Firefly C-BG Build

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

TheAviator

Well-Known Member
Joined
Jan 18, 2009
Messages
935
Reaction score
121
This is my take on the now-classic Hummingbird boost glider. Originally for 13mm A or B impulse motors, I've upsized mine a little for 18mm B or C motors. It uses the tip panels from my flapped S8E glider (that I have yet to build...). Target weight is 30g on glide.

Like the Hummingbird, it features movable flaps that deploy at apogee. The right flap is cut with a lower chord at the tip than the left to give the effect of slight right aileron input. The left wing tip will be weighted for a left-hand turn. The idea is that should a spiral dive develop, the increased speed with allow the aileron input to overcome the tip weight and roll the glider back to stable turning flight.

I've deviated from the original Hummingbird a little for the flap actuation mechanism. Instead of the original pivot pod, I've opted for a more traditional hook style. In place of the pivot pod is a bellcrank that is held down by the hook of the pod. When the pod slides back and is jettisoned, the bellcrank is released, allowing the flaps to deploy.

I've attached pictures of my work so far. The bellcrank is colored red and the pushrod black to make it easier to see. More information to come at a later date.

Solid Model.jpg

Boost Config.jpg

Glide Config.jpg

DSC_1112_Reduced.jpg

DSC_1113_Reduced.jpg

DSC_1114_Reduced.jpg
 
Last edited:
More work done today. Got the wing installed on the fuselage and adjusted the flap mechanism. I shimmed the left side of the pushrod to give the left flap slightly more up throw than the right. This will induce a (hopefully slow) left roll on boost. The difference is about one degree. Both flaps stop against the fuselage giving the same downwards throw of about three degrees.

After checking for fit and function, I bonded the left side of the fuselage to the model, completely enclosing the flap mechanism. This makes the fuselage incredibly clean and should give a measurable increase in boost over an external mechanism. Additionally, the built up fuselage is incredibly stiff, both in bending and torsion.

I've attached pictures of the mechanism in place, both with and without the fuselage side in place. Also, since I forgot yesterday, a picture of the parts in included.

DSC_1111_Reduced.jpg

DSC_1118_Reduced.jpg

DSC_1119_Reduced.jpg

DSC_1121_Reduced.jpg

DSC_1122_Reduced.jpg
 
Finished up the hard parts of the glider last night. The tail feathers are cut from 3/32" 6lb./ft^3 balsa and hand sanded to an airfoil shape. The tapered side of the airfoil was glued to the boom to give incidence on both the horizontal stabilizer and the fin. The effect is a small amount of up elevator and left rudder input.

I did some soft hand tosses in the meager lawn of my apartment building. It seems to need about 2.5g of nose weight, but this gives a nice loping left hand turn. I'll need more space to get it properly trimmed, but this is a good start. I need to add the glider hook and nose block yet, so I will ensure that these come to 2.5g before attaching.

I can't seem to find my 18mm tubing, so the pop pod will be one I'm reusing from a NARAM-54 glider unless I can find it. First flights will be on an A8-3 and we'll see where we go from there.

DSC_1127_Reduced.jpg

DSC_1123_Reduced.jpg
 
Finished up the glider tonight. Found a piece of 3/16" square plastruct tubing for the glider hook and some scraps of BT-20 for the spooler pod. The hook is a pylon with a 1/8" square spruce piece that slots into the plastruct tube. The spooler pod separates into two parts, each with its own streamer. The design slows the streamer deployment, allowing the glider to fly free before entanglement becomes a possibility.

Final glide weight is 34g, with liftoff mass 59g with an A8-3 installed.

It will be flying this weekend at the Ray Harmon Memorial Regional in Cleveland. Unfortunately, I will not be there, but I have been promised pictures and video, so stay tuned!

DSC_1131_Reduced.jpg

DSC_1132_Reduced.jpg

DSC_1133_Reduced.jpg

DSC_1134_Reduced.jpg

DSC_1137_Reduced.jpg

DSC_1138_Reduced.jpg
 
Last weekend, I wasn't able to get pictures or video, but a test flight did occur, and I was informed that the glider was tail heavy. So this weekend, I got an opportunity to go to MMRR in Columbus where they were flying B-BG. I got the longitudinal trim worked out and boosted on a B4-2; this flight the glider entered a light spiral dive from which it did not recover, so the time was only about 40 seconds. The spiral, however, didn't get any worse, so I figured the asymmetric flaps were helping at least a little. Also, I lost the forward half of the pop-pod so I had to cobble another one on the field.

Anyways, so I trimmed out the glider with a little more right tip weight to try to flatten the glide. The second boost was as arrow straight as the first, and it started a spiral dive again. However, this time, more half way down, the glide flattened to a typical BG circle. From there it was off to the races. I could tell I had found a thermal, but the extent of that thermal was unknown to me at the time. I chased the glider for about 1/3 of a mile through a wheat field; all this time it was maintaining or slightly increasing altitude. At the end of the wheat field was a tree line, I stopped there to watch as my glider caught an elevator to that big runway in the sky. After 12 minutes, I lost sight of it at an estimated 3kft altitude and climbing. Line of sight was never broken, I just could no longer distinguish the little black speck in the sky.
 
1.) Enclosed hollow balsa fuselages are advantageous for models with moving mechanisms. At least, there didn't seem to be any performance detriment, and the fuselage was significantly stiffer than a simple spruce stick for a little more weight.

2.) The slight roll on boost didn't seem to stress the glider and it ensured any pitch trim problems were averaged out over the boost. After it left the rod, there was no discernible deviation from a straight boost. With 1* of aileron throw, I got about 2 rolls a second.

3.) The asymmetric flap system seemed to help prevent or correct the spiral dive. I need to make it slightly more effective, so the next model will have a smaller right flap.

4.)A longer nose would be effective in reducing the amount of nose weight required. Reducing the thickness of the rear fuse might help too.

5.) Thermals are fun.
 
One last farewell photo. Unfortunately, the videos didn't turn out, so this is it.

DSC_1240_Reduced.jpg
 
The asymmetric flap was NOT at all for spiral dive prevention!

The flap setup mimics the old "wash in on the inboard main wing panel on the inside of the turn" setup that was commonly used on free flight model planes ages ago.. The claim was that it would cause models to stay in thermals better. I finally figured out an aerodynamically plausible explanation of how it might work (asymmetric stall behavior in turbulence that causes a turn to tighten).

Anyway, ever since I started using it, the percentage of my models that got good thermals went way up... and it seems to have been true on all the Hummingbirds I have made.

2 rolls per second is about right. I prefer to think of it as a roll about every 75 ft of boost altitude. (thats a constant, rolls per second varies with airspeed). With the original airfoils and good construction there is no drag penalty on boost, and the structural loads from it are infinitesimal.

Looking at the photos, it looks like your model had a lot less dihedral than my Hummingbirds did. THAT is why you had spiral dive problems. Dihedral is a pretty minor effect on glide drag (the vertical component of the wing acts sort of like a winglet). But its a big effect on spiral dive stability, and large dihedral is also a big help on visibility so the timers can keep it in sight longer as its disappearing downwind in the thermal. Anyway, the large fuselage, dihedral and tail sizes on the original were all pretty carefully sorted out and are important for a bunch of reasons. If you need to change things, try a stock one first.

Bob
 
Last edited:
The asymmetric flap was NOT at all for spiral dive prevention!

The flap setup mimics the old "wash in on the inboard main wing panel on the inside of the turn" setup that was commonly used on free flight model planes ages ago.. The claim was that it would cause models to stay in thermals better. I finally figured out an aerodynamically plausible explanation of how it might work (asymmetric stall behavior in turbulence that causes a turn to tighten).

Anyway, ever since I started using it, the percentage of my models that got good thermals went way up... and it seems to have been true on all the Hummingbirds I have made.

2 rolls per second is about right. I prefer to think of it as a roll about every 75 ft of boost altitude. (thats a constant, rolls per second varies with airspeed). With the original airfoils and good construction there is no drag penalty on boost, and the structural loads from it are infinitesimal.

Looking at the photos, it looks like your model had a lot less dihedral than my Hummingbirds did. THAT is why you had spiral dive problems. Dihedral is a pretty minor effect on glide drag (the vertical component of the wing acts sort of like a winglet). But its a big effect on spiral dive stability, and large dihedral is also a big help on visibility so the timers can keep it in sight longer as its disappearing downwind in the thermal. Anyway, the large fuselage, dihedral and tail sizes on the original were all pretty carefully sorted out and are important for a bunch of reasons. If you need to change things, try a stock one first.

Bob

July/August 1993 High Power Rocketry said:
Since this seemed co be a major failure mode at the '90 and '91 flyoffs, special care was used to eliminate the problem. First, Hummingbird has a relatively high dihedral angle at 17*. It also has a low vertical TVC, small span and long tail moment. All of these tend to stabilize the spiral divergence mode. The use of flaps allows a large pitch stability margin, and there is a lot of incidence difference between the wing and stab during glide. This also helps avoid the spiral problem. Finally, the shaping of the flaps, with the larger flap on the left wing gives both tip washout on both wings and also mid panel washin on the left wing. A left turn is set up with very slight left rudder and left tip weight. If a spiral dive should start, as the speed increases, the built in "right aileron" setting will tend to overpower the tip weight and stop the turn.
Emphasis mine. Could you explain what you meant here then?

Also, I forgot to mention that the lower dihedral was unintentional. I failed the measure twice cut once rule, so ended up with 13* instead of 17* as intended. I understand that the spiral dive was started as a consequence of my goof.

This model was more a proof of concept for me, but I did enjoy building and flying it. The airfoils were wrong and poorly made, I futzed the dihedral, and the boom was repaired from a bit of hangar rash. The next model will have airfoils specifically chosen for the job and the flaps deployed for minimal drag at zero lift launch. So far, for this size, I've found that the Zone-V2 DLG tip foils work pretty well for the application with between 1* and 2* up deflection on boost and give full laminar flow at least to the hinge.

For reference:
Span: 380mm
Chords: 140mm->85mm linear taper, tips hand rounded
Area: 8.55 dm^2
Root airfoil: Zone-V2, 50-50 blend of 40k and 20k foils
Tip airfoil: Zone-V2-20k
Flaps: 1*/2* up on boost, 3* down on glide
HTVC: ~.5
VTVC: ~.035
Glide Weight: ~35g
Root Re*sqrt(CL): ~30k
Estimated boost Re: 700k
 
Last edited:
Back
Top