Descon Sprinkles

Scratch - Sprinkles {Scratch}

Contributed by Bob Ellis

Manufacturer: Scratch
Contributed by - Bob Ellis

Sprinkles, the Fire Hydrant
by Bob Ellis
While walking through Wal-Mart, trying to find out where they'd moved the rocket stuff, I saw Sprinkles, the Fire Hydrant sitting on the top shelf of the kids water toys area.  He is supposed to "provide hours of fun in the back yard with spinning action".  The instruction card did not include any directions, other than to use it on a soft level lawn for safety.  This sounded like it had to be intended for a rocket!
I went through several days of drawing fins and trying to get a good sim run, but all the usual ideas for fins changed Sprinkles from a Fire Hydrant to something strange.  Finally, while wandering through a hardware store my attention was caught by the Egg Crate Light Diffusers for fluorescent lights.  A bell went off, remembering discussions of grid fins, but never having tried them because of the work involved.  A couple of quick calculations and the grid looked like the way to go.  The calculations showed that for a 1/2 inch grid extending two squares (one inch) from the body I would get the equivalent of a 1/2 inch high fin with a seven inch span.  A quick sim run said that this brought the CP back to almost ? caliber stability, more than enough for a very stubby fire hydrant.  The best thing was that it wouldn't change Sprinkles looks; he could just sit on top of the grid.

Sprinkles is manufactured in China for Shelcore, a U.K. company.  They have a web site at http://www.shelcore.com.  The part number is 02650 Sprinkles.  The light grid doesn't even have a manufactures name on it, just an Ace Hardware price tag.

The only obvious modification to Sprinkles to get him ready for flight was to remove the hose fitting on the side, this left a hole which was filled with a red plastic whistle salvaged from a Foxtail (a soft baseball with 4 whistles and a long nylon tail.  It whistles when swung by the tail or thrown).

 

At last a true use for singing fish
Several design decisions now had to be made.  The six inch diameter main body was rigid enough, but was definitely not straight enough anywhere to allow internal braces for engine mounts or couplers for separation joints.  In addition, the six inch diameter would be difficult to pressurize with the small powder cup on a 24mm RMS.  Therefore, the decision was made to use the outer body as a shell and provide a center core for the motor mounts and parachute, no fin attachment was required since the grid fin would simply be glued to the bottom of the body.  For simplicity I decided to use a standard BT-60 tube with a 24mm motor tube.  In keeping with the Flounder Parts spirit, I figured that I should use some truly Flounder centering rings.  I cut them from the box for a "Big Mouth Billy Bass" singing fish that I received for Christmas.  (The fish may eventually fly, if I can figure out how to compensate for the CP changes when the head and tail start moving.)
Parts layout, the T-nuts and bolts are not visible
The non-visible modifications that had to be made were still fairly simple.

1. The Yellow cap was removed (one screw) to allow access to Sprinkles main body.
2. The motor mount was assembled, pretty much normally, except for trying to fit T-nuts into the cardboard centering rings.  I didnÇt want to collapse the corrugations and weaken the cardboard. After crushing and bending two sing rings, I wound up using CA to glue two rings together and soaked them with CA to stiffen them.  I drilled holes for the T-nuts and glued them into the holes, then slid the double ring onto the motor tube.
3. The shock cord was carefully measured at over eight times the body length and attached thru the upper centering rings.  The motor mount was glued into a 9.5 inch piece of BT-60.
4. Two holes were marked off and cut in the top and bottom of Sprinkles main body; these were sized to fit a piece of BT-60.  A fly cutter made quick work of the bottom hole, but the top had to be cut free hand and sanded out.  The disks cut from the top and bottom were saved for use with the nose cone coupler so that the Yellow cap could be replaced.
5. Two quarter inch holes were drilled just outside of the BT-60 holes to allow the launch rod to pass completely through the body shell.  These were placed to allow the rod to go through one of the existing holes in the Yellow cap.
6. The surface inside and outside around the large holes was roughened up with a Dremel sanding drum to create a better surface for the glue to stick to.  The entire bottom surface was also roughened to prepare a glue surface for the grid fin.  Do this outside!
7. The BT-60 was then pre-glued on each end and slid into the main shell.  This is when I found something that stinks even worse than dope or CA fumes.  That "something" is Devcon High Strength Plastic Welder.  It does work, but the smell is terrible, and the working time is less than five minutes.  Once it starts to set, it leaves strings everywhere (like hot cheese), then changes to something like three day old rubber cement, then finally sets hard.  You canÇt smooth it out while youÇre working, you canÇt sand it, but you can use an Exacto knife and cut away parts that you donÇt like.
8. The internal "nose cone" is a BT-60 coupler with the disks from the top and bottom cutouts glued into the top.  The top has a ? inch nipple that the Yellow cap screws onto.  This screw was replaced with an eyebolt from the inside for the shock cord mount, with a locknut on the top to hold the Yellow cap on.  Some minor cutting was required to get enough clearance around the nut to allow the Yellow cap to spin.
9. The parachute and Nomex protector were attached to the shock cord, and everything was fitted together, there was no room for wadding of any kind.
10. The grid was cut out in a 13 by 13 square, I had planned to make a circle, but the grid was too brittle without closed ends.  I first tried cutting the grid with a Dremel cutoff wheel, but that just melted the plastic.  A one-inch, fine tooth razor saw worked out just fine, it just needed a little sanding with a coarse emery board to get rid of some sharp edges.  I cut out the center 3 by 3 squares to clear the BT-60 where it came out of the bottom of the shell and the grid fin was ready to go.  Who said grid fins were a lot of work?
11. Sprinkles then stood on his head while the bottom of the shell was coated with Plastic Welder (outside this time) and the grid was set into the glue.

Major parts after assembly, before the grid fin is attached, 
the eyebolt can be seen in the nose cone/coupler
Sprinkles weighs in at 435 Grams after 
his swing test, no nose weight was needed.
Sprinkles last step before being certified for flight was a swing test out in the backyard.  He passed that test with flying colors, transitioning quickly into forward flight, even when started backwards or spinning and with a F39 RMS load in place.

Sprinkles was now ready to fly.  I really liked building Sprinkles, no spirals to fill, no launch lugs or fins to line up, no filling and sanding, no painting and best of all the decals were already on.

Sprinkles sitting on the pad, ready for his first flight
Sprinkles first flight on a D24-4, 18mm reload in the backyard was a "qualified" success.  He made a great takeoff, straight up for a simulated 120 feet, arced over and headed back down.  Unfortunately, the four-second delay was too long.  The parachute came out and un-rolled but was still not fully deployed when Sprinkles hit the ground.  He almost had enough time, in fact he hit fins first, indicating that the parachute was almost open.  In Sprinkles defense, the parachute was an old heavy one that I no longer use; it just barely fit into the BT-60, and had to be rolled tightly.  This was my fault, not Sprinkles.  The grid fin took some heavy damage, breaking off almost all the grid on two sides, but Sprinkles showed no other signs of damage.
Just  after ignition for Sprinkles first flight
Sprinkles coming down on a streaming parachute
The aftermath of using the wrong parachute.
Sprinkles went back to the work bench.  The grid fin proved very easy to repair, just cut out some pieces of grid, sand everything flush and use PVC cement to glue the new pieces in place.  Some extra thin strips were cut and used to beef up the break lines and the fin feels as good as new.
Sprinkles on the work bench, I found a use for Estes aluminum launch rods, they can be used to adjust tension on a rubber band.
Sprinkles next flight was on an E28-4, 24mm reload, at a C.R.A.S.H. club launch during our Ninth Annual Colorado Aerial Rocket Circus (CARCIS IX) NAR Regional contest.  He had a new parachute, thinner rip-stop nylon that took up much less space and did not have to be rolled tightly.   Sprinkles launched into a momentary lull in the high winds that had just postponed the contest.  He had a good liftoff with only slight weathercocking.  Ejection was well after apogee, but this time he had a good parachute deployment and a fast drift downwind.  He did have minor damage to the same parts of the grid that had been repaired from the first flight.  It looks like the PVC glue didnÇt bond to the grid; it just held it in place.  I'll have to try some other solvents to see if I can get a good repair.

Just after ignition for Sprinkles second flite, who says you can't see the Blue in a 24mm Blue Thunder reload.
Sprinkles under a good chute, the grid fin is 
intact at this point, but broke again on landing.
Sprinkles does seem to be quite happy in his new role as a rocket instead of his previous life as a lawn toy.  In fact, that grin never leaves his face.

For my part, I like the reactions, starting with  "Where's the fins?" and "That can't be stable!", then changing to  "That was a great flight!"  and "I don"t believe it flew that well in this wind" after the launch.  We still need a good day with no wind noise to see if the whistle works.  I have one other rocket with two of the whistles and they can be heard at apogee but require an underpowered flite.  I'm hoping that Sprinkles lower speed and altitude will make the whistles easier to hear.

comment Post a Comment