 PM, Tyler Pope wrote:

> I can physically see the head zigzag as it goes around arcs. I am also
> exporting my files out of Solidworks and haven't had issues with arcs until
> using sailfish.

You may be hitting a speed range which induces vibrations. That cantilevered
platform and xy gantry seem to have an issue that way. (For me, it's in
a 20 - 25 mm/s speed range.) Printing towards the back of the build platform
helps somewhat.

> I did however have issues with MBI's 6.0 firmware. I was getting a bad
> finish after the first 1/4". I don't know if they could be related in some
> way, but my bot seems to be off since I upgraded from 5.5.

The finish bad after 10mm was a bug in MBI's 6.0 firmware which they fixed
with their 6.1 release.

Dan

[Dan Newman]

On 18 Oct 2012 , at 7:33 PM, Dan Newman wrote:

>
> On 18 Oct 2012 , at 7:19 PM, Tyler Pope wrote:
>
>> I'm also having problems printing arcs and circles. I'm using default
>> "fine" settings and the straight lines look great but any arc or circle
>> shutters back and forth. The parts in the below picture should have been
>> round. Any ideas?
>>
>> <https://lh4.googleusercontent.com/-syyGbl5Fdio/UIC4ketiwMI/AAAAAAAADUE/XmutRxkr2ag/s1600/20121017_201927.jpg>
>
> I'll bet that those "stutters" correspond to the edges of the facets
> which make up the STL model. That is, you're seeing slightly more
> plastic at just the spots where the print goes a little slower: at the
> junction point between line segments.
>
> Things you can try:
>
> 1. If you made that model in OpenSCAD then set $fn to a larger value (e.g., 100). You
> can do that globally as well as locally in function calls. I.e., you can just put "$fn = 100;" at
> the start of the OpenSCAD file.
>
> 2. Enable Altshell

And make sure that the "Use M320 / M321" box is checked.

Dan

[Cymon]

What is "Use M320 / M321"? I have it unchecked (the instructions didn't say to check it).

[Dan Newman]

On 19 Oct 2012 , at 8:14 AM, Cymon wrote:

> What is "Use M320 / M321"? I have it unchecked (the instructions didn't say
> to check it).

Actually, in the wiki they do as of last night. (Or did my changes get lost.)
Also, the next RepG release will have them checked by default for SF-47 and 50.

If you are using Sailfish or MBI firmware, you want it checked. Otherwise, nothing
happens.

If you are using Jetty Firmware 3.5 or earlier, you do not want it checked.

It's kind of hard to get useful help into the SF plugin setting's UI.

Dan

[Dan Newman]

On 19 Oct 2012 , at 8:25 AM, Tobias Hall wrote:

> I installed it, and after each succesfull print my bot(Replicator Dual)
> removes the endcap settings, which is kind of scary... Am I doing something
> wrong?

"Endcap settings"? Additional details, please.

Dan

[Tobias Hall]

Machine->Onboard preferences->Endstops/Axis Inversion->Invert Endstops == "No endstops installed"

So when my bot is homing it will go to far and I have to shut it down.

I'm just guessing that this is the cause, but it happens and I'm not sure of what to do about it.

[Dan Newman]

Tobias Hall wrote:
> Machine->Onboard preferences->Endstops/Axis Inversion->Invert Endstops
> == "No endstops installed"

That's actually correct, as counter intuitive as it might seem. It's
what my Replicator shows in the
attached screen shot. As to why things keep on getting set to that, I'm
not too sure (unless RepG
has code to explicitly change a Replicator to that.)

When you say that it homes too far, is it even homing in the right
direction? If you accidentally have
Cupcake or ToM start.gcode, the axes will be homed in the wrong
direction for a Replicator, and
vice versa. Since there are only endstops for one direction of travel
per axis, homing in the wrong
direction doesn't work -- there's nothing telling the bot to stop.

So make sure that your machine type (driver) is set to "The Replicator
Dual" and that in
Print-o-Matic you have checked the "use default start/end gcode".

Dan

[Tobias Hall]

Driver is set to Replicator Dual.

"Use default start/end gcode" is checked.

"Use Print-O-Matic ( stepper extruders only )" is also checked.

It seems to be homing in the correct direction. To the right, before it's about to move the extruder to the left and start heating.  It goes to far to the right...

I re-flashed and used the leveling script from RepG instead of using the machine's leveling script and since then it has not happend. 

Dan

--
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+unsubscribe@googlegroups.com.

[Dan Newman]

Tobias Hall wrote:
> Driver is set to Replicator Dual.
> "Use default start/end gcode" is checked.
> "Use Print-O-Matic ( stepper extruders only )" is also checked.
>
> It seems to be homing in the correct direction. To the right, before
> it's about to move the extruder to the left and start heating. It
> goes to far to the right...
>
> I re-flashed and used the leveling script from RepG instead of using
> the machine's leveling script and since then it has not happend.

Well, I'm glad to hear that things are working. No insight at the
moment as to what
you were experiencing though.

Dan

[Tobias Hall]

It started again...  Then stopped, then started. I really don't know what triggers it. It goes to far to the left and the right... Don't know what to make of this. I've gotten two small prints out of it so far, and I really like the acceleration, it looks, feels and sounds awesome. But I need to do something about this, but I don't know what.

Dan

[Dan Newman]

Tobias Hall wrote:
> It started again... Then stopped, then started. I really don't know
> what triggers it. It goes to far to the left and the right...

This is a stock, MBI Replicator 1 with standard stepper motors and
electronics?

Are you using a custom machine definition (xml description) or is your
Machine > Machine Type (driver)
the standard "The Replicator Dual"?

And you're using "ReplicatorG 39 - Sailfish" downloaded from the
Sailfish Thingiverse thing?

Dan

[Tobias Hall]

It's stock, I had to swap one fan, but that's all.

Ok, so I'm trying everything here! I installed MBI 6.0 firmware and guess what! It happens there as well! This means it's not on your side!

But I got this the same day I installed your firmware, that's why I made that connection. So I started to mess around with the endstop on the right side. And it did activate when I clicked on the trigger. But not always. It seems to be some glitch in the cable... 

I'm a noob, what is the easiest way for me to fix this glitch?

And I'm sorry if I sent you on a wild goose hunt for bugs! 

Thank you.

// Tobias

Dan

[Dan Newman]

Tobias Hall wrote:
> It's stock, I had to swap one fan, but that's all.
> Ok, so I'm trying everything here! I installed MBI 6.0 firmware and
> guess what! It happens there as well! This means it's not on your side!
> But I got this the same day I installed your firmware, that's why I
> made that connection. So I started to mess around with the endstop on
> the right side. And it did activate when I clicked on the trigger. But
> not always. It seems to be some glitch in the cable...

Ummmm, call MBI support ASAP. They have had problems with damaged
endstop cables
causing damage to the motherboards. That the problem is intermittent
suggests a possible
short in the cable. You may (or may not) be at risk for damage to the
motherboard. MBI
support can advise you. And based upon reports in this forum, they are
very responsive
and will replace damaged parts such as you describe.

Dan

[Bartholomew]

In installing Sailfish, I see machine type (driver) selections for various types (Thingomatic w/HBP and Mk7 (sailfish), etc), but no selection for my rig ((Thingomatic w/HBP and Mk6+).

Should I just select the mk7 version?

Many thanks,

Bart

[Dan Newman]

Bartholomew wrote:
> In installing Sailfish, I see machine type (driver) selections for
> various types (Thingomatic w/HBP and Mk7 (sailfish), etc), but no
> selection for my rig ((Thingomatic w/HBP and Mk6+).
> Should I just select the mk7 version?

That's an oversite (e.g., boo boo). I'll update the file and send you
it with directions.

Dan

[Bartholomew]

Thanks!

[Dan Newman]

Bartholomew wrote:
> Thanks!
Okay, you can use the Mk7 variant they're equivalent as far as Sailfish
is concerned.

However, I've attached an updated file and if you want to give it a
spin, let me know if it
works. You should see "Thingomatic w/ HBP MK6 (Sailfish)" appear. This
file
should be used to replace machines/thingomatic-sailfish.xml in the
ReplicatorG directory
tree. (On OS X, it's
/Applications/ReplicatorG.app/Contents/Resources/machines/ .)
You need to exit and restart RepG for the new file to appear.

Dan

[Bartholomew]

Good to hear. I am in the middle of a test build, so when that completes, I'll install the new file and send a message.

Thanks again for the fast response.