Funky wavieness with 6.0 firmware?

[Zak Kus]

So, i've done this print dozens, maybe even hundreds, of times. I upgraded to 6.0 recently (did some other prints that were successful), but when i printed out this guy it produced a weird wavy effect that i had never seen before. (See picture)

It was the same gcode i had used many times before (so no settings error). I thought it was maybe something funky with my filament (it is sorta old), however the pattern seems very regular and "intentional". All the peices have a little bit of wave, but its most noticeable on the dramatic slopes and curves. 

Has anyone else seen this?

[Jetty]

I'm guessing you're using acceleration? The causes are 2 things:

1. The smaller vibration effect is Vibration of the HBP.
Printing near the back of the platform will make it better, or reduce
your overall speed
and acceleration too (if you're using it).

2. The larger effect is due to too little filament being extruded at
the start of the each move,
and too much at the end of the move.

Certain shapes make the above problems worse, due to the increased
vibration
and resonance.

1. will start to become visible on speeds > 30mm/s (accelerated or
non).

[Tyler Pope]

I'm getting the same thing after upgrading to 6.0. I tried going back to 5.5 but that didn't help. Is it possible that 6.0 added something that doesn't get cleared when loading 5.5?

I don't think it's a vibration problem because like Zak said, the same program running in the same place on the table is getting different results before and after upgrading.

[Jetty]

I believe the acceleration settings were changed to. They maybe
making it worse.

> > <https://lh3.googleusercontent.com/-hIfYdNKGQBI/UG9ub8-gkTI/AAAAAAAAAE...>

[Matthew Marlowe]

On Fri, Oct 5, 2012 at 6:32 PM, Jetty <clell...@gmail.com> wrote:
> I believe the acceleration settings were changed to. They maybe
> making it worse.
>

That's probably the case. I am printing with 6.0 and getting near
perfect prints.
However, I've noticed/changed:
a) updated acceleration firmware settings - See attached pic with
those that I use.
b) limiting print speed to 60/60/60 with some other minor non-default
skeinforge changes
c) Noticed that somehow prints are more sensitive to how close the
build plate is to the extruder, whenever I'm slightly too close I'm
now seeing tiny little dots showing on the exterior of my prints. I
don't remember seeing the same effect before. Simple fix though and
makes levelling the build plate easy now - just keep printing out
small prints with each having a skirt. If one of the corners is too
close, the exterior wall of those prints will have the dots on the
surface. If the corner is too far, the skirt will not stick in a
straight line or not be perfectly flat. So, all I have to do is
monitor these items on each print and adjust each corner slightly
until I get perfectly level.

Matt

--
https://www.twitter.com/deploylinux

[Moceri]

I'm having the same problem with 6.0 right now.

I've tried several acceleration configurations as suggested from Tony Buser a while back, "Joe's MakerBot", PlanetaryGear, what was suggested by Matt in this thread, and my own.

What I have observed:
     1) no matter the acceleration settings I used, the prints would start out fine for about half or a quarter of the print
     2) after half/quarter of print, the X/Y seem to get very jittery and print quality suffers
     3) I get the same artifacts as Zak
     4) the issue is not isolated within the gcode, I've tried generating the same models in both MakerWare/Miracle Grue and RepG/skeinforge and get the same jittery result. However, the artifacts are less apparent in my skeinforge profile because I reduced my perimeter feed/flow rates to .33.
     5) does not matter if print in in the front or back of the plate/position

A few notes:
     1) I updated the firmware via MakerWare initially
     2) I have been going back and forth between RepG and MakerWare changing the acceleration values
     3) MakerWare has a few more options than RepG, the x/y jerk are separate
     4) this may be an unrelated issue, but possibly connected, when I am connected to RepG I get this message in the console: "You are running an unverified MightyBoard, your machine is not a verified MakerBot Replicator", but I'm still able to operate the bot normally from control panel/preferences
     5) no issue connecting in MakerWare

Attached are a few pics of my results. Would a video suffice instead?

On Friday, October 5, 2012 6:41:42 PM UTC-5, Zak Kus wrote:

[Jetty]

Do you have a link to the items you printed on Thingiverse, wouldn't
mind
having a go at printing them myself.

On Oct 7, 4:16 pm, Moceri <xmoce...@gmail.com> wrote:
> I'm having the same problem with 6.0 right now.
>
> I've tried several acceleration configurations as suggested from Tony Buser

> a while back <http://www.flickr.com/photos/tbuser/7645393408/>, "Joe's
> MakerBot"<http://joesmakerbot.blogspot.com/2012/07/acceleration-and-you.html>,
> PlanetaryGear<http://www.planetarygear.org/2012/09/makerbot-acceleration-settings.html>,

> > <https://lh3.googleusercontent.com/-hIfYdNKGQBI/UG9ub8-gkTI/AAAAAAAAAE...>
>
>
>
>  Untitled-1.jpg
> 71KViewDownload
>
>  Untitled-2.jpg
> 68KViewDownload
>
>  Untitled-3.jpg
> 109KViewDownload

[Moceri]

Do you have a link to the items you printed on Thingiverse, wouldn't
mind
having a go at printing them myself.

Sure, I've attached a .stl that is not available on Thingiverse yet, ("untitled-1.jpg").

"untitled-2.jpg" is of my Sue the T-Rex
"Untitled-3.jpg" is the Companion Cube with a flat bottom (not sure what I scaled it to) and the Bucket of Octopi (from the MakerWare examples)

Would the altshell plugin for SF50 that was just released do any good (any difference than the previous altshell)? Just a thought.

[Moceri]

Also, here is my current profile for SF50 just in case if you or anybody else want to give it a shot.

[Matthew Marlowe]

Moceri,

I printed out the bucket of octopi along with other sea creature
designs last night on replicator running 6.0 - in preparation for a
kids pool party today - no problems at all with any of the prints.
The only time I see dots/exterior blemishes is when the extruder is
too close to the build plate.

Matt

> --
> 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/-/MJsMPOqnM0UJ.
>
> 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.



--
https://www.twitter.com/deploylinux

[Moceri]

The only time I see dots/exterior blemishes is when the extruder is
too close to the build plate.

I'm definitely sure the plate is level. However the issue I'm having is unrelated to the build plate. The stepper motors for the X/Y axis seem to be very jittery/shaky while laying down the shells/perimeter.

Are you noticing this at all during your prints? I'm currently using your acceleration settings, and I'm still getting this effect.

I'm going to try repg39 and activate altshell for SF50 to see if it does anything for me. More later...

[Tyler Pope]

I  also used Makerware to upgrade to 6.0. Could that be it?

--
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/-/Jyv2BjCHmQ4J.

[Moceri]

I  also used Makerware to upgrade to 6.0. Could that be it?

Tyler, try connecting your bot in RepG. Do you see that same error message I'm getting about an "unverified mightboard"? It very well could be a bug from MakerWare.

[Matthew Marlowe]

I've had to tighten the belt on the x axis about every 72hrs of printing lately....If I don't do that, then as the extruder speeds up - there can be minor defects.  However, it only takes 5 minutes to do and is relatively easy.
As far as makerware/replicatorg, I haven't touched makerware at all yet...especially as my desktop runs gentoo linux, and I haven't gotten around to writing a gentoo compatible wrapper around the ubuntu tailored package.

--
https://www.twitter.com/deploylinux                    

 

[Tyler Pope]

Yes I do. I noticed that last night. Makerbot support suggested that I change A and B Axis Junction Jerk in my onboard preferences to 2.0 but I couldn't get the settings to stay.

[Tyler Pope]

Makerbot support sent me this fix:

I'm going to have you download ReplicatorG 39, which just was released a few days ago.

http://code.google.com/p/replicatorg/downloads/list

Once that's installed, re-upload firmware 6.0 from ReplicatorG, regenerate your code, and let me know how the print comes out!

Best,

Alex
MakerBot Support

So far it looks like it fixed the problem on new gcode. I hope old gcode runs without issues. I don't want to regenerate it all again.

On Friday, October 5, 2012 6:41:42 PM UTC-5, Zak Kus wrote:

[Tyler Pope]

It worked on a part with straight walls but when I tried one with curved walls it failed again. I sent my part, gcode and a pic to Makerbot support last night. Hopefully they figure it out.

On Oct 9, 2012 12:26 PM, "Jürgen Gründmayer" <jgruen...@gmail.com> wrote:

Hi Tyler,

I tried this fix, but it didn't solve the problem with the jittery x/y-axis. Did it really work for you? Which operating system did you use?

--
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/-/fCpEUBodjsQJ.

[Jürgen Gründmayer]

I am also in contact with MakerBot Support (Gavin), his last suggestion was to lubricate the Z-rod... Of course this did not change anything.

I've told him that his colleague Alex is dealing with the same issue, I hope this helps that they take the issue seriously.

[Chuck Joga]

I have now experienced this waviness issue also.  I have a piece that has flat sides and a curved end.  The waviness starts part way up.  I thought that something came loose on the machine so I tightened all bolts and tried to reprint.  I got the same results but this time I was watching it when the waviness started.

When it starts, the sound and pattern changes noticeably.  See the pic above.  On the normal part of the print, the extruder head was making a very smooth arc around the end and going very fast on the first shell and then slowing on the outer shell.  When the waviness started, the extruder head started "jerking" around the curve, got louder, and went slower on the first shell.  The outer shell speed seemed to go the same speed as the inner shell and made the same sound.

This is not a problem loose bolts and must have been introduced with the firmware.  I'm going to try the new ReplicatorG 39 first and then try downgrading the firmware to 5.5 to see if there is any change.

-Chuck 

[Chuck Joga]

Let us know if you get any new information form MBI.  I'm convinced that this is a software/firmware issue.  I'm going to try to determine if this is a problem with acceleration or the new firmware.

-Chuck

[Jürgen Gründmayer]

I don't think that it has anything to do with acceleration in general, because you can hear the different sound above a certain z height even with acceleration turned off. Although the effects on the print quality are not so pronounced and you have to look really close.

[Justin Ribeiro]

We're having the same waviness issue with 6.0. We've used both stock RepG 39 and our build of RepG and it hasn't made any difference. We've re-loaded 6.0, we've reset to defaults, tried a number of profiles. The waviness always re-appears.

[Chuck Joga]

Summary: In some circumstances, FW 6.0 introduces the waviness issue that was not present in FW 5.5.  In FW 6.0, waviness presents itself in different degrees whether acceleration is on or off.  It is more detectable when acceleration is turned on and when printing at any speed.  Turning acceleration off and printing at the default slow speed greatly reduces this waviness issue.

I ran some tests with FW 6.0 and FW 5.5 with different combinations of acceleration and feedrate.  All gcode files were generated with Replicator G 37.  All generated objects used defaults from the default tab to switch between default speed/temp and accelerated temp/speed.  Filament diameter was set to 1.74mm and object infill was 3%.

My conclusion is that if you turn on acceleration in FW 6.0, the waviness reported here may be produced using both accelerated and normal speed defaults in Replicator G.  Additionally, very slight waviness can be detected upon very close inspection when acceleration is turned off in FW 6.0 at 40mm/sec print speed.  Using FW 5.5 with acceleration turned on and 40mm/sec print speed shows no detectable waviness.  While FW 5.5 @ 40mm/sec produced no detectable waviness, there are issues with extruder start/stop that seem to have been improved in FW 6.0.

Recommendation if you experience this so called waviness: For best quality, use FW 5.5 @ 40mm/sec and infill of > 10%.  For best print quality using FW 6.0, turn off acceleration and set print speed to 40mm/sec.

-Chuck

--------------------------------------------------------------------------------------------------------------------------------------------------------

Firmware 6.0 vs 5.5

• Acceleration: ON
  
• Feedrate: 80mm/sec
  
• Notes: Waviness in present with FW 6.0 on left.  There is no waviness with FW 5.5 on right.  Dimensional quality of FW 6.0 using acceleration is superior.  Print with FW 5.5 shows distorted/contracted bands at bottom and top.  There were noticeable gaps and dropped loops using FW 5.5.  Extrusion starts were inaccurate with FW 5.5.  "Air printing" was noticeable with FW 5.5 which left gaps and mangled support structures.  Distortion in FW 5.5 print may be due to low infill of 3%.

--------------------------------------------------------------------------------------------------------------------------------------------------------

Firmware 6.0 Feedrate 80mm/sec vs. 40mm/sec

• Acceleration: ON
• Notes:  Decreasing the print speed by 50% using FW 6.0 improved the waviness, but was still noticeable.

--------------------------------------------------------------------------------------------------------------------------------------------------------

Firmware 6.0 Acceleration On vs. Off

• Feedrate: 40mm/sec
• Notes:  Turning acceleration off in FW 6.0 further decreased waviness, but was noticeable only under close inspection.

--------------------------------------------------------------------------------------------------------------------------------------------------------

Firmware 5.5 Feedrate 80mm/sec vs. 40mm/sec

• Acceleration: ON
• Notes: Waviness at 80mm/sec on left was noticeable upon very close inspection.  There was no waviness detected at 40mm/sec print speed on right.  Comparing FW 5.5 80mm/sec print to FW 6.0 40mm/sec print showed slightly less waviness using FW 5.5.  However FW 5.5 print quality was inferior to FW 6.0.

[Zak Kus]

Awesome write-up Chuck, thanks.

A funny thought did occur to me though. I was building thing for a buddy which consisted of a rod, ball, and socket (all of which experience heavy waviness). I had decided to basically ignore the issue (since i was prototyping), however the waviness did actually help the ball/socket maintain good friction that i was going for, and the buddy loved the "textured finished look i included".

The bug is definitely annoying and i cant wait for a fix, but it makes me think about a new category of finishing options. I guess in principle the proper place for something like "turn of skin waviness" would be in whatever cad software you're using, but its an interesting...

-Zak

[Chuck Joga]

This does not work!

I just upgraded to RepG 39 and upgraded the FW from 5.5 to 6.0.  The problem still persists.  :(

-Chuck

[Chuck Joga]

FYI.  The upgrade will overwrite any custom begin/end gcode.  Make a backup before upgrading.  

-Chuck

[Chuck Joga]

Hey Zak,

I'm pressing bearings into this piece and like you said, it the waviness does grip it really well.  :)

-Chuck

[Mark Cohen]

Wow this is just like the old days when they were releasing software that was not QA tested. All they had to do was print something to see there was a problem.

--

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/-/lRC8yRnHeHkJ.

[Matthew Marlowe]

On Thu, Oct 11, 2012 at 4:20 PM, Mark Cohen <markc...@gmail.com> wrote:
> Wow this is just like the old days when they were releasing software that
> was not QA tested. All they had to do was print something to see there was a
> problem.
>

Still seems strange as I must be the only one not seeing this issue on 6.0.

--
https://www.twitter.com/deploylinux

[Chuck Joga]

Count yourself fortunate that you're not hitting this.  Looks like this is some kind of edge case and I'm not sure what the exact scenario is that causes it.  All I know is that there are at least a few examples of people reproducing the problem.  I've been using FW 6.0 since they released it and this is my first encounter. YMMV.

-Chuck

[Finhead]

This wavieness has been happening to my prints on the 6.0 firmware as well. I have tried RepG 0037-0039 as well Makerware all the same if it's running accel. Back to 5.5 and slow printing for me :(  Really a bad deal since the fist 2cm or so print so very nice.

[RocketSled]

Assuming you have a dualstrusion, which extruder are you using?  I've been printing with 6.0 on the right extruder and the quality of my prints have improved from 5.5, and I haven't seen the wavy problem (and the objects I've been printing are fairly large, inches in size).  Last night, I did a job (twice to be sure) with the left extruder and they're *awful*.  Terrible wavyness. I'm using a different color ABS (white vs. black), but I ran a second print with reduced HBP and Extruder temps and filament diameter (down by 0.04mm) and there was no meaningful difference.  

Can't think of a reason why which extruder would matter to the FW, but a possibility none-the-less... so while I might just be wasting more plastic, I'm about to try a rerun on the right extruder, in black.  If that looks good, I'll swap filaments and try one more time with white (though color mattering to the FW is even more outlandish possibility).

[Chuck Joga]

Single extruder for me.

-Chuck

[Mark Cohen]

Dual using right in black

On Oct 12, 2012 2:05 PM, "RocketSled" <rscoh...@gmail.com> wrote:

Assuming you have a dualstrusion, which extruder are you using?  I've been printing with 6.0 on the right extruder and the quality of my prints have improved from 5.5, and I haven't seen the wavy problem (and the objects I've been printing are fairly large, inches in size).  Last night, I did a job (twice to be sure) with the left extruder and they're *awful*.  Terrible wavyness. I'm using a different color ABS (white vs. black), but I ran a second print with reduced HBP and Extruder temps and filament diameter (down by 0.04mm) and there was no meaningful difference.  

Can't think of a reason why which extruder would matter to the FW, but a possibility none-the-less... so while I might just be wasting more plastic, I'm about to try a rerun on the right extruder, in black.  If that looks good, I'll swap filaments and try one more time with white (though color mattering to the FW is even more outlandish possibility).

--

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/-/gcpf7etHV8cJ.

[Mark Cohen]

Thats great. Although it looked very nice on Nandi the Bull, It was driving me crazy.

On Fri, Oct 12, 2012 at 4:56 PM, Ben McCallum <ben.mc...@makerbot.com> wrote:

Hey Everyone,

This is Ben from MakerBot support.

I just wanted to let you all know that we've figured out what's causing this waviness. We're testing a fix and we're going to be sending out very soon. 

Thank you to everyone who pointed this issue out - we really appreciate the heads up. 

-Ben

On Friday, October 5, 2012 7:41:42 PM UTC-4, Zak Kus wrote:

So, i've done this print dozens, maybe even hundreds, of times. I upgraded to 6.0 recently (did some other prints that were successful), but when i printed out this guy it produced a weird wavy effect that i had never seen before. (See picture)

It was the same gcode i had used many times before (so no settings error). I thought it was maybe something funky with my filament (it is sorta old), however the pattern seems very regular and "intentional". All the peices have a little bit of wave, but its most noticeable on the dramatic slopes and curves. 

Has anyone else seen this?

--
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/-/qsiXqpIEwesJ.

[Shawn]

So, what was the problem? Is the fix something we can make simple
changes to work around? Or is this a solid software issue that needs
new f/w and/or software?

Telling me you found the problem is great. Telling me what the problem
was is even better. :)

Shawn

On 12-10-12 02:56 PM, Ben McCallum wrote:
> Hey Everyone,
>
> This is Ben from MakerBot support.
>
> I just wanted to let you all know that we've figured out what's causing
> this waviness. We're testing a fix and we're going to be sending out
> very soon.
>
> Thank you to everyone who pointed this issue out - we really appreciate
> the heads up.
>
> -Ben
>
>
>
> On Friday, October 5, 2012 7:41:42 PM UTC-4, Zak Kus wrote:
>
> So, i've done this print dozens, maybe even hundreds, of times. I
> upgraded to 6.0 recently (did some other prints that were
> successful), but when i printed out this guy it produced
> a weird wavy effect that i had never seen before. (See picture)
>
> It was the same gcode i had used many times before (so no settings
> error). I thought it was maybe something funky with my filament (it
> is sorta old), however the pattern seems very regular and
> "intentional". All the peices have a little bit of wave, but its
> most noticeable on the dramatic slopes and curves.
>
> Has anyone else seen this?
>

> <https://lh3.googleusercontent.com/-hIfYdNKGQBI/UG9ub8-gkTI/AAAAAAAAAEM/Dx1K61KxDb8/s1600/photo.JPG>

[Andrew Plumb]

Translation: It was a firmware issue and they're testing the updated firmware before pushing out an update.

I was seeing the same thing happen half way thru a simple 20mm diameter pumpkin print.  I asked and it turns out the bug manifests pretty consistently at about 10mm into many builds.

…looks like this might have been the commit that fixed this particular issue:

https://github.com/makerbot/MightyBoardFirmware/commit/3842322d2de9f23a94d7083e69c8c2e6c684f86c#firmware/src/MightyBoard/Motherboard/Command.cc

Gotta love open source software development. ;-)

Andrew.

--

"The future is already here.  It's just not very evenly distributed" -- William Gibson

Me: http://clothbot.com/wiki/

[Doogiekr]

Nice find and fix... good thing there is still a community working on these issues =)

[Finhead]

Great you guys found the issue. :) Being brand new to this can I ask is this a fix that can be put into place now or must we wait for a firmware update to eliminate this problem?

If we can use this would anyone be able to explain how to do this? Again great job guys very cool of you all to be on top of the issues so quickly.

[Tyler Pope]

Got it fixed on my machine with the help of Makerbot support. Just reloading 5.5 doesn't fix the issue but resetting the motherboard completely and then loading 5.5 fixed it for me. 

The X offset is still off but I will recalibrate that later.

The same prints that were causing me problems are running without issue now.

[Mark Cohen]

The question I would like answered is if this is so consistant and
obvious, how did the get through QA? Is there no QA anymore?

[Jetty]

Just a heads up. Sailfish doesn't have this problem.

[Zak Kus]

How are you completely wiping the motherboard? My google searches have not been helpful...

[Tyler Pope]

The same window that you use to update the firmware also has a reset the board completely button. 

I'm trying Sailfish out now. It's showing signs of being really good at 120 mm/s.

[Zak Kus]

Yeah a reset + sailfish seems to have taken care of the problem

[RocketSled]

Well, my right hand extruder was producing better prints.  My test object was actually two objects, side by side, one printhead.  Curiously, the further right object printed better, slice-for-slice, than the left object.  But both objects had clear defects in the shell, waves or blobs, couldn't tell which.

I just updated to Sailfish, and am running my first retry of the above print.  It's 75% complete and it looks *perfect*. 

It's great that MB has found their problem and is testing a fix, but you can get Sailfish now and it unquestionably fixes the problem.  And the printer overall seems to be running much better.  Much less mechanical excitation during printing, reducing both the HBP "diving board" effect as well as the noise level.  And I'm running at 50% faster speed than with 6.0.  Absolutely a higher recommended upgrade!

[DB]

Hi Andrew-

Any chance you can instruct on how to implement this fix?  I've got zero programming experience.

Thanks,

Dan

[Chuck Joga]

Hi Ben,

Any word on when the update will be available?

Thanks,

-Chuck

On Friday, October 12, 2012 2:56:44 PM UTC-6, Ben McCallum wrote:

Hey Everyone,

This is Ben from MakerBot support.

I just wanted to let you all know that we've figured out what's causing this waviness. We're testing a fix and we're going to be sending out very soon. 

Thank you to everyone who pointed this issue out - we really appreciate the heads up. 

-Ben

On Friday, October 5, 2012 7:41:42 PM UTC-4, Zak Kus wrote:

So, i've done this print dozens, maybe even hundreds, of times. I upgraded to 6.0 recently (did some other prints that were successful), but when i printed out this guy it produced a weird wavy effect that i had never seen before. (See picture)

It was the same gcode i had used many times before (so no settings error). I thought it was maybe something funky with my filament (it is sorta old), however the pattern seems very regular and "intentional". All the peices have a little bit of wave, but its most noticeable on the dramatic slopes and curves. 

Has anyone else seen this?