Running PETG on Cube3

[TommyDee]

You've noticed that you can make Ekocycle print files, right?
... Just select the Ekocycle as your offline printer and yadayadayada...
(Note here that I've seen distinction between ABS clear and ABS tinted in 3DS literature. Clear was noted to run 5 degrees cooler.)

Considering we need to change the file header, this is a process that will involve the file converters.

Export your build file or save your build file directly to disc.
Decode the file (drag and drop on cube-decoder.exe)
Remove the Kanji from the bottom of the generated file. Sometimes you will see Kanji at the top of the file. Just delete with notepad.exe.
Search/replace: M104 S265 and replace with M104 S260
Search/replace: M204 S265 and replace with M204 S260
This makes sure you don't push your Cube3 past the original limit. I cannot say for sure there is any difference, but Ekocycle runs the heaters up to 265*C and Cube3 up to 260*C. To me, this is a safeguard. There are only a couple of calls at this temperature using clear as a material for the slicer.

Change the following header lines:
^Firmware:V1.14B
^Minfirmware:V1.14B
^PrinterModel:CUBE3
^MaterialCodeE1:92 (for instance... this is teal PLA; this must match your cartridge chip for material type) 108 is a coral ABS

The rest doesn't matter.
Save and encode the file, and print on your Cube3 using an SD drive.

Do you want to see what the part looks like in the file? Open it with CubePro

There is the obvious "just change the temperature" option, but I haven't bought into the rumor that Cube3 uses the best of 3DS slicers.
Going through some comparisons of some key printing elements, and experiencing the finer details of how to affect prints, I can say with certainty that the Ekocycle slicer is very much the same as Cube3, but the feed and speed profiles are much more aligned with an earlier CubePro release.

f.PNG

NovaMaker clear PETG printed on Cube3/Lokbuild using the method described above.

g.PNG Cubepro V1.87 opening the converted file (right side of the screen -OPEN-)

[bobskinner]

Tommy! That looks like it turned out quite nicely! Did you have any difficulties with your foray into PETG on the Cube 3?

[TommyDee]

Probably the least remarkable thing I've done on the Cube to date

Thank you for your unrelenting "encouragement"

[bobskinner]

Woohoo! I am glad it worked out. It would have been soul crushing to have recommended it so strongly only to have it suck on the Cube 3...

- - - - - - - - - -

Plus, it's SO SHINY!!!

[TommyDee]

It is a mesmerizing material while it's printing. You can see the heat dissipate as it goes from glassy to matte. Good troubleshooting tool.

But I hear you. I send junk filament to do weed-eater duty

- - - - - - - - - -

I've been printing a good number of things with the PETG running the EkoCycle slicer.

I gave up changing the temperature so it is running at 265*C for a few calls in the print-file.
I just change the 4 header calls and put the file on a USB drive.

The LokBuild is a little touchier about PETG.
I need to keep the build plate very clean; wash with mild dish-soap and dry with a clean towel.
A slightly elevated temperature of the plate helps too. A 10-second stove warming is sufficient.
Larger areas of 1st layer prints helps but if you only have a small contact area, I suggest a little Cube-Glue or similar.
I've put a single streak of glue on the plate and just wipe it with a wet finger after a print and it is ready to go again.

I also set the material in the EkoCycle config to clear since I have clear PETG.
There may be differences between clear and colored 3DS PETG settings.
Regardless, these files are printing nicely.

I've run into one part that prints oversized. I had a similar issue with ABS.
The Bowden tube coupling nut to the nozzle (B2B) is tighter than it should be.
PLA solves this perfectly, but this is a strange phenomena.
Otherwise, I'd say PETG is the perfect union of PLA and ABS for basic printing.

On the downside is unsupported things like shafts. As they get taller, they get weaker.
Compared to PLA, PETG is softer. This causes the print to move with the nozzle.
The result should be obvious.

Now as to the difficulties this type of material substitution. First thing was to pick a chip for PETG.
PLA doesn't get hot enough for purge or test prints. But you will never carbonize the tip with carbon from excessive heat.
ABS may be suitable. The max temp in Cube3 is 260*C and the top end of PETG gets real close to that.
This means purge should work well and even the test print. So an ABS chip seems to be more useful.

- - - - - - - - - -

...and for another consideration for Ekocycle owners:

I'm running my Inland ABS yellow (yes, the same old spool!) using an Ekocycle profile.
Remember that this Inland ABS was terrible at putting little blow-holes all over my prints using the Cube3 ABS slicer.
To solve this, I reduced the temperature and reduced the retractions.
Well, Ekocycle does both of these.
Not as drastic a reduction on the retract as I would like, but still better than what CubePrint does for Cube3 slice files.
Temperatures are perfect, and I suspect many 3rd party ABS will think 245*C is a lot more comfortable than 260*C.
So for Ekocycle owners, just load up some 3rd party ABS that likes 245*C and see what happens.
You may be pleasantly surprised.
As to Cube3 owners... just change those 4 headers in the Ekocycle generated print file and run it on your 3rd party ABS.

- - - - - - - - - -

These are the 3 materials, and they all exhibit different characteristics even though it is the same print.

h.PNG

With regard to how CubePrint v4.03 works with our printers, there are differences in the very same print.
You probably recognize this as the B2B small spool post.
Initially, I printed them with PLA. Perfectly good for the purpose.
I printed ABS nuts to stay inside the B2B cartridges. The PLA post spins those ABS nuts on without any resistance.
Then I printed the some PETG nuts and I have some PLA nuts for the spool-side.

Funny, I print the PETG post and the PETG nuts just lock onto it. They spin on okay at the top, but the lower threads are tight.
I put the ABS nut on there, nice snug fit.
Put the PLA nut on the PETG post; no issue. Not free running but loose enough.
So I printed an ABS post using the PETG print file.
This is such a sweet print! This time the PETG nuts work, a little tight, but acceptable.
PLA nuts, no issue. ABS nut, again, a little tight but acceptable.

This has been a fairly common occurrence in all the time I've been running these printers.
Now we add PETG and it too has its own characteristics.
One of which is "cold welding". This is when 2 materials "meld" together without any added effort.
This is how screws tend to cease. However, it appears that PETG too has a tendency to "stick" significantly.
The engineering fix for this is dissimilar materials. In this case, PETG and ABS make some pretty nice partners if PLA it restricted for your application.

- - - - - - - - - -

Lokbuild and glue...

I've run 10 prints on this small streak of glue. On a stock build plate, the glue would have remained with the part. Lokbuild tends to hold the glue -much- better.

i.PNG

- - - - - - - - - -

I got some Amazon Basics black PETG...

First thing I noticed is that the temperature range is super small... 230-240*C.
Never seen anyone offering anything this specific in range.

It is nicely opaque... surprisingly, actually. Which means that this has a lot of color pellets in the mix.
Sure enough, this is a little fussier. I ran a 235*C print and it came apart like spaghetti.
Then I ran a normal Ekocycle print and it did much better.
But... due to the extra color pellets, it has a much greater tendency to warp as well.
And has more trouble sticking to LokBuild.

Bottom line, you need to treat this filament more like ABS.

The clear PETG I got earlier is much more compliant. This will be a learning adventure

[JohnTee]

This is a great bunch of observations and data points, TD. We are all learning from you. Thanks for that!


On switching between Ekocycle slicer and Cube 3 slicer: I've found the only parameters that must be changed are the PrinterModel and MaterialCodeE1 (or 2). The machine doesn't seem to care about Firmware and Minfirmware as long as they are the same value.


The .cube3 file decoder I am using automagically eliminates the kanji from the front of the file, and I don't bother with that at the bottom - I just leave it. The encoder and machine don't seem to care. This avoids one editing step.


I use Notepad++ to edit the files. It's Regex pattern matching and 'find all in file' make it really easy to find and fix commands.


On cartridge chips: Do you think the machine does something different depending on which chip is installed - i.e., does it execute the bfb instructions differently? Or does it simply check to verify the chip agrees with the "MaterialCodeEx" header value, then just run whatever bfb code is in the file? I haven't tried to check this, but I think I will.


I have printed Ekocycle (PET + coke bottles) filament as ABS in Cube 3, and it did well. It's nice to know you have done it the other way and it works!
------
Have you (or someone) posted a list of the MaterialCode values vs colors somewhere? If not, I have a list that Tomuro provided with the chip programmer.

[TommyDee]

Hello John.

I only know that there is a min-firmware XML file on the Cube's SD. That may only support the CubePrint app. come to think about it.

The garble at the bottom of the file just makes it slow to process. Easy enough to dump even in a 50mb file. I do remembe rthe system just ignoring the extra stuff.
The real take-away is that it does not re-code into a usable image, and that is why we need to use the SD rather than WiFi.

Can't you make a script in Notepad++ to remove the lower section?

ABS is about 10 degrees hotter than PETG. I was going the other way around with great success... 3rd party ABS running an Ekocycle profile.
...and here is why:

Cube3 slicers are hell on retracts. They run an M227 using 450 for the retract speed/distance. This entrained air into the melt when using ABS. ...but 3DS' special formula ABS is tolerant of this motion even after our bowden upgrades. Shapemaker and Inland are -not-. Why such an aggressive move? Because the original cartridge bowden tubes stretched! And they stretched a lot. The aggressive retract is to make up for this... part is the relaxing of the tube, followed by actual small amount of retract. This is the Cube3.
Ekocycle on the other hand stopped M227 at 375 for retract speed and distance. This was enough to solve the issue with Inland. I did something very similar with the Makeshaper ABS, where it went from unusable to perfectly gorgeous prints. So it comes down to the unique slicer used for PETG and the development level that it reached.
What is different in our situation is the stiffer bowden tube. It doesn't stretch as much. I suspect that where our little "blob" comes from in PLA as well. A halving of the retracts in PLA files should also yield smoother seems. Maybe now you are seeing why I'd like to automate this process as much as possible.

Ah!... the chips and materials! Yes, this is the key to understanding the slicers and the available configuration files. 3DS has a serious material plan in the longrun. You can tell it all fell apart, but the hooks are already there. They locked in 4 materials: ABS, PLA, RPET, and WSS (Recycled PET and Washable Support Structures).

Chips are parsed twice; once for the CubePrint 4.03 app and once for the slicer. The config for CubePrint has a profile for every material and within each material, there is an entry for the number of walls (hollow, strong, solid). This provides certain input for the slicer and the app itself. Chips are also parsed for the slicer with a build-config. Here, the material divisions include layer thickness and looses the print quality settings (already established in the app). This makes these two configs easy to understand. And it is these two configs that we want access too to make meaningful changes.

F/W operations based on chips = zero Software operations based on chips = everything

Both config files are basically profile files. They manage multiple profiles with possible external hooks. Structural supports seem to need tweaking so they kept these as external hooks.
The material calls come from the Cubify app. Most of the -exotic- materials were never implemented. I suspect there are hooks in the config files that would open up more materials than the 4 we all know and love. However, it may not matter, because in order to use the exotics you also need a chip to match it. I know we can do that, but that would be a very unique chip which is not what we do generally. We want to keep this simple, right?

So what can you do with one chip? Actually, what can't you do with one chip... you cannot reliably purge or run a test print. You have to be very careful not to burn any filament in the tip as carbon can destroy a tip. With that limitation understood, there are no limits if you only have one chip. The reason I say this is the fact that we have all the profiles in hand! What I am getting at is that in config, you can set PLA to look like anything! Even if we cannot wake up the hidden buttons in the app, we can print a flex profile.

I say this with great confidence from what I've seen so far in the current Cube3 files. And there is light at the end of tunnel with Bob's success in modifying a CubePro config file. If you have the chance, John, take a look at decoding and re-coding the Cube3ConfigurationCube123.XML and Cube3Build_Config_cube_3.XML (Windows programs x86 / 3dsystems folder)
The current decoder destroys the header but Bob gt there somehow. I haven't got it to work reliably yet.

The next step to making the config files accessible is to do the same with CubePro v1.87. That version supports the recoded files and we should be able to mimic Cube3 with the CubePro app by tweaking the configs to look like Cube3. There is also a switch in there that I'd like to try... "Cube3 true"

If you can offer guidance on this config conversion, life will move forward.

And yes, I've seen the material codes somewhere. I just need to remember 2... 92 (PLA) and 108 (ABS) I just need to remember "200"

- - - - - - - - - -

There is one other notable aspect of the config files... seems there is distinction between all-caps variables and conventional variables. I've seen this to mean which variables are truly user variable and others are system settings.

- - - - - - - - - -

Notepad++ script...

replace 4 instances common on all files
look for M107
move cursor to the end of only "M107" instance
delete to EOF

- - - - - - - - - -

Thanks John for getting me the shortcut to reach the EOF() with a 3-finger'd keystroke. Works wonders in notepad and notepad++.

How do you tell Notepad++ to delete the current line in script?

Anyway, back to printing the black Amazon.com PETG. This is the stuff that curled on me about as much as ABS does on CubeGlue/LokBuild.
Again, suspecting the color pellets for the less than stellar performance, I decided to try an ABS profile and modifying the following:
Temp: M104 S245 (was S260)
Retract: M227 P350/P300 (was 450/400) S350/S300 (was 450/400)
whateverthisdoes: M228 P0 S350/S300 (was S450/S400)

This time I am printing the B2B printhead nozzle with the purge tower (cooling tower).
Little chance of warping with this footprint. The idea is to dial in the temperature and the retractions for this particular filament.
I will have to treat this stuff a lot more like ABS.
I am not sure if there is a significant difference between an ABS slicer profile and an Ekocycle slicer profile.
I'll have to look into that now that I think of it.

impressive!
d.PNG
...and not so much!
e.PNG
245 degrees using an ABS profile... and some seemingly extremely sticky glue.
but it did have some heat related issues too:
f.PNG

Now that print upset me a lot! So I went back and made a Ekocycle of just the nozzle housing.
No purge tower this time so apples and oranges do apply.

But with ABS prints, I've never been able to print this part as a lone part...
guess what, PETG can!

g.PNG
Insanely accurate! Thread was perfect. I needed to sand down the inner wall for the ABS core, but hey, this is amazing!

So amazing as a matter of fact that I am also running the MakeShaper ABS to see if this is a simple fluke.
Now I am really curious as to which of the settings below are responsible for the huge difference.

- - - - - - - - - -

I like this, and I am also quite surprised at how much difference there is between the ABS and PETG settings:
please note that this forum does not always faithfully reproduce the passed values leaving out or adding minus symbols (-).
Please be careful cutting and pasting from these posts.

The values below are based on Cube3 V4.03 configuration file. There are 156 entries in both and their variable names are all consistent.
However, it seems half the values are different; some a small difference and some significant. Have a look. First value is ABS and second is RPET (Recycled PET) ; the T/F logic should be obvious:

CUBE_LAYER_THICKNESS		0.1925	0.1925	TRUE
LayerStartingOffset		0.22	0.32	FALSE
QuickLayerPause		-1	-1	TRUE
ZJumpDistance		2	2	TRUE
ZJumpVelocity		60	60	TRUE
ZJumpMinVectorThreshold		5	5	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_TEMP		260	245	FALSE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_BORDER_EXTRUDER_RPM		40	40	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_FILL_EXTRUDER_RPM		36	36	TRUE
BoundaryEndPoinRetraction		0.05	0.05	TRUE
BoundaryEndPoinRetractionExtra		0.05	0.05	TRUE
QuickLayerPartFanOn		TRUE	TRUE	TRUE
PartFanOnLayer		0	10	FALSE
QuickLayerTemperature		260	250	FALSE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_BORDER_SPEED		1800	1800	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_FILL_SPEED		1800	1800	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_SUPPORT_SPEED		1250	1250	TRUE
CUBE_INITIAL_DWELL_TIME		1	1	TRUE
CUBE_HATCH_RETRACTION		0.3	0.4	FALSE
CUBE_FILL_RETRACTION		0.15	0.1	FALSE
CUBE_NEW_RETRACTION		TRUE	TRUE	TRUE
CUBE_MIN_LAYER_TIME		23	23	TRUE
CUBE_MIN_LAYER_DIST		10	13.5	FALSE
CUBE_JUMP_SPEED		8000	8500	FALSE
CUBE_DRAW_BORDER_SPEED		1800	1500	FALSE
CUBE_DRAW_SUPPORT_SPEED		2500	2000	FALSE
CUBE_DRAW_HATCH_SPEED		3850	2500	FALSE
CUBE_DRAW_FILL_SPEED_UP		2350	2000	FALSE
CUBE_DRAW_FILL_SPEED_DOWN		2350	2000	FALSE
CUBE_DRAW_RAFT_SPEED		1050	1050	TRUE
CUBE_DRAW_RAFT_PLATFORM_SPEED		600	900	FALSE
CUBE_DRAW_RAFT_PART_SPEED		2100	2100	TRUE
CUBE_DRAW_SHORT_SPEED		2500	2000	FALSE
PrimeSteps		450	375	FALSE
ReverseSteps		450	375	FALSE
PrimeExtruder		600	400	FALSE
ReverseExtruder		800	800	TRUE
PrimeHoldOff		0	0	TRUE
ReverseHoldOff		450	375	FALSE
FanPWM		100	100	TRUE
CUBE_EXTRUDER_PREWARM_TEMP		245	240	FALSE
CUBE_EXTRUDER_BUILD_TEMP		260	250	FALSE
CUBE_EXTRUDER_STANDBY_TEMP		210	210	TRUE
CUBE_EXTRUDER_SPIKE_TEMP		260	265	FALSE
CUBE_DRAW_BORDER_EXTRUDER_SPEED		21	18	FALSE
CUBE_DRAW_SUPPORT_EXTRUDER_SPEED		32	25	FALSE
CUBE_DRAW_HATCH_EXTRUDER_SPEED		65	30	FALSE
CUBE_DRAW_FILL_EXTRUDER_SPEED_UP		36	30	FALSE
CUBE_DRAW_FILL_EXTRUDER_SPEED_DOWN		36	33	FALSE
CUBE_DRAW_RAFT_EXTRUDER_SPEED		29	32	FALSE
CUBE_DRAW_RAFT_PLATFORM_EXTRUDER_SPEED		29	35	FALSE
CUBE_DRAW_RAFT_PART_EXTRUDER_SPEED		29	32	FALSE
CUBE_DRAW_SHORT_EXTRUDER_SPEED		14	20	FALSE
CUBE_FILL_DOWN_ANGLE		-45	-45	TRUE
CUBE_FILL_DOWN_SPACING		0.6	0.6	TRUE
CUBE_FILL_SPACING		0.55	0.55	TRUE
CUBE_MIN_FILL_VECTOR		0.75	1.25	FALSE
QuickLayerPartFanPWM		100	100	TRUE
QuickLayerJumpSpeed		7000	7000	TRUE
PreToolChangeReverseNotRetractingSteps		750	700	FALSE
PreToolChangeReverseRetractingSteps		300	350	FALSE
PostToolChangeReverseNumberofSteps		10	10	TRUE
ToolChangeReverseRPM		500	500	TRUE
ToolChangePurgeNumberofSteps		750	700	FALSE
ToolChangePurgeRPM		50	50	TRUE
MaintReverseNotRetractingSteps		75	350	FALSE
MaintReverseRetractingSteps		50	150	FALSE
PurgeStripRPM		40	40	TRUE
PurgeStripOffset		1.8	1.8	TRUE
PurgeStripSpeed		800	800	TRUE
GlueDryTime		60	60	TRUE
SeamToBack		TRUE	TRUE	TRUE
PurgePattern		2	2	TRUE
PrimeBorderFirst		TRUE	TRUE	TRUE
PurgeRaftSpeed		827	827	TRUE
PurgeRaftRPM		45	45	TRUE
SmartHatchandFillPath		TRUE	TRUE	TRUE
RetractMinThreshold		3	4	FALSE
DiamondRetract		0.05	0.05	TRUE
SpanningUpFillSpeed		2000	2000	TRUE
SpanningUpFillRPM		38	38	TRUE
SpanningDownFillSpeed		2000	2000	TRUE
SpanningDownFillRPM		38	38	TRUE
SpanningDownFillSpacing		12	12	TRUE
SpanningDownFillSpacingOnWSS		12	0.6	FALSE
AccelStart		0	0	TRUE
DecelStop		0	0	TRUE
AccelRate		5000	5000	TRUE
DecelRate		5000	5000	TRUE
AccelTheshold		1000	1000	TRUE
RepurgeCount		50	10	FALSE
StartupPurgeSteps		600	700	FALSE
StartupPurgeRPM		50	35	FALSE
QLHatchLimit		0.5	0.5	TRUE
QLFillLimit		0.8	0.8	TRUE
QLBoundaryLimit		0.5	0.5	TRUE
MiniReverseSteps		10	10	TRUE
MiniReverseRPM		50	50	TRUE
SlowHatchSpeed		3000	3000	TRUE
NumOfSlowHatchLayers		3	3	TRUE
RetractMinThresholdSupport		20	20	TRUE
CUBE_DRAW_HATCH_EXTRUDER_SPEED_HEX		40	28	FALSE
CUBE_DRAW_HATCH_EXTRUDER_SPEED_DIAMOND		40	28	FALSE
AdjustQuickLayerSpeed		TRUE	TRUE	TRUE
MulticolorTempOffset		0	0	TRUE
FirstLayerRPMBorder		40	37	FALSE
FirstLayerRPMBorderOnABS		21	21	TRUE
FirstLayerRPMBorderOnPLA		21	21	TRUE
FirstLayerSpeedBorder		1302	1800	FALSE
FirstLayerSpeedBorderOnABS		1800	1800	TRUE
FirstLayerSpeedBorderOnPLA		1800	1800	TRUE
FirstLayerTempBorder		260	250	FALSE
FirstLayerTempBorderOnABS		257	255	FALSE
FirstLayerTempBorderOnPLA		257	255	FALSE
FirstLayerRPMSupport		34	34	TRUE
FirstLayerRPMSupportOnABS		34	34	TRUE
FirstLayerRPMSupportOnPLA		34	34	TRUE
FirstLayerSpeedSupport		700	500	FALSE
FirstLayerSpeedSupportOnABS		2500	2500	TRUE
FirstLayerSpeedSupportOnPLA		2500	2500	TRUE
FirstLayerTempSupport		257	250	FALSE
FirstLayerTempSupportOnABS		257	255	FALSE
FirstLayerTempSupportOnPLA		257	255	FALSE
FirstLayerRPMFill		40	29	FALSE
FirstLayerRPMFillOnABS		36	36	TRUE
FirstLayerRPMFillOnPLA		36	36	TRUE
FirstLayerSpeedFill		1800	1400	FALSE
FirstLayerSpeedFillOnABS		2350	2350	TRUE
FirstLayerSpeedFillOnPLA		2350	2350	TRUE
FirstLayerTempFill		260	250	FALSE
FirstLayerTempFillOnABS		260	255	FALSE
FirstLayerTempFillOnPLA		260	255	FALSE
RaftPlatformSpacing		3	3	TRUE
RaftSpacing		2	2	TRUE
RaftInterfaceSpacing		1	1	TRUE
ExtruderCurrent		450	500	FALSE
MinLineHatchLength		2	3	FALSE
QLPrimeSteps		400	300	FALSE
QLReverseSteps		400	300	FALSE
QLPrimeExtruder		750	400	FALSE
QLReverseExtruder		1000	800	FALSE
QLPrimeHoldOff		0	0	TRUE
QLReverseHoldOff		400	300	FALSE
MinIslandArea		2	2	TRUE
ShortFillLength		3	3	TRUE
ShortFillRPM		27	20	FALSE
SkirtFillSpeedFirst		1299	1400	FALSE
SkirtFillRPMFirst		33	29	FALSE
SkirtFillSpeedSecond		1800	1700	FALSE
SkirtFillRPMSecond		37	29	FALSE
SmallestFill		0.7	1.1	FALSE
SmallestFillRPM		32	20	FALSE
SmallestFillSpeed		2350	2000	FALSE
ToolChangeStandbyTempWait		FALSE	FALSE	TRUE

- - - - - - - - - -

THESE ARE ALL THE ENTRIES THAT ARE DIFFERENT (FALSE from list above): ABS on the left and RPET on the right.

LayerStartingOffset		0.22	0.32
CUBE_DRAW_NO_RAFT_FIRSTLAYER_TEMP		260	245
PartFanOnLayer		0	10
QuickLayerTemperature		260	250
CUBE_HATCH_RETRACTION		0.3	0.4
CUBE_FILL_RETRACTION		0.15	0.1
CUBE_MIN_LAYER_DIST		10	13.5
CUBE_JUMP_SPEED		8000	8500
CUBE_DRAW_BORDER_SPEED		1800	1500
CUBE_DRAW_SUPPORT_SPEED		2500	2000
CUBE_DRAW_HATCH_SPEED		3850	2500
CUBE_DRAW_FILL_SPEED_UP		2350	2000
CUBE_DRAW_FILL_SPEED_DOWN		2350	2000
CUBE_DRAW_RAFT_PLATFORM_SPEED		600	900
CUBE_DRAW_SHORT_SPEED		2500	2000
PrimeSteps		450	375
ReverseSteps		450	375
PrimeExtruder		600	400
ReverseHoldOff		450	375
CUBE_EXTRUDER_PREWARM_TEMP		245	240
CUBE_EXTRUDER_BUILD_TEMP		260	250
CUBE_EXTRUDER_SPIKE_TEMP		260	265
CUBE_DRAW_BORDER_EXTRUDER_SPEED		21	18
CUBE_DRAW_SUPPORT_EXTRUDER_SPEED		32	25
CUBE_DRAW_HATCH_EXTRUDER_SPEED		65	30
CUBE_DRAW_FILL_EXTRUDER_SPEED_UP		36	30
CUBE_DRAW_FILL_EXTRUDER_SPEED_DOWN		36	33
CUBE_DRAW_RAFT_EXTRUDER_SPEED		29	32
CUBE_DRAW_RAFT_PLATFORM_EXTRUDER_SPEED		29	35
CUBE_DRAW_RAFT_PART_EXTRUDER_SPEED		29	32
CUBE_DRAW_SHORT_EXTRUDER_SPEED		14	20
CUBE_MIN_FILL_VECTOR		0.75	1.25
PreToolChangeReverseNotRetractingSteps		750	700
PreToolChangeReverseRetractingSteps		300	350
ToolChangePurgeNumberofSteps		750	700
MaintReverseNotRetractingSteps		75	350
MaintReverseRetractingSteps		50	150
RetractMinThreshold		3	4
SpanningDownFillSpacingOnWSS		12	0.6
RepurgeCount		50	10
StartupPurgeSteps		600	700
StartupPurgeRPM		50	35
CUBE_DRAW_HATCH_EXTRUDER_SPEED_HEX		40	28
CUBE_DRAW_HATCH_EXTRUDER_SPEED_DIAMOND		40	28
FirstLayerRPMBorder		40	37
FirstLayerSpeedBorder		1302	1800
FirstLayerTempBorder		260	250
FirstLayerTempBorderOnABS		257	255
FirstLayerTempBorderOnPLA		257	255
FirstLayerSpeedSupport		700	500
FirstLayerTempSupport		257	250
FirstLayerTempSupportOnABS		257	255
FirstLayerTempSupportOnPLA		257	255
FirstLayerRPMFill		40	29
FirstLayerSpeedFill		1800	1400
FirstLayerTempFill		260	250
FirstLayerTempFillOnABS		260	255
FirstLayerTempFillOnPLA		260	255
ExtruderCurrent		450	500
MinLineHatchLength		2	3
QLPrimeSteps		400	300
QLReverseSteps		400	300
QLPrimeExtruder		750	400
QLReverseExtruder		1000	800
QLReverseHoldOff		400	300
ShortFillRPM		27	20
SkirtFillSpeedFirst		1299	1400
SkirtFillRPMFirst		33	29
SkirtFillSpeedSecond		1800	1700
SkirtFillRPMSecond		37	29
SmallestFill		0.7	1.1
SmallestFillRPM		32	20
SmallestFillSpeed		2350	2000

- - - - - - - - - -

Again, notice the fully capitalized variables v. the non.
Direct g-code v. modifiers?
These data are what are passed onto the slicer.
Notice that fully capitalized lines read like primary inputs to the slicer.
It should be very easy to see how these are applied in the final output file.

Call > ABS > PETG > match?

CUBE_LAYER_THICKNESS		0.1925	0.1925	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_TEMP		260	245	FALSE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_BORDER_EXTRUDER_RPM		40	40	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_FILL_EXTRUDER_RPM		36	36	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_BORDER_SPEED		1800	1800	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_FILL_SPEED		1800	1800	TRUE
CUBE_DRAW_NO_RAFT_FIRSTLAYER_SUPPORT_SPEED		1250	1250	TRUE
CUBE_INITIAL_DWELL_TIME		1	1	TRUE
CUBE_HATCH_RETRACTION		0.3	0.4	FALSE
CUBE_FILL_RETRACTION		0.15	0.1	FALSE
CUBE_NEW_RETRACTION		TRUE	TRUE	TRUE
CUBE_MIN_LAYER_TIME		23	23	TRUE
CUBE_MIN_LAYER_DIST		10	13.5	FALSE
CUBE_JUMP_SPEED		8000	8500	FALSE
CUBE_DRAW_BORDER_SPEED		1800	1500	FALSE
CUBE_DRAW_SUPPORT_SPEED		2500	2000	FALSE
CUBE_DRAW_HATCH_SPEED		3850	2500	FALSE
CUBE_DRAW_FILL_SPEED_UP		2350	2000	FALSE
CUBE_DRAW_FILL_SPEED_DOWN		2350	2000	FALSE
CUBE_DRAW_RAFT_SPEED		1050	1050	TRUE
CUBE_DRAW_RAFT_PLATFORM_SPEED		600	900	FALSE
CUBE_DRAW_RAFT_PART_SPEED		2100	2100	TRUE
CUBE_DRAW_SHORT_SPEED		2500	2000	FALSE
CUBE_EXTRUDER_PREWARM_TEMP		245	240	FALSE
CUBE_EXTRUDER_BUILD_TEMP		260	250	FALSE
CUBE_EXTRUDER_STANDBY_TEMP		210	210	TRUE
CUBE_EXTRUDER_SPIKE_TEMP		260	265	FALSE
CUBE_DRAW_BORDER_EXTRUDER_SPEED		21	18	FALSE
CUBE_DRAW_SUPPORT_EXTRUDER_SPEED		32	25	FALSE
CUBE_DRAW_HATCH_EXTRUDER_SPEED		65	30	FALSE
CUBE_DRAW_FILL_EXTRUDER_SPEED_UP		36	30	FALSE
CUBE_DRAW_FILL_EXTRUDER_SPEED_DOWN		36	33	FALSE
CUBE_DRAW_RAFT_EXTRUDER_SPEED		29	32	FALSE
CUBE_DRAW_RAFT_PLATFORM_EXTRUDER_SPEED		29	35	FALSE
CUBE_DRAW_RAFT_PART_EXTRUDER_SPEED		29	32	FALSE
CUBE_DRAW_SHORT_EXTRUDER_SPEED		14	20	FALSE
CUBE_FILL_DOWN_ANGLE		-45	-45	TRUE
CUBE_FILL_DOWN_SPACING		0.6	0.6	TRUE
CUBE_FILL_SPACING		0.55	0.55	TRUE
CUBE_MIN_FILL_VECTOR		0.75	1.25	FALSE
CUBE_DRAW_HATCH_EXTRUDER_SPEED_HEX		40	28	FALSE
CUBE_DRAW_HATCH_EXTRUDER_SPEED_DIAMOND		40	28	FALSE

- - - - - - - - - -

Okay, now I am really confused

h.PNG

Talk about -in-your-face- ! -

Remember all that fuss and work to get Cube3 ABS profile to print 3rd party ABS -correctly-?

Well, here it is in all its glory, USE EKOCYCLE!

I took the same file I printed the black PETG with and printed this MakeShaper natural ABS. And it is PERFECT!
The only thing that got in my way was the different offset for PETG. -Very- light touch to the build plate on my machine.

- - - - - - - - - -

I'm getting close to tweak all the B2B parts for PETG printing.
Down to the bulkheads right now.
Amazing how sticky threads can get in PETG!

Also a quick updated cartridge saver... Now has the Cube3 logo.

cartsaverii.PNG
https://www.thingiverse.com/thing:2931422

- - - - - - - - - -

This is insane!

Everyone that has ever tried to create Marlin on a Cube3 has failed without extreme measures.
ABS Marvin turns his head into a pile of goo.
PLA Marvin has blobitis along with hook-deformities.
PETG Marvin...


b.PNG

Perfection!

This Marvin looks as good as the ABS Marvin I can print from 3D Builder.
That means at one time, the ABS formula for Cube3 was pretty much dialed in.
Makes one wonder what happened.

- - - - - - - - - -

Another week in on making black PETG parts with the Amazon filament. This stuff curls! Just to get this out there. The NovaMaker clear is much more "relaxed" on the print bed.

I have been making a very consistent change in all my print files for running this PETG. Over a dozen applications so far and absolutely excellent prints.
One caveat; PETG, 3DS ABS, and PLA all come out slightly different in geometry! This means that if something fits with PLA, it may not fit with PETG.
This is not new, however. I've been chasing this issue with the Cube3 when it comes to mating ABS parts together that work perfectly with default PLA prints.

First of all I print using the Ekocycle profile in CubePrint 4.03. I export the file and decode it.
I remove the image from the bottom (find the last M107 and delete to EOF - select end of M107 and ctrl-shift-end del)
Replace the 4 header requirements; firmware, platform, and material(s)
Remove the motion from the wiper box (second G1 call moving to X-95.500...)
Change the velocity of the first G1 call onto the plate from 8500 to 5000 (G1 X-nnn Y-nnn Znnn F8500.0)
Change all the temperature calls to 250 max (M104 S265 > M104 S250 & M204 S265 > M204 S250)
Encode and use a USB drive to print.

Without any other setting changes, these prints exhibit no blobs and prints reliably. If you have your gap set to perfection, you also get a beautiful blend of the first few layers with the rest of the print. This is something I've never achieved reliably with PLA or ABS settings.

Again, be aware! Each material has a different initial offset -beyond- the gap you set. So if your gap is a healthy 250um, it will add whatever the first Z-setting is in your print file. In the case of PETG, this setting is 0.32mm. For PLA it is 0.22 and for ABS it is 0.25. The difference for PETG is significant. It is actually enough to add blue tape to your build plate to make up the difference if you want to keep a common gap setting for all materials.

The Amazon PETG needs glue! Although I've made a few prints directly on LokBuild, the likelihood of it staying there is dubious at best. Not so with the NovaMaker PETG. It likes LokBuild just fine. I've been using a blue tape known for a stupidly tenacious hold. And to that I add some watered down Cube Glue. The only squirming the print can do now is to peel up the adhesive on the back side of the tape as the material curls. For round objects, this is not a huge issue. But do treat this particular material similar to ABS.

[buddybu]

Hi TD,

First of all I print using the Ekocycle profile in CubePrint 4.03. I export the file and decode it.
I remove the image from the bottom (find the last M107 and delete to EOF - select end of M107 and ctrl-shift-end del)
Replace the 4 header requirements; firmware, platform, and material(s)
Remove the motion from the wiper box (second G1 call moving to X-95.500...)
Change the velocity of the first G1 call onto the plate from 8500 to 5000 (G1 X-nnn Y-nnn Znnn F8500.0)
Change all the temperature calls to 250 max (M104 S265 > M104 S250 & M204 S265 > M204 S250)
Encode and use a USB drive to print.

These velocity, movement and temperatures you change are for EKO -> PETG, correct? Do you make any similar changes for PLA and ABS

Temperature

EKO	PETG	PLA	ABS
265	250	?	?

Velocity -- does this apply to X, Y and Z? Or to F?

EKO	PETG	PLA	ABS
8500	5000	?	?

For the motion call do you remove it for all PLA, PETG, ABS? And is it G1 and G2 or only G1?

I know so many noob questions.

[TommyDee]

I should clarify some of these actions, yes. I was afraid of running the Cube3 at 265 at first. This has not proven to be an issue. I was also waffling between 245 and 250*C and found the default Ekocycle to work just fine for PETG and 3rd party ABS.
I also leave the PLA profile alone... and for 3DS ABS, I leave its profile alone.

The velocity change is only for the initial trek to the start point. A bit more of a chance to get a good start tack on the build plate.

I also remove the call just outside the wiper because it adds to drool-time.

Header changes should be pretty self-explanatory as it has to do with what printer is accepting the file.

I am glad to see others working down to this level of print management. So far, the only real things that make a difference, for explained reasons, is retraction and temperature. Retraction in the Cube series profiles is based on a stretching bowden tube. That thin stock tubing can stretch up to 10mm (3/8") with ease. This is what provides the drool-preload. With the stiffer 4mm bowden tubes, there is no stretch so the compensation in the profiles do benefit from tweaking this value. There are the M227 and M228 calls. I only mess with the first 2 parameters in the M227 call. I can cut them nearly in half and only marginally increase stringing. The Ekocycle settings are much more in line with the new tubes than the Cube3 profiles. I can run Ekocycle prints on PETG or 3rd party ABS with changing only the header info, again, running on a Cube3 platform.

Now here is where I hadn't gone yet... I can see how a unbalanced M227 call could help. One value is the retract, and the second is a resume of sorts. They are always alike. But if you want to remove that head-pressure, offsetting one from the other may reduce the blob syndrome even more without inducing stringing.

One of the difficult things in the aftermath of slicing is to surgically changes speed & feed values to something useful. I've mapped these things out on graphs only to find it daunting to tweak. The only reference help you have is the "Vector" calls which state which type of extrusion is happening in that section. This is why I want to get in front of the slicer and change the configs for this type of recipe enhancement.

Also remember that the Cube3/Ekocycle is a special printer in that it shares a heater-block for 2 extruders. There is actually a bias temperature assigned to the second extruder. I hadn't tried to change a dual-color print yet, but it should be possible to carefully modify the temperatures with some level of sanity. However, if you try this with 3rd party slicers, you might be in for some effort in managing extruder swaps per layer.

I've dug a long way into the working of the Cube3/Ekocycle slicer code as well as the CubePro code and I find a lot of possibilities there. Let me know what you would like more information on and I can share what I have. Obviously, implementing a 3rd party slicer is on top of the list but even the stock slicer has more options than one would first think. Making the changes is where we are stuck right now. For Cube3/Ekocycle, we are probably ahead of the CubePro with the file conversions as we can manage the latest version. But do drop some of those config(dot)XML files onto the decoder... you might like what you see in there. Basically, they consist of many profiles dialed in for the app selections made for slicing. There you can change many aspects of the slicer for every different kind of extrusion profile. Recompiling the retraction settings alone would safe tons of time in editing files and can print from the CubePrint app. Basically, you can manage recipes by changing config files out. I've done one apparently successful recompile of a config file. You have to reconstruct the header before encoding it. My test was to simply decode and recode the same file after fixing the header-line to see if they were still identical, and they were.

[mperroud]

Hello everyone, I was able to successfully print PETG. from simplify 3D I made a 2 layer raft with PLA (printed at 220ºC) after that, change the filament to PETG and continue printing at 235ºC without layer fan.