Send via SMS

Prologue

The Tern-2VS was designed to be a 2 meter sailplane capable of thermal endurance, but with a design that favored penetrating flight at moderate speed (~20-30 mph). The plane was sized to withstand a 50 lb winch pull, and used the SD7037 airfoil. It was designed to fly as an RES model.

22 February, 2006

Design of Spars

In working on the Tern 2VS, I have come across a lot of information on the internet on how to size a spar for bending moment. Sadly, a lot of it seems to be generally wrong, and relies upon the underprediction of material properties to avoid disaster. I decided to evaluate how wrong the common method was, so I derived the correct equations. I put the writeup together here.

23 February, 2006

Day 1 with the Tern: Tail Feathers

I started the build of the Tern today, chosing to begin with the tail feathers. The assembly of the stabs and ruddervators went fine, but I suppose that I would feel pretty bad about myself if I couldn't assemble some balsa stick parts over a plan. This was the first time that I used my GP Magic Magnet system, and it worked really well. It isn't tremendously better than T-pins for this type of build, but I do look forward to using it on the fuse. Here is a picture of the stab in work.



You can see the magnetic hold downs that I was using around the frame.

The part of today's work that I was the happiest with was the fabrication of the stab mount block. I planned this as a trapezoidal piece of balsa that would act as both a mount for the stabs (they are offset from horizontal at 39 degrees), and as a receiver for the composite tail boom. I cut a piece of balsa to the right thickness (5/8), and then cut the bevels in the sides with my bandsaw. It turned out nearly perfectly on the first try.



I just couldn't resist taking the stabs and taping the rudders to them and then pinning both assemblies to the mount just to see how it would look. As I hoped, it came out very well. Not bad for a night's work.


24 February, 2006

Day 2 with the Tern: Fuselage Part 1

Today was my day off, and I was much more productive than normal - mostly because the weather was not nice enough to golf.

I began with a quick finish of the tail feathers. I drilled the mount plate and rounded the forward and aft edges, and then took some care to mount the stabs at the same height on the plate and the same decalage. After that was done, I moved on to the fuse.

Nothing makes me revise my thinking like actually building a fuse. I had planned to use a thin spruce plate for the towhook, and run it through a mousehole in two bulkheads. I decided that this wasn't the best plan, so I eliminated one of the bulkheads and made the towhook plate out of basswood. Much as is done on the Spirit, I used 6-32 blind nuts in three locations to give some Cg flexibility for the tow point.

The only other change that I made was to use basswood for B4, which is the main loadbearing bulkhead. It takes most of the load from the hookplate, and is the tie to the wing bolt. The aft deck went together nicely, and the Magic Magnet system made the height offset a breeze.



The aft wing bolt point is just between the full fuse bulkhead and the first bulkhead on the aft deck. While the double triangle stock at the aft deck attachment point has stiffened that up quite a bit, I am counting on the wing box to carry some of the tail load for big positive moment maneuvers. Given the stiffness of the tailboom, this shouldn't be a problem.

Once everything was test fit, the assembly of the fuse sides with the Magic Magnet system was the easiest that I have ever seen. I used the vertical stands to provide pressure at each bulkhead (replacing my fingers), and was able to get it perfect in two shots (one for each side). As you can see, I used some presses to hold the fuse down in the center so that it wouldn't twist when I had only one side on. I have to say, that worked even better than I had hoped, and this seems a truly great tool for making slab sided fuselages.



Overall, I call this a pretty good second day. The only problem was a lack of planning that caused me to glue some bulkheads in place before I had made cutouts for the pushrods. That may come back to bite me, as I couldn't get the pushrod holes cut at the angle that I wanted, but I don't think it will be anything more than cosmetic.

25 February, 2006

Day 3 with the Tern: Fuselage Part 2 - Control problems.

The day started a bit strangely. I was blown out of the water by Mrs. Tillman in match play this morning. She had me down 2 on the front nine, and I managed to squeak back to even by the turn with a par and a birdie on 8 and 9. I think that the hot dog did me in, because straight out of the gate on the back I was down 1, then 2, then 3 by the 12th hole! Every time I would nail a great shot, she would sink an impossible putt to push the hole. I just couldn't catch a break, and she closed me out 3 and 2 after 16.

Perhaps that was just the beginning of the control problems. When I got back to working on the Tern, I started tinkering with my pushrod arrangements. I had planned to put the servos in sideways and stacked in between bulkheads 3 and 4 (see yesterday's post for the cutaway). I quickly realized that this wasn't going to work.

With some quick (OK it was an hour and a half of tinkering) thinking, I decided to move one servo aft of the towhook block, and keep one forward. I could then run both pushrods out over the aft deck and bring them out with the tail boom.

After getting that all sorted out, I closed up the top of the fuse and got B1 ready for the nose block.



To help me keep track of all the stuff I am trying to get done this winter, I made a to-do list.

27 February, 2006

Days 4 & 5 with the Tern: Fuselage Part 3

Today I have come very close to finishing the fuse. It is covered, and I am in the process of attaching the control linkages. I forgot how hard it was to cover tail feathers once they are installed on the fuse, I won't make that mistake again.

Not many pictures, I don't want to give too much about the color scheme away before it is done, but here is a tease. These are the covered ruddervators.


01 March, 2006

Day 6 with the Tern: Fuselage finale and Wing Part 1

Yesterday I finished the fuse, including the controls. I was having a bit of a problem with control cable buckling, until I slapped a couple of nice black cable ties onto the pushrod at the nodes, and then the problem went away. I guess those are now primary structure.

I started the inner wing panel last night, got all of the ribs for that panel cut and the spar glued in place. Looks like my trailing edge concept is going to work very well, I'll try to take/post some pictures of that later. I have also decided to add wingtip extensions (3" span each and a tip chord of 4"). That will bring the total span of the model just under 78", and may drop the wing loading ever so slightly.

What it really does is make a dramatic change in the spiral and yaw stability based on the model that I am using. It should change the EDA to around 13 degrees for the whole plane, which more than compensates for the lack of dihedral at the root.

I also finished beautifying the excel spreadsheet that I use for spar sizing on my models. I have put it on my design page, along with some instructions.

Day 7 with the Tern: The Wing is Spoiled

I came close to finishing the center wing panel today. The LE and TE went on with no problems, and I was correct that my TE idea would work. My thinking was that a traditional 1.5 inch TE was to much for most of the wing, but just right for over the bolt block in the aft fuse. So I figured that I would use the 1.5 inch stock inside of BL3 and use smaller TE stock outboard. As you can see, it worked well, and just requires a little shaping to get them to match perfectly.



I also added the spoiler bays in the wings. I decided to go with a smaller spoiler that is only 6 inches wide, but is 1.5 inches deep. It will sit flush with the top of the wing when the cap strips are installed. The bays are framed, and the spoiler has a step to rest on when not deployed. I am going to use a set of dual belcranks to magnify the servo output from the S3108, and I may use springs to hold the servos shut. I haven't decided yet on all of that.



I ended up deciding that a #16 rubber band through an eyelet on the bottom of the spoiler that was then wrapped around the CFRP tube spar was the perfect solution. These provide a good amount of tension, but are pretty easy to install, even with a pair of hemostats through the servo bay.

04 March, 2006

Days 8 and 9 with the Tern: 3 Wingtips, 2 Spoiler Servos, and an outer wing panel.

What an interesting couple of days this has been. Two days ago, I started building the wing tips. One of them came off just fine, and then I broke the other one. For some reason, this sent me into a funk, and it didn't dawn on me until yesterday that I could just rebuild it (duh). So I did.



Problem number two arose with my selection of an S3108 for the spoilers. That servo just doesn't have enough action. It has plenty of torque, but just not enough throw with the standard arm, and I couldn't find a replacement arm. So I am switching to an HS-80MG for the spoilers. I'll drop the arm through the bottom of the wing into the fuse, and that will give me plenty of throw. Along the way, I drafted a simple XL spreadsheet to analyze the geometry and the loads required to actuate the servo. You can get it on my Design Page, there are some instructions there too.

I finished the left outer wing, and joined the wingtip to it. It came out very nicely. I'll get some pix up as soon as I can.

Day 10 with the Tern: The Doldrums

If I had to guess, I would say that I am certainly at the center of this build. I think that the light at the end of the tunnel and the light at the beginning are equidistant. I think that the struggle of having to make a lot of little tweaks to the plan is starting to get to me, and I need to step back a little bit to avoid burning out.

I tackled some of the things that I hate doing today, and still didn't manage to get any of them done because of other requirements on my time. I started the wing bolt blocks (forward done) which had to be finished before I sheeted the wing. I also started the spoiler servo mount, which also has to be done before I finish sheeting the wing. While both certainly went well, they just took far too much time. I did manage to set up the spoilers with the Hitec servo, and they work just fine.

For some reason that I don't understand, I really like the look of cap strips on ribs. I think that they add a lot of classic look to a plane. Of course, they also serve some great structural value, especially when you are working with 1/16 balsa. It is just a shame that you can't see them through the monokote. I make way too many boo-boos to even think about clear monokote...

I have gotten decidedly lazy about bothering to take any pictures, plus there is so little to take pictures of.

05 March, 2006

Day 11 with the Tern: Renewed Fire

Well, maybe not fire, but at least I could work on it again.

I finished the center wing for the most part today. I realized that the cutout in the center rib and the hatch were holding the spoiler servo well enough, and I actually didn't need to screw it to anything. I may add a couple more bracing blocks on the hatch, but that will be easy before covering. After sheeting and cap stripping the center wing, I finished sanding the leading edge of the left wing and rounded the left wingtip. They both came out really well. With the dihegral, the final span should be just over 76 inches.



As you can tell from the picture, the sweep on the outer wing panel is quite large (14.5%). My calculations show that I will have plenty of lift, and the sweep will add a lot of stability, but it will really help the Tern get up to high speed during penetration flight.

I did a quick weight check. My target weight for the model, balanced, was 32 oz, which would have given a wing loading of just a touch under 8 oz/sqft. Covered, with radio and battery, the fuse weighs right at 11 oz. The center wing with servo weighs 3.5 oz, and the left wing weighs 2 oz. With another 2 oz for the right wing and then 25% addition to the total wing weight for covering, spoilers and joining, that brings the estimated weight of the plane to 20.5 oz before balancing. If I can't balance this thing with 12 oz of lead, we are in trouble.

10 March, 2006

Days 12 and 13 with the Tern: What a Wing

On Wednesday, I finished the right wing. On Thursday, I bonded the whole thing together and glassed the joints. Because of the way that I did the joints (I used TE stock as a large bonded surface and to give me the proper dihedral), there wasn't much load tie between the stiff spar on the outer wing and the stiff spar on the inner wing, so I glassed around the LE sheeting using 4 plies of 7781 glass (one full wrap ply, one joint ply on the top and one on the bottom, and then a doubler ply that was twice as wide). I also used a glass wrap around the body of the rib itself. I wrapped the glass around the rib several times, then wrapped that in FEP, and then held it all in compression with many tight wraps of masking tape. It worked really well.




I used wood filler where the LE met the sheeting, and over all of the LE glass. Sanded that down this morning and it is ready for covering. The wing weight (without covering) is right at 10 oz, including the spoilers and servo (which weigh an ounce combined). That should put me balanced at around 25 oz. I am pretty happy with that. I don't think that it will be the 2M HLG floater that I was hoping for, but it will be a terror off the hi-start.



Perhaps I could market these to IKEA as some sort of Scandahoovian modern art? Call it the Ingvar or the Tarlap.

11 March, 2006

Day 14 with the Tern: Bad golf but good progress

I am really close now. I finished covering the wing, I picked a sunburst scheme on top that just took a lot of meticulous cutting and covering and way too much intensity. I made a huge mistake in forgetting to rig both the spoiler wires and the rubber bands that hold the spoilers closed before I covered. I had to cut into the bottom of the covering for the wires, and rigged up a couple of hooks to run the rubber band around the spar with. Just took a little more time.

I used some DuBro electric hinge tape for the spoilers. According to the inside of the roll, it was made by 3M, and I wonder what it was originally used for. It is a really tough, thick polymer film with a very good PSA on the back. Many good uses for that sort of thing, but it was still pretty cheap ($6 for 10 or 20 yards).

All that remains is to drill and tap the aft wing bolt block, add the Velcro to hold the hatch down, and then balance the whole deal out. According to Curtis Suter's great sizing spreadsheet, my balance point should be 3.2 inches from the LE at the root. There is a link to this spreadsheet at Colorado Gliders.

I'm not even going to talk about the golf, but the weather was wonderful today.

12 March, 2006

Day 15 with the Tern: Fin


And by that I mean fin, as in end in latin, not fin, as in vertical.

I can't really believe that it only took me 15 days of building. All I had left today was a final weight and balance (came out at 31 oz with 5 oz of ballast), some radio installation/placement concerns, and a final trim of the aft wing bolts.

I ended up keeping the switch velcroed to the side of one of the servo bays. It would have been too much of a space and weight waste to put in a large switch with a mount. If I decide to change that later, it won't be a big deal. I made the hatch easy enough to get off that it won't be a problem. Because I didn't initially plan on using my computer radio for this ship, I installed a GP Elevon mixer for the Slinger (at Tower). It works well, but won't be needed if I switch over to a radio with onboard mixing.

Good thing that there wasn't much else so that we had time to go out and play a horrendously long round of golf.

I'll finish writing the build instructions and then I'll post them and the plans on my design page.

13 March, 2006

Trim flights

The wind was blowing well today, so I took out the Tern for a couple of trim tosses on the soccer field. It took well to the air, and the Cg was obviously pretty well on the nose. No hard landings or other flight problems.

One design problem reared its head. The shaft that I used for the tailboom didn't have anywhere near enough torsional resistance. When I hit the rudder, the v-tail would try to roll the tail, and it would spin the shaft instead of the plane.

After a few tosses, I went to the hobby shop and picked up a super light aluminum helicopter shaft and installed it. I think that will do the trick just fine. Should be able to finish up with that tomorrow and be back for some trim flights later in the week. I was tipped off to a sight where they sell composite tailbooms (at pretty good rates too). It is at www.tailboom.com, and they have many booms designed for aircraft from Mark Drela.

The tailboom that I got was weight neutral with my existing tail. I ended up adding about 0.5 to 1 oz of glass to reinforce it at the stab and at the fuse. I think that I may be able to take an ounce of ballast out of the nose because I shifted the overall weight of the plane forward with a bigger boom but less of a stab mounting plate. We'll see.

22 March, 2006

Rx Glitching, Power Models, and more...

Last weekend I took the Tern and the Sophisticated Dude out to the park for a quick set of trim flights. The Dude lept right out of my hands, and was quickly about 200 feet downwind. With no penetrating ability, it was a lost cause. I quickly switched to the Tern.

The day before, I had installed a GWS Naro six channel Rx so that I could use my Futaba 7U Super for all the great mixing. I turned everything on and began a quick range check. I was less than 30 feet from the model when everything started glitching. I extended the antenna on the Tx, but that only delayed the glitching by about 10 feet. Fortunately, I had no desire to crash or lose the model after a hand launch, so I packed up and went home. Sure enough, it was doing the same thing in the basement.

I went over to my local shop, and they indicated that this was a problem with these Rx units, and nothing was wrong with the channel that I was on. They hooked me up with a nice, caseless JR unit and crystal that works just fine and is actually smaller.

Since then, I fired up my really ugly stik and ran some gas through the enormously overpowered engine. I am also nearly done with my EZ LR250 Formula 1 ARF.

Of course, the weather is still rotten.

24 March, 2006

Beautiful, yet not very robust.

Well, some depressing news today. The Tern suffered a tremendous accident. I put it on the standard up start for a few launches. It handled like a dream, and had unbelievable response to both pitch and yaw. More importantly, it could convert altitude into speed and keep that speed up for high speed glides. I could do 4 to 5 laps around the field with no problems with just 100 foot of launch. All at 25+ mph of course.

Well, like the idiot that I am, I got myself in trouble. I steepened a turn too close to the ground (I still can't figure out why), and got nailed with a big tip stall. That swept outboard wing loses all its lift at once, and I had no altitude to recover (plenty of speed though).

The resulting impact sheared the wing bolts off, buckled the D tube in three places, and just about shattered every piece in the forward fuse. The wing, interestingly enough, showed little adverse affect, but will require quite a bit of monocoting work to restore properly.

I am going to finish a few more projects first and then built a new fuse for the Tern Mk 2. At least it flew better than I hoped it would.

At least all I lost was time, monocote and balsa.


Epilogue

While I enjoyed flying the Tern immensely - it was very quick to build speed off launch and was a wonderfully low drag ship - it was unsuited to my locale. Since it is hard to find a location with a large enough field for a full launch, it never came into its own as a goal-return ship, and has largely been replaced by electric assist sailplanes as my first choice. That being said, it was a wonderful design experience, and I learned a lot about scratch building over the two week build.

Back to Dr. T's Sailplanes

Copyrights


This site is Copyright 2006 by Matthew Tillman, a Creative Commons License has been established for this work.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License.

This page is powered by Blogger. Isn't yours?

In fact, I just got off the treadmill...