Flexy Dually - almost there (print quality)

Things are starting to look pretty good. I’ve started to make a few custom models. I’m presently surprised how straight forward and relatively simple it’s been to design models for the dual materials (I was worried that would be the hardest part).

I’ve made many adjustments to temp, flow, print speeds, etc. My latest print happens to be one of the fastest:

235 ic3d abs
225 ninja flex
85 bed w/ glue stick
Ooze guard
20 mms print speed
30 infill speed
10 first layer
98% flow
Adjusted fillament diameters based on average
2.82 ninja
2.91 abs
.18 layer height

The results are pretty good. Ninja tends to ooze but the guard helps.

I would like to figure out a way to keep the abs particles out of the ninja flex though (and vise versa, but the former is bothering me more).

I might tweak temps and maybe increase retraction? (Already set to 35 though).

Any other tips?

Looks good!

Maybe enable a wipe and coast in addition to retraction… not sure if Cura has those settings (different from the wipe & prime tower). The other thing may be to reduce the ABS extrusion temp 1 - 5C, that may help the ooze during the printing of T1.

Keep tweaking, I’m taking notes. :slight_smile:

Some tweaks. Slight improvement.

First - tried lowering ninja flex temp. Anything under 225 results in a failed print. 225 seems to be a good temp. The ninja does ooze quit a bit though. Anything colder gets jammed up a few layers into the print. Still 225 results I’m happy with.

Next - lowered ic3d abs temp to 232. Improved slightly the thing is that the ooze guard results in small particles being dropped. These are then picked up by the nozel and moved onto model during other extruder movements. Most of the particles in the model come from this type of thing.

Finally - I increased retraction to 40. Not a very noticeable difference. Probably not worth the change.

I haven’t tried the wipe and prime tower. That’s my next tweak. Cura doesn’t have a wipe and coast option. Should I look to slic3r? I hear people say good things about it, but am comfortable with Cura. Does slic3r support the flexy dually?

The interesting things about all the prints. The imperfections are all in the same locations (except for the occasional fillament imperfection). The sloppy ninja side is near the ninja seem. The abs particles on the opposite side are about the T0 nozel distance away from that same seem.

What I’m not showing is the ninja abs shadow ooze. Basically the pattern of the abs center outside the model formed by the inactive extruder. Ninja likes to ooze. The ooze guard helps at keeping these away from the print. I cannot seem to address. Any cooler and ninja runs into extrude problems. Additional Retraction didn’t help. This isn’t an issue per say, but does need to be taken into consideration when modelling. I also think the quality would improve if the extruder changed more frequently. This model is basically to parts. (Tire and rim).

Pics show latest revision and all three revisions (excluding failed prints).

I should also mention I’ve done some additional bed leveling (all hooked up now with a dial indicator) and Z stop adjustments (and tool head nozel leveling).

Ive seen some Ninja extruder issues when zstop is too close to bed.

I’m general, still feeling my way around the perfect Z stop height. Switched from folded paper (0.25 width) to a business card (0.30 width). Started getting it so I can slide the card in and out with slight resistance and no tool head movement. Still think it might be too close.

(If it’s not obvious, I’m in love and having a blast with all this :wink:

You’re doing everything right. The ninjaflex ooze is hard to control. The larger nozzle diameter may be contributing to the ooze… but without experimenting its only a speculation. The ABS ooze is easier to control through retraction and temperatures.

Using ABS and calibration cubes is probably the best way to dial in the initial nozzle height. Always use T0 for initial calibration. With the dual extruder, I’ll start with the business card to get a starting point then start printing a small calibration cube (20x20). When the height gets to 2mm, I’ll cancel the print and remove the wafer to look at the bottom. I’ll tweak the Z-endstop by quarter turns until the bottom looks smooth, with little striations from the ABS extrusion. Then start the same process for T1… generally with I know that if I can see 2-3 threads on the adjustment screw, I’m close to getting T1 level with T0… again its all with the calibration cube / wafer. I printed a M5 thumbscrew for the leveling screw to help with quick adjustments.

I personally don’t use the wipe/prime tower with the dual extruder. But it could help with the ninjaflex if you decide to experiment with more extreme retraction settings… for instance 3mm retract. But otherwise the ooze shield around the print should “prime” the nozzle for consistent extrusion.

Overall I think your prints look good.

Curious, Why not just print a 20x20x2 cube instead of stopping it mid print? Does that change the bottom layer results?

Do you expect to see the same extruder paths in the ninja bottom layer? My ninja bottom layer is pretty flat and smooth. The abs is smooth but I can still see the extruder paths.

If bottom layer is set to 100% instead of 125% should I expect 0 elephant foot behavior? It’s definitely reduced but not gone. I have the filament flow set to 98%. I get more accurate prints that way. Does the bottom layer extruder width of 100% take that into account?

I’m going to print some thumbs screws for the bed and nozzle. That seems like a nice convenience.

Thanks again for all your help and tips!

20x20x2 is approximately same as stopping a 20x20x10 mid-print? One less model to keep track of… :slight_smile: Printing 1.5 -2mm is enough to see how the bottom is printing.

Elephant foot is really the initial nozzle height too close. I use the flow rate / multiplier to gauge dimensional accuracy of the printed part. Since there are M5 socket cap screws on the printer, I print a thumbscrew cap that snaps onto the screw… the benefit of designing one is the known tolerance of +.02mm. The thumbcap also fits on the leveling screw for T1. So yes… multiple uses.

For the initial layer, I use 90% of the layer height and 100% width. Again, if you’re getting elephant foot its the initial nozzle height, try using the software Z-offset to tweak rather than the z-endstop screw (if the Z-offset gets to .3mm, go back and check the endstop screw).

Tweaking and adjusting is part of the experience. Glad you’re patient with the process.