Page 1 of 1

Ninja flex issues

PostPosted: 2017-Feb-Tue-12-Feb
by cyborg5
I have a Printrbot Metal Plus that's about 20 months old. It has the upgraded aluminum extruder that is theoretically Ninja Flex capable. It has the old ceramic hot end (I apologize for not knowing part numbers). Several weeks ago I printed my first Ninja Flex parts. They were large flat pieces (a lid for a cup) and a printed very well. I was pleasantly surprised at how easy it was.

Fast-forward a few weeks and I tried to print another one and have had no success. The filament keeps jamming. It's not an issue of tension between the extruder and the spool. The extruder makes a loud clicking noise as it tries to feed it jams and the stepper motor jumps backwards.

I thought perhaps it was a nozzle clog but the filament does seem to drip freely while heating up. I also switched back to PLA I was able to print several parts normally so if I had a nozzle issue I would've thought my PLA prints would mess up.

I printed at 235° and 35 mm/s for both my successful prints and my later failed attempts.

Can ninja flex absorb moisture from the atmosphere? I'm wondering if maybe it was successful with the filament fresh out of the package and now several weeks later it has gotten stickier.

Any advice is appreciated.

Re: Ninja flex issues

PostPosted: 2017-Feb-Tue-14-Feb
by ktfergus
Upgrading to a 13s hotend might help, but even with one of those I could never get Ninjaflex to print as well as I needed.

My other recommendation is to print reeaaalllyy slowly; see if that helps.

Unless you need the part to be super flexible, I suggest using Ninja "semiflex." I had much better results using it. However, I've since switched to Sain Smart's TPU filament for all of my flexible parts. It's half the cost for almost twice as much filament & I can print it at ABS speeds.

Not sure about water absorption, but it couldn't hurt to "cook" the ninjaflex to dehydrate it. I suggest baking a small amount first, just see if it makes a difference.

Finally, I've attached a pic below showing the difference I encountered printing ninja semiflex after switching to a 13s hotend using the same settings. I highly recommend one, it's amazing how much better they print.


Re: Ninja flex issues

PostPosted: 2017-Feb-Tue-14-Feb
by cyborg5
Thanks for the response. I thought about switching to semiflex. You're also giving me incentive to think about upgrading the hot end.

Re: Ninja flex issues

PostPosted: 2017-Feb-Tue-15-Feb
by RetireeJay
I've printed a rather large part in Semiflex, but I'll admit it wasn't easy. I have about six partial prints under my workbench that came before I settled on exactly the right combination of nozzle size (0.8), trace width (0.8), layer height (0.4), printing temperature (235), printing speed (volumetric, 10mm^3/sec), extrusion multiplier (1.12), and part geometry. Even then, the final result was not quite perfect; it was good enough to use but required some doctoring to fully achieve my goals (e.g. leakproof). BTW, the Extrusion Multiplier is needed because the filament compresses inside the driving point of the extruder and a command to extrude 10mm of filament won't actually extrude that much filament without the multiplier. That's given, of course, that I'm using an accurately measured diameter for the filament size and that the extruder is correctly calibrated for normal filaments.

For smaller prints (that would fit inside a 100mm cube) I use a 0.4mm nozzle and a significantly lower volumetric extrusion rate.

I don't see any option for "volumetric rate" in Cura, but it is available in Slic3r. You can use Slic3r to create G-code even if you use Cura (instead of the manifestly superior Repetier) to control your machine and send the G-code.

Re: Ninja flex issues

PostPosted: 2017-Feb-Tue-16-Feb
by musk
Much trial and error has led me to perfect and reliable ninjaflex prints on my metal plus.

The hardware I use:
Ubis13 (not S), though this is probably similar to using the S
Aluextruder V2

Hardware notes:
You need to make sure that the lower part of the extruder bracket (the part that holds the hotend) is pushed up as tight as possible to the drive gear. When the bolts are loosened, you can see that there can be quite a bit of variance in how snugly that bottom piece fits the contour of the drive gear. You'll want to push it all the way up and tighten it, ensuring the tightest possible fit so that the ninjaflex has nowhere to slip out the side.

Slicing notes:
1. Don't bother using retraction. Ninjaflex will drool, there's no way of getting around it. Retraction will just make any problems worse. If you really want to us retraction, make it very very small. (0.3mm)
2. Because Ninjaflex will buckle in the extruder pinchwheel area, the main thing to consider is the MAXIMUM FLOW RATE that your extruder can sustain without gradually building up too much pressure and causing a problem. You will find that this limitation means you'll need to print extremely slowly, but the benefit is that this limitation effectively means you get higher resolution for "free", as you can print thinner layers without incurring extra print time. I believe my super reliable settings are 0.16mm layers at 20mm/second. If I were to print thicker layers, I'd need to print much more slowly to compensate, so there's a nice benefit that everything I print with ninjaflex looks really, really crisp.
3. Ensure that you do not overextrude your first few solid layers. If your Z offset calibration or extrusion multiplier cause you to extrude too much rubber for the available amount of space, it's like holding your hand in front of the nozzle to choke off the flow. This backpressure will cause problems feeding. You want the volume to be either perfect or ever so slightly underextruded.
4. I use 240c.

Ninjaflex notes:
When you have a clog, make sure that there were no oversize lumps in the feedstock. I've had three separate rolls of Ninjaflex which had large lumps throuought, and these lumps would be slightly too big for the top of the extruder, causing a clog. This can be really frustrating when you're doing tests to figure out what went wrong. Everytime you have a clog, check the filament first to see if it was the feedstock's fault.