-
You need Win10 and absolute virgin install of the toolkit. It is like cat experiment thing... its both dead and alive until you look at it. And when you do, it no longer registers as a connected Cubify printer. Touchy as all getout! But I can connect (right side port of course). I've used it. But I break the driver the moment I query it.
Not thrilled being dongle'd to the printer with my laptop. Needs a way around that too ;p
Thing is, 3D Builder is exactly the same slicers* as CubePrint. I've inspected the lot and yep, 3DB is a new front-end for the old slicer. *=outdated allowing rafts.
-
I have posted a new version of the cubeutils with cube3editor. This new version, when you click "Generate Cube3" will generate the combined file with updated index file and cube3.
I have not tried using the script to generate a file, but the save routine for both UI and command line is the same process. The end file is a little smaller since the included XML file has some of the whitespace removed.
Link to updated files
- - - - - - - - - -
One more thought. This implementation only updates the type, materials and CRC tags. I haven't added any code for support/sidewalks...
-
Got it. That material matching for supports and sidewalks could be a manual edit in the "RAW BFB". If it was the last thing that kept WiFi from working - because the app couldn't parse the file... well then now you know where to look ;p
-
Okay, last update for today. :) But there could be more. I modified the RawBFBView to update the in memory BFB with changes that are made in the editor, so no need to save the BFB, encode as a CUBE3. You can make your changes, and then just close the editor and they will be applied automatically. I suppose I should add an apply button so that you can force the changes to be applied and allow changes to be discarded if necessary. That's tomorrow's change. :)
You get it here.
-
Nice! Yes, that is part of the workflow i cannot see. When things are committed. But that is what is helping me to stumble around in the dark to see what I'd 'like' to do. You certainly caught a glimpse of this in this update. Excellent work!
-
Okay, here is the Raw BFB Editor with the apply button to allow for live updating.
or: https://drive.google.com/open?id=1gd...mhQ_f28-DhOI6e
- - - - - - - - - -
Sidewalks and supports can only be set to left or right, not simultaneously. And it gets set to the color on that side. So, if your Extruder1 color is Dark Grey(101) and your Extruder2 is Yellow PLA (85), and support is set to Dark Grey and sidewalk is set to Yellow, then Supports is 101 and Sidewalks is 85.
-
Yes, the logic is certainly there. And if either or both is -1, those are left alone.
The challenge of course is to know which color was assigned originally to E1 and E2 and lock any changes reliably to the sidewalk/support settings when not -1.
- - - - - - - - - -
Okay, give me a Use-Case that I can throw at this thing... anybody.
The most logical one to pursue is PETG to PLA (EKOCYCLE to CUBE3) as I am really starting to like that slicer a lot more
seeing how it is implemented with more finesse than the PLA slicer. Of course, EKOCYCLE is limited in options. No Infinity rinse; no 70-micron prints; - but everything I print with PETG using this slicers comes out very nice with only a few tweaks! The toolkit is perfect for this type of thing. I appreciate all the work you are putting into this. My goal is to contribute through some characterization of the variety of slicers we do have in BFB format, primarily the various Cubify slicers.
Do any work in Cura?
-
Interesting. So, I see what you mean about the color. If I change the left extruder color from red to blue and I had sidewalks on for the left extruder, then I will need to update the left extruder color too.... Hmm. Yeah, the logic I see is this:
Code:
if I change extruder n color then
if sidewalks is not -1 and sidewalk color equals extruder n original color then
update sidewalk color to match extruder n new color
end if
if supports is not -1 and supports color equals extruder n original color then
update supports color to match extruder n new color
end if
end if
I have done a little bit with Cura, but it is a bit overwhelming. Do you have a basic setup for cura?
-
6 Attachment(s)
Yes... correct on the logic. Loop that twice for each extruder and you got it. That's why it is good to use in the color parsing algo.
Cura is missing something I really like in Cubify... a type of internal wall support when the overhand goes negative... say upper hemisphere.
However, Cura has a BFB output with post-processing. Worth noting ;p
- - - - - - - - - -
I am just stumped at how useful this little tool can be. I continued my conversion 'therapy' with a project I am currently working on; a tray for Argentine ant bait.
These little pests are unbelievably prolific. They once found a hummingbird feeder and completely devoured it! Hence my hummingbird feeder trap on Thingiverse.
Well, I have since learned how you must deal with these pests. Something about boric acid at levels under 2% with whatever you can get the ants to come to. Well, I've learned that there is very little these ants won't come to... but certain things help. I've decided that a 1% solution to 49.5% sugar and 49.5% water would produce a bait that will not exceed 2% and is nominally delivered at 1%. My first tests were to fill bottle caps. Unfortunately I've killed a lot of worker-ants that should be bringing the bait back to the hive. With this particular tray, the ants have a much better chance to swarm the tray without drowning. This is kind of like drowning in beer for an alcoholic. A good friend of mine had a pretty good joke about that.
Anyway, I printed 2 of the trays using CubePro v2.02 as a baseline. The Cubepro does a pretty nice job of this piece. One thing Cubepro doesn't do well is make for a solid bottom. You'll see in the images that the fill for the bottom is fairly sparse. I suspect they did this to help with warping but that's just a guess. In my world, I like solid bottoms even if you can see every trace. This baseline makes for very functional parts.
Attachment 3319
300um Cubpro v2.02 print; baseline 1
Next I put the same two models in CubePrint V4.03. I ran them as ABS with sidewalks. Knowing how these print with ABS on the Cube3 will get me the second baseline I need. Needless to say, these two models are still a world apart in similarity. The Cubepro-sliced print is smaller and the details are more 'correct'. The Cube3 print on Cubepro was 'fat' or 'thick'.
I changed the Cube3 file using Buddy's most excellent cubeutils. The temperatures were already current but 'M104' needed to be removed.
I used the editor and went straight for the BFB viewer. I copied the content and removed all and pasted that into notepad. I searched out all the M104's and removed them. Then I re-selected everything and pasted it back in the viewer's UI. Immediately the temperature fields updated to show only 2x M104 present at 0 degrees.
The second part of the modifications required me to understand the settings that Cubepro would use for retractions. I ran these same models in Cubepro v1.87 (different computer) and saw that the retractions were 250 and the two modifiers are 300 and 700. I changed the cube3 file to conform to these values using the editor.
Then I needed headers. I set this all up as a right extruder print as I wanted yellow trays form the Cubepro. This meant I needed a right side 'doner file'. I sliced any-ol'-file using the right extruder only without sidewalks or supports (forcing -1 to these entries) in the Cubepro v1.87 app. Considering these values have no meaning in conversion (slicer and app settings only) this keeps the header simple. I cropped everything between the 'init complete' and the last time-stamp replacing the void with the BFB from the cube3 file in between, cropped similarly.
Long story short, I opened the file in Cubepro v2.02 and it provided the preview expected. I then sent the file onto the printer to be run. The output was one solid puck of Ant Tray! If you want a fast way to get ABS prints to be "thick", this is certainly it!
I am taking another run at these prints this morning. I've decided that I would try backing off the extruder pressure to 75%. There is maybe 7 or 8 entries ranging from 40 down. I ran the script maker and pasted the values in Excel to do the calculations. Since all the numbers were whole numbers, I replaced the values with their rounded whole number value. I also decided to place a tape dot under the z-level switch to get a little more gap out of the first layer. ...the quick way to increase the Cubepro gap.
I am convinced that there is a proper pressure percentage that will get a Cube3 ABS print to look identical on Cubepro. I don't know that the correlation of the existing data is simple as both pressure and velocity work together and the Cubepro profiles are completely different from the Cube3 profiles. However, this percentage-correction should cover all basic for this type of conversion. When I get to the point of getting the final percentage number, I will need a different test model with 'bridges'. I'm just trying to get to a ballpark using this method and being productive at the same time.
Attachment 3318
Cube3 ABS w/ retraction correction applied.
I have long thought on using these pressure changes to make files compatible across platforms. There is no easy way to do this without the help of something like the cubeutils. Now the change is as simple as knowing the conversion factor you want to apply. Just know that some prints only have a few variations... others have many more.
So this is my first go at platform conversion with ABS. As I only run ABS in the Cubepro duo, a full library of conversion will not be published but for anyone trying this for themselves will understand the process and hopefully share their findings as well.
I did run into one bug that was not reproducible, Buddy. I was changing extrusion pressure values in the editor UI and when I got the second-to-last entry, it would not take it nor let me move to the another value. It was acting as if a buffer filled. I exited out of the program and re-opened the file and this time worked the pressures from the bottom up and all worked as expected. Just in case you thought you'd seen this before, this might suggest you actually had. No additional errors were observed in any other operation in this experiment.
Cubepro shortcoming...
Attachment 3320
left: Overextruded cube3 file - right, CubePro v2.02 @300um
- - - - - - - - - -
Seems 75% is a very good number.
Attachment 3321
Very comparable to the Cube3 printed ABS version. The z-level dot alteration helped. The part was still very stuck to the build plate.
Here is a comparison of all three trial pieces:
Attachment 3322
top right; Cubepro v2.02
left; straight conversion Cube3-ABS to Cubepro-ABS
Bottom right; left version with retractions reduced to 75%
- - - - - - - - - -
And if you're still wondering how to merge the walls? Try increasing extruder pressure to say 125% on the Cube3/Ekocycle platform. You might increase the heater block bias temperature by a percentage as well to make sure there is enough melt available.
- - - - - - - - - -
Definitely finding a bug in the extrusion pressure algo. Two means by which this attempt failed and one work-around.
The attached file has a script file and a native Cube3 ABS print file. When I open the file with the cube3editor and import the script (load script), it gives me an apparent 'null' error. ... "[]". However, it does take some of the data and wrongly updates the extrusion pressure data. Basically it replaces the wrong numbers when 2 entries fail; third and forth. Also updated are the retraction values as input with the script. What is unique in this file set is that I am not affecting the temperatures with the script file.
The other bug I noted yesterday is definitely there. You start filling in the extrusion pressures and at some point the 'curser gets stuck'. You can escape out of it but no further editing can resume on other fields. I again edited from the bottom up (lots of entries!) and it worked fine.
-
Wow, you have been busy. Sorry for my lack of response -- been dealing with trying to get Cura building. :) I am close, but not successful, yet. I will take a look at the extrusion pressure bug you found.
