Birch V2

[Nylocke]

I got 3 rolls of NinjaFlex (Midnight, Fire, and Saphire) for Christmas. I also got a pack of 4 steppers. Looks like I'll be getting to 3+ extruders and some Ninjaflex this break.

My local maker space is doing something for New Years. I think we're going to hang out for a full day. I got some Tupperware to waterproof my water cooling. I'm going to do that tomorrow, with some rewording, so I can take it with me places.

Also, I never got those rubber feet with my printer. SeeMe doesn't sell them standalone. Looks like that can be my first Ninjaflex print

[Nylocke]

Also I have plans to swap back to Marlin.

[Eaglezsoar]

Are you switching to Marlin because of its supposed auto leveling?

[Nylocke]

Part of the reason. Also those rumors of SeeMe swapping back. I started out with Marlin and its always been kinda my favorite.

I've been looking around and it just seems like the auto leveling is a lot more mature on Marlin than it is on Repetier, and I don't want to wait for .92. The multiple points of probing seems like a much better solution in comparison to the 4 defined by repetier. I know I have to work hard to get the machine really well calibrated, but I think things will work real nice with marlin and the FSRs.

[Nylocke]

So i figured out my extruder DOESNT WORK AT ALL with Ninjaflex. I need to modify the design a bit to clamp more on the filament. I have it designed to the closest the bearing and drive gear are supposed to come together is 1mm. Guess thats too large of a gap. Guess that means I have to print a V4. Sorry to anyone who didn't have success with it.

[Nylocke]

Ive gone through 2 revisions since the initial test. R4 didn't work much better than R3 (first release). I have modified the extruder even further with R5, which is printing now. I won't get to testing NinjaFlex with it until a bit later however. R4 was a simple change, I had designed it (for some odd reason) so the closest the bearing and drive gear should ever come would be 2mm apart, so I cut away material so the gap was only 1mm. A new print resulted in this change also not working, though I got more of a "part" than with the original release version (it didn't even extrude any material originally).

R5 has moved it so there is a .3mm gap between the bearing and the effective diameter of the drive gear. I also changed the angle that the spring goes on once again, this way there should be significantly more pressure on the filament (this was another issue I identified while testing R4, even with PLA/ABS). I will upload the file changed and report back with my results ASAP (probably won't be until Friday at the earliest)

(edit: R5 is live on Youmagine and Repables)

[Nylocke]

I made another little change to the tensioning arm. I beefed it up even more and then I trimmed a bit of material off that was interfering with the bearing making contact with the drive gear while the extruder was unloaded. I'v also changed to a new version system that describes which version each piece is on. Right now it is on 5.5, the first number meaning which revision the body is on (5) and the second number being the revision of the extruder arm (also 5).

[Nylocke]

While creating a Marlin config for my printer, I realized what was probably causing most of my issues with printing flat. My carriage horizontal offset is different than everyone else's because of the birch being thinner than the melamine. For some reason in the .91 repetier there isn't an explicit representation of this value, but I'm guessing it gets calculated in some way.

Looking into things, Marlin is getting pretty well cleaned up. There are a lot of features and other goodies that are being added to Marlin that aren't getting added to Repetier (or working in Repetier). I think that looking into Marlin is a good idea for anyone (especially those who are thinking of adding a leveling probe system, filament diameter sensor, etc)

[BenTheRighteous]

If you're going to be going through a bunch of revisions on your EZFlexStruder, might I suggest imprinting the version onto the part itself? A small "r5.5" on the arm and body or something.

I went through just 4 revisions of my (unpublished) cell phone charging cradle. Even then I couldn't keep track of which print came from which model before not too long. Started regretting not even putting a simple versioning scheme on each print, even an increasing number of dots somewhere inconspicuous would've helped!

Anyway, hurry up and finalize your design already. I want to try printing it with trimmer line!

EDIT: Oh also, why are you trying to design a minimum gap between the bearing and the gear? If I were designing it, I would just let them touch (but I would probably be oblivious to any damage that might cause as well). Just curious.

[Nylocke]

I can't remember the exact reason why, I think it was to help with constraining the filament properly, but it was more of a "perfect world" design. I have since changed it, R5.5 has abolished this. I'm going to test it fully this weekend, the plan is to get all the kinks out of the design and get it working flawlessly.

Whats up with your trimmer line? Ive printed plenty of Nylon with the default EZstruder.

[BenTheRighteous]

I meant that your EZFlexStruder is going to be made out of trimmer line. Then I'm going to print some soft filament with it!

I printed one of your previous revisions out of ABS and it probably would've turned out ok if I'd read your instructions, but instead I cracked the print trying to get the pass through fitting in there. I just figure nylon will probably hold up to being drilled/threaded a bit better. We'll see!

[Nylocke]

I knew thats what you meant the second time I read it. Then I read it again and forgot lol. You probably will want to make the arm out of ABS or PLA or something more rigid than Nylon, it may flex after time.

I should probably adjust the diameter (or make that hole taper) a bit more. Ive gotten 2-3 made without cracking but I was messing around with one today made on a poorly aligned Rep2 (schools printer) and I cracked it. When I fix every little thing thats wrong with it I'll write a more detailed installation guide so y'all aren't in the dark on the less obvious procedures for installation/printing.

[BenTheRighteous]

Good thinking on making the arm from a rigid material, I hadn't considered long term flex. I will probably still print the body from nylon but perhaps the arm from something else.

If I may make a request, could you add a small ridge to the base of the bearing shaft? I noticed that when I had the retaining screw fully tightened, there was no space between the outer ring of the bearing and the inner wall of the arm, and the bearing didn't roll well.

I suppose I could just print a washer of the appropriate size though too, now that I think about it...

[Nylocke]

Did you over tighten the screw? I have like 5 of these printed, all different versions, and even the Makerbutt printed one allows the bearing to move smoothly. I can probably fillet that a bit more though, that should help.

[Nylocke]

Ive spent the past few hours trying to get a working Marlin config. Everything works except the endstops. For some reason the Y and Z endstops are set up wrong and I can't for the life of me figure out how to fix it. They are both for some reason set to Y min and Z min, even though I have it set to go to Z max and Y max on home. X is working fine. I can fix it by swiping the pin definitions, but if I do that the auto leveling doesn't work, since it uses Z min as the probe pin. I have a lot of confidence in the leveling routine working, it does several iterations of probing making minor adjustments each time, and if what it says in the config file it should get the machine within a .03mm total tolerance of level (+/- .015mm).

Once I figure out all this Marlin business (someone mentioned I need to disable X_MIN and Y_MIN somewhere, but I don't know which file they were referring to in particular) I'm going to move onto the extruder. I really want to make some feet for my printer, amongst other things.

[Nylocke]

And it turns out it was just me being dumb. I had my FSR board plugged into the X min, not Z min, so even when I did have the endstops configured correctly it wasn't going to work. Doh.

[Nylocke]

I got everything figured out. I ran G29 and it just probed the bed. It was pretty cool. I need to figure out all the commands and how they work but this is gonna be awesome.

[BenTheRighteous]

Did you over tighten the screw? I have like 5 of these printed, all different versions, and even the Makerbutt printed one allows the bearing to move smoothly. I can probably fillet that a bit more though, that should help.

Of course I did! I'm paranoid about moving parts and loose screws.

A fillet could work but I also wonder if that might introduce some wobble into the bearing's motion. I don't know what the proper engineering term is (I had to look up "fillet...") but my thought was similar to just sliding a small flat washer onto the base of the shaft before installing the bearing. So the edges would only be 90 degree angles.

Of course you should design it how you want, if you don't want to change it up in the design then I'll just print my own washer like I mentioned earlier for my own little mod. I'll be happy either way!

[Nylocke]

I made some changes. I compressed the spring an additional 2mm and added a bit of a ridge that should help with your binding issues.

Looking deeper into things, Marlin is a bit fragmented. The main Marlin (official release) does not support leveling with deltas atm. Johann's branch is more of a correctional sort of leveling, it doesn't actually adjust any parameters, it just gets it to print flat. Rich Cattell has a branch that actually adjusts the parameters over several iterations, but its a bit old and it doesn't have the X3 or X3 Pro on it. If one of you is feeling adventurous it has RAMBo support. It sounds like Rich's helps with calibrating the machine a lot, but there is still some manual calibration to be done if you're not lucky.

https://github.com/RichCattell/Marlin" onclick="window.open(this.href);return false; The development branch was updated 2 days ago, so it seems he is still tweaking, but no X3 support is annoying.

https://github.com/jcrocholl/Marlin" onclick="window.open(this.href);return false; Johann's branch.

Ive been having issues with my DELTA_SEGMENTS_PER_SECOND and the Acceleration values being to high. I was getting stuttering with 160 Segments. I dropped it to 100, but 9000 accel is too high and it ends up ripping up what it just laid down. I don't know why I kept it that high. I'm trying again with 2000 Accel right now.

[Nylocke]

V6.6 Seems pretty much perfect. I just printed one last night and mounted it on a Mini Kossel I have with me at the moment. I put some PLA into it, turned the gear, and it fed through nicely. I took some of my NinjaFlex down, squeezed the lever, fed it in, and then pulled on the filament to try to drag it out and it didn't budge. I turned the gear and it fed through consistently. I may make some slight tweaks but V6.6 is certainly ready for some serious testing/printing.

[Nylocke]

It does work!

First Ninjaflex print

Need to refine my settings but it made a thing.

[BenTheRighteous]

Nice! Is that a smiling creeper? What speed did you print it at?

[Nylocke]

Its a creeper keychain, one side is smiling one side is frowning. Speed was 20mm'sec. I wanna try some more with higher speeds, but I've been working on this Mini Kossel and Ive been printing parts for that all day.

[BenTheRighteous]

Nice, 20 mm/s is pretty good for a flexible filament. I could only get to between 8 and 12 mm/sec with the stock EZStruder before it would jam. Too bad you didn't try faster though! I bet you could've handled it!

[Nylocke]

I bumped it up to 30 with this coolio keychain that I've wanted for some time. Its got some embossed text cut in and extruded out, so it should be a good test of complex perimeters and retraction. So far its been running fine. haven't gotten to the text yet.