Building a full-auto Nerf Demolisher Part 5: I make a flywheel cage without the goal of a lift in FPS.

Oh you have a new flywheel cage do you?

It is no big secret that the flywheel cage is an extremely frequent topic in Nerfing. This one is canted, that one is CNC’d from a solid block of aluminum, and there are several now in open source flavors. My own foray into the caged wheel came about from a goof, an inability I have to deal with imprecision, and the fact that I screwed up the stock one while filing down the pressure fit splines and now I can’t get a motor to seat properly. I actually did intend to use an aftermarket cage for my build, going as far as to buy all the cages that Dr. Snikkas has available because canted. Well, I did that too fast because the one thing I didn’t notice is that the cage for the Demolisher requires shell cutting. “But Blasterdad you’ve already cut your Demolisher to all hell why not just cut it more?” I dunno, I think I’ve already addressed the fact that I just do stuff for no apparent reason. Back to our story.

Can we get these wheels off here please?

Even if I was gonna go with my own cage I was still gonna need the stock wheels to come out of the stock cage. But how to get them off without weakening the hub? Well what you need in this situation is something like a Jaw Puller, or a Bearing Puller, or a Gear Puller, or a Tie Rod Puller… you get the idea. This thing is known by many names and it is incredibly effective at getting stuff off a shaft. Time to make a tool!

After realizing that having a fairly sharp edge where I’d be sticking my finger wasn’t optimal I made some refinements and now we have a puller that won’t cut your finger when you use it. If you are 3D print-abled feel free to download the STL file at Thingiverse. Keep in mind that I had to narrow the arms to 2mm so I could slip it in the cage wells and under the wheels in the cage I drew up, so the puller arms have been weakened a bit, I’ve had a few that were one-and-dones but they still got the job done.

flywheel puller

Now with a rounded over handle!

Hello, my name is Blasterdad and I like to make drawings.

Yup, I didn’t draw this cage to increase crush, induce dart spin, or increase FPS. I drew it to get a set of 130s nicely mounted into the shell without cutting open the side of the Demolisher. What seems like a super wide shell is actually quite tight when it comes to the tolerances around the flywheel cage mount. The 130 cans specific to the MTB Honeybadger and MTB Rhino are also slightly longer that the stock 130s the Demolisher ships with creating even more real estate issues. Couple that with the imprecision of grinding down the splines in the stock cage to allow for an aftermarket 130 to slip in and what you wind up with is either mis-alinged motors or worse you go beyond the point of no return and remove too much material and you get zero pressure in the fit – which is bad because fitting aftermarket motors in the stock cage relies on pressure to fit.

Besides, you don’t go after FPS lift with a 3D printed cage.

For obvious reasons, 3D printers are not exactly what you should consider the gold standard for dimensional accuracy. The FDM process is inherently imprecise, it’s fast melting plastic that’s squirting out of a nozzle in the most basic sense. My aim was to produce something that could exploit the strengths of the MP Select Mini and work around the drawbacks of FDM 3D printing – this meant making my focus calibrating that little sucker and lots of trial and error in slicing to get as clean a print as possible even if it meant a painfully long print time.

All done, it only took 7+ hours!

To tube or not to tube.

It seems there have been some debates around the usefulness of the dart tube in the flywheel cage context. Of note are the OFP cages that have a considerably low-profile design. One of my aims is to help address the reliability of the Demolisher, which admittedly might have to do more with the feeding end of the cage. What can I say I’m a sucker for symmetry, and besides it just looks cooler with a tube. Enjoy many examples of why it just looks cooler below, along with a little reflecting on exactly how it arrived at my latest design iteration.

It’s a Flywheel Cage!

I started with a measurement template and after 5 revisions landed on my current design.

One of the last refinements I made was to the feeding geometry.

My fourth iteration incorporated a uniform countersink at the mouth of the cage with a duckbill at the top.

The duckbill now sits inline with the top of the tube.

There is a considerable amount of post-processing required to get the tube clean enough.

Straight shot through the wheels.

That’s nice, but stats or no love.

There we have it, a cage that isn’t meant to contribute to a lift in FPS, add spin or crush the dart. Instead what I have is a cage that gets 130s into the shell of the Demolisher nice and neat, provides a clean mounting point for 130 motors, and is optimized toward encouraging more reliability from an auto-feeding mechanism. It doesn’t mean however that I didn’t get it on the chronograph.

All the results below are based on the following build specs:
2s 20-30c Lipo at roughly 7.6-8.3 volts
Stock Demolisher Flywheels
MTB Honeybadgers in the flywheel cage (we talked about the fact that Rhinos were unavailable at the time of the build right?)
MTB Honeybadger in the pusher gearmotor

I cycled 9 18 round magazines loaded with a mixture of Elite dart knockoffs, Koosh gen 4 darts and genuine Nerf Elite darts through a Caldwell Ballistic Precision Chronograph. I also threw out 6 results that were wildly outside of the standard deviation range while 8 shots failed to register. Leaving the test with a total of 148 registered shots, that’s plenty to get a good average.

Feel free to dive into the numbers here.

The results above make it pretty clear why MTB Honeybadgers are meant for the pusher, the flywheel cage not so much – if memory serves me right MTB spec’d ’em that way. However for a 2s setup after about ~1500 darts run through the prototype prior to the chronograph testing, on a mixture of newish and older darts I’m not terribly mad at the results. For me the big metric is Darts per Second coming in with a range of 9-11 darts, and fairly reliably at that I might add – throughout the chronograph testing I had very few jams, maybe 4 total through 9 cycles of an 18 round mag. I’ll take 4 jams out of 162 darts and be happy. The fact that 32% of the darts came in at 90+ FPS is another thing I’m okay with, when combined with the 80+ FPS range you get 56% of the darts in the 80-90+ FPS range. These are numbers that I can live with.

Next time.

Building a full-auto Nerf Demolisher Part 6: It’s not the voltage that kills you, it’s the amps.

1 Comment

1 Comment

Leave a Reply

Your email address will not be published. Required fields are marked *