Replicator Skeinforge Retraction Settings

[Matthew Marlowe]

All,

The replicator is definitely a tool which needs
practice/experimentation to learn how to make perfect prints. Mine
have been getting significantly better every week since I received it
~6 weeks ago - I'm finally at a stage where I can print w/o a raft
nearly 100% of the time and have the prints come out with confidence
that nearly all defects, if any, will be attributable to the design
rather than the printer or its environment. I'm stuck with two
issues though and I think the first has something to do with the
retraction settings:

1) The best way to explain it is that when the replicator is printing
the floor of an object, it will create the skin for one half of the
floor starting in the middle and stopping just before the side wall,
then it will stop extruding, travel back to the middle and start
printing the other half of the floor. However - it seems to move
1-2mm before it begins extruding again so there is a tiny barely
visible gap left in the floor. I don't see this occurring elsewhere,
but that may just be because it is easiest to see for floors and not
so much with walls which also have a jitter distance. My current
speed settings are 60/60 with retraction also set at 60 with a
retraction distance of 1. Does anyone know what value needs to be
modified to prevent the gaps?

2) Like nearly everyone else, I am fighting corners of objects rising
- but this is only a problem for long prints(6hrs+) and even then,
only worth worrying about with the 9hr+ ones where the bottom of a
large flat object may gain some minor curvature. My build
environment is covered on all sides and the HBP is 110. I've seen
recommendations to increase the HBP temp to 115, add insulation to the
floor, cover air vents in the bottom of the case, or even use hair
spray or other adhesives. I'm not sure what the proper solution is.
I'm leaning against adding any chemicals to the plate, and leaning
towards using the skirt function with layers set to 5 to create a
small insulation protecting wall near the bottom of objects while they
are printed. Is anyone else doing this? I spoke to someone on the
makerbot irc channel, and they recommended against increasing the HBP
temp as the replicator has enough issues with power management as is
and replicatorg has a strong warning against modifying it. Thoughts?
Am I just being timid by avoiding increasing the HBP above 110?

Thanks in advance!
Matt
--
https://www.twitter.com/deploylinux

[Shawn]

For the retraction speed, check your Skeinforge settings under the
Dimensions tab. The "Extruder Retraction Speed" setting is set to
20mm/s by default. If you are using acceleration, you probably want
this at 80mm/s. And you may want to experiment to find the sweet spot
for your machine/environment.

Also on the same tab is the "Retraction Distance", which defaults to
1mm. I haven't had to change this (yet), but have heard that when you
get down to the fine detail, dropping this to 0 or 0.2 might be reasonable.

For the lifting corners - raising your bed temperature is probably the
fastes/easiest fix. I have mine set at 115. At 110, the sides peel too
much and waste prints. Raising the temp immediately solved that problem
(in my case). A warning here though: the HBP is only rated for 120C
max (at least this seems to be the specs I found when I looked for it).
Don't set it above this as you'll possibly do damage, or shorten the
life of the bed.

The other solutions still apply too. I can still get peeling from
cooler air blowing into the chamber, so I make use of at least panels.
Lately my chamber has been heating to 60ish degrees - a little too warm
in my eyes, so I've taken to printing without the hood.

Your environment varies from mine, so you should do what works for you
and your patience level. :)

HTH

[Doogiekr]

I started having the same retraction issue with 5.5 firmware... i tried every combination in the retraction setting to try to fix it but nothing worked. What did work is turning acceleration off in the machines LCD. Of course it means i have to print slower, but it completely fixed my retraction issues so to me it was a fair trade. I will keep it off until the firmware is fixed.

[Adan Akerman]

That is precisely how I would describe my experience.

--
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/-/PdunrqloLUwJ.
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.

[66tbird]

I had the same issue when my nozzle was not burned out perfectly clear and clean and my filament diameter was off. I've never changed retraction. I've noticed some color are more prone than others on the skip. Normally I bump nozzle temp or filament diameter to get perfection.

Chamber temps,, I run 65C on up and have maxed at 80C (my shop can be 45C+ which is my personal 'it's just to hot to be here' limit) and corner lift is not an issue. If fact I design in a place or two with a lift just so I can get the darn prints off. On a side note I haven't cleaned my HBP with acetone in weeks. I set the nozzle/HBP leveling at .11mm (.008' on an standard feeler guage) as my sweet spot. Anymore and it may corner warp, any less and the tape get torn because the print won't budge.

[RocketSled]

On Sunday, September 16, 2012 10:47:46 AM UTC-4, 66tbird wrote:

Chamber temps,, I run 65C on up and have maxed at 80C (my shop can be 45C+ which is my personal 'it's just to hot to be here' limit) and corner lift is not an issue. If fact I design in a place or two with a lift just so I can get the darn prints off. On a side note I haven't cleaned my HBP with acetone in weeks. I set the nozzle/HBP leveling at .11mm (.008' on an standard feeler guage) as my sweet spot. Anymore and it may corner warp, any less and the tape get torn because the print won't budge.

A couple of thoughts come to mind.

The electronics of the MightyBoard carry I suspect mostly a commercial-grade rating.  Commercial electronics are typically rated for 0-70ºC operating temperature. The Replicator sucks air in from above the baseplate, to exhaust over the surface of the MB.  If you have 80ºC temps in the build area, and have not changed the position of the fan, you are running the MB too hot, which is accelerating your electronics to an early and untimely demise.  Heat is the enemy of long term reliability

The largest print I've done is almost 6"x8" (a structural base plate piece for a radio controlled tri-copter).  But I've done a couple of dozen different jobs now (including a job that prints multiple separate parts all in the same run).  I have literally never had a corner lift.  I've replaced the Kapton on my machine a few times now because I'm always tearing it trying to get my prints off the damn HBP.  

I print at 230ºC. I run the HBP at 115ºC.  I have acceleration on, but de-tuned a bit, and print at 60mm/s no problemo.  My machine is not enclosed.  I've done prints from 0.1 to 0.2 layer thickness with equal success. 

I don't really consider any of these things as very special or insightful, but... I did modify SF's settings to print the first layer much slower, like at 50% (I think it was defaulted to 90%).  That made a huge difference in the quality of the first layer.  Particularly for text and smaller diameter circles.  And I also use a steel feeler gauge, a 0.005".   One thing I do do, that seems to be critical to success, is warm up the HBP a good 10 minutes *before* leveling.  And I don't use the built-in utility.  Using RepG, I home Z (move up until it stops) and then step down 0.1mm.  Then I turn off the stepper motor control and move the print head around however I want while I'm doing the leveling.  And from my experience it is much better to be too close than too far away..

I do realize that the prints are also susceptible to lifting if there's a lot of airflow around the HBP.  But I have to wonder if build plate leveling is also a contributing factor?

I've also been thinking about printing my models with a small "undercut" at one corner to help me get something under the print to pry it away...

  

[66tbird]

We have similar projects in RC. I'm currently doing a FPV camera mount on a 525 Quad and a bubblebee shell for that little HK micro quad.

I agree with you, I'm  running my mb to hot. I need to toss an 80mm in or around there soon. But I'm bad about pushing my equipment, Just ask my Lipos. lol.

[Dan Newman]

> A couple of thoughts come to mind.
>
> The electronics of the MightyBoard carry I suspect mostly a
> commercial-grade rating. Commercial electronics are typically rated for
> 0-70ºC operating temperature.

While I do not know which grade is used, this is one of those applications
where automotive grade is likely the better choice: meant for harsher conditions,
namely higher temps and more vibration. All spelled out in the Society of
Automotive Engineering's J1879 standard,

http://standards.sae.org/j1879_200710/

Unfortunately, they charge money for their standards and I don't know offhand of
a source for free copies of that standard.

Dan

[RocketSled]

For electronics, particularly silicon, there's really no difference between C-temp, I-temp, and M-temp parts.  They all come off the same wafers.  But they get sorted at ATE for characteristics that make them better at higher temps. The higher temp parts cost more because they yield fewer parts as they tighten the test criteria.  

I agree that vibration is a factor worth considering.  I haven't looked but would be pleased if the MB is on iso-mounts that damp vibration from the Replicator's structure.  But Surface Mount PCBAs are really very robust.  They're not like leaded components, there's a lot of solder relative to the size of the package connections. If the board's not flexing a lot, it can take a lot of shock and vibe. 

I do rotary and fixed wing RC (just came back from the field, actually).  I too abuse my LiPos.  I don't understand the guys who baby their packs.  Sure, they last longer, but you don't fly nearly as long per pack.  I'd rather spend more time in the air and replace the packs more often.  After all, it's not flying if you're not flying!

[Matthew Marlowe]

All,

Wow...what a diverse and amazing amount of feedback to my questions --
thank you.

As for the retraction settings concern, I'm pretty sure it isn't
related to plate leveling unless I really want to get super aggressive
with the z offset (dropping to under 0.1 from the current ~0.15) ---
I'm getting nearly perfect prints even with just 00.17 - .20 layer
height when I take the 20-30 minutes to do the leveling procedure as
described by rocketsled. I haven't tried to go any lower than 0.17
layers - but the delay in restarting extrusion after travel in certain
cases always happens and is only really evident if one looks for it in
the floor of whatever object is being built. I do have some minor
concerns about the smoothness of the HBP plate - it seems to be
slightly higher midway between all 4 screws and always highest in the
center. l should research what the max variance of the plate should
be.

I also haven't yet tried turning acceleration off.....I really like
the idea of slow changes to the extruder speed rather than having it
fixed and I've already reduced my default accel speed to 60 for flow,
travel, and retract. I'll have to do find a good test to see if my
filament diameter is set correctly...right now I'm setting it to the
max of several readings or avg filament reading + 0.03 whichever is
higher. I may also need to do a test of what happens with retraction
distance is reduced to 0.

With regard to HBP temp, I'm going to test increasing to 112...but the
HBP temp seems to overshoot by up to 4 degrees when warming up and if
120 is the max limit, than perhaps 116 degrees would therefore be the
actual max safe value. It isn't clear what other effects increasing
HBP temp might have, so I'll take baby steps first to 112 and then to
114 if still needed to prevent corner warping on long prints.

Another question did turn up over the last 48hrs and that is more a
question of the impact on the nozzle when one gets aggressive with z
offset and layer height and ones plate itself is not perfectly
smooth..I've seen a number of tiny flakes on the first layer and the
more aggressive I get, the dirtier I find the nozzle after each long
print. I assume there is some kind of trade off between lifetime of
the nozzle and print detail. At the moment, I'm sticking with 0.2
layer height and cleaning off the nozzle after each print while
retaining a z offset near 0.15. I see several recommendations for z
offset of 0.1 or less but I worry that it is going to result in a very
short lived nozzle.

Matt