ISON Tandem AirBike Builder's Log

November, 2002 -- I'm baaaaack


Wow... no updates from me for almost a year. Other projects and commitments kept me out of the shop for over a year, and now it feels really good to get back at it. As much as I hate how cold and short the days are this time of year, it does make it much easier to hunker down in the shop and build some airplane. My second wing panel is going together on the big table now... I won't bore you with pictures of that for a while.


One of my distractions this summer was joining in a partnership of five fellow aviators to form Adventure AirSports, LLC, mainly to provide aerotow launches for hang glider pilots in the Kansas City region. Our acquisition of this Bailey-Moyes Dragonfly has allowed me to become a tug pilot sooner than expected (before my AirBike tug is ready for service). The Dragonfly is a strange flying airplane, and I'm still getting better at flying it with each flight. I look forward to flying my Tandem AirBike, and expect it will make a nicer handling tug.


Back in February I was able to get the Junkers flaperons from Zenith 99% done. After the spar/rib assembly was riveted to the lower part of the skin, the upper part of the skin was pulled tight with strap clamps and then drilled, clecoed and riveted in place.


Having the assembly clamped to a flat surface is crucial for this step, because riveting the trailing edge completely locks the torsional alignment of the flaperon forever.


The four flaperon surfaces, ready to be joined into two long flaperons. That step will be done when they are mounted to the new hinge brackets on the wing. The flaperon halves weigh about 2 lbs. each. With the new hinge brackets, the total add-on weight will be around 9 lbs. (although they are lifting their own weight and more in flight).


End view of finished flaperon, resting on the right wing now going together.


Remember the extra gusset Richard suggested adding to the back side of the jury strut spar blocks? (See Nov. 00 update.) Don't glue them to the rear spar until you've installed ribs 4-7!!! Ooops.


I visited John V. in March, and admired the excellent progress in his metal assemblies. John reports that his Geo engine has been mated to the fuselage since these photos were taken.

The arrow points to a hook welded onto the outboard end of the upper foot support on the left rear rudder pedal. John added these to prevent his feet from slipping and rotating off the pedal, something that otherwise will happen naturally when your legs are relaxed. I think this is an excellent idea and I will be doing the same on my rudder pedals.


I towed my fuselage on its own gear to the hangar where I am employed as an aircraft fabricator/builder assistance technician. I've promised myself that even if it is ALL I do on the project this year, I will prime and paint the metal weldments before the year is over. I have better facilities for sandblasting and spray painting at work.

A 1/2" bolt fits snugly in the tailwheel spring and loosely in my ball mount hole. This gave plenty of motion and security for the short drive, and the "Tandem Trailer" rode just fine.


Compared to the ultra-complex Thunder Mustang project with its 16,000+ parts and ~7,000 hour build time, the Tandem AirBike is a walk in the park! Seeing the Tandem in the corner of my eye all day is really helping me get motivated to work on it again. I find myself yearning for the AirBike's simplicity after being bogged down in the Mustang's complexity and slow progress all day.


Nice spectrum of aircraft represented by these two projects -- one designed to fly very, very fast (~400mph), and the other intended to fly as slowly as possible (~27 mph)! Robert would like the Thunder Mustang, as it has a four-stroke engine (640hp 12 cylinder) and holds 51 gallons of fuel in each wing!


Heads-up to all AirBike pre-welded fuselage owners... I discovered that the lift strut brackets on my fuselage were not drilled as specified in the fabrication drawings. The edge-to-center dimension for these holes (shown in this photo as .295") is specified in the drawings as .375", which is intended to provide .250" (or one full hole diameter) of metal outside the hole. As you can see, my brackets have only .170" of metal outside the holes, which is only 68% of the hole diameter. These brackets are among the most highly loaded and critical fittings on the airplane, so this discrepancy concerned me greatly.

I contacted Wayne about this situation, and he concurred with my decision to remove the improperly drilled brackets and replace them with new ones which he will provide (he also offered to replace them for me if I sent the forward fuselage half back). Wayne is not aware of any other lift strut brackets being shipped out with improperly drilled holes, but I recommend you check yours carefully.

I made the decision to replace these brackets even though the removal and re-welding will be a pain in the butt, and even though the improperly drilled holes still seem to satisfy the basic safety factors designed into the Tandem AirBike. Jim Collie's stress analysis for the Tandem shows the front strut as carrying 2171.5 lbs and the rear strut 1850.2 lbs (I assume these figures are at gross weight and 4.4 Gs, but don't know for sure). The lift strut bracket is reportedly limited by (weakest in) its bearing strength (the ability of the edges of the holes to restrain the force of the strut bolt without plastic deformation), with a safety factor of 1.45 front and 1.7 rear. The bearing load surface in positive G flight is illustrated by the heavy line just above the arrow in this drawing:

Bearing load

I reverse-calculated the above figures and discovered that Mr. Collie had correctly calculated the bearing load based on a .090" thick bracket (I'm guessing he started with the .100" metal thickness and subtracted a .005" chamfer on each side of the hole). Here's the bearing load formula:

.250" (hole dia) x .090" (bracket thickness) = .022 in^2 (area)
.022 x 140,000 psi* = 3150 lbs hole bearing strength
3150/2171.5 (front strut load) = 1.45 safety factor

(*4130 Ultimate bearing strength = 140,000 psi)

This tells me that the ability of the material to resist the worst-case (4.4 Gs) pull of the strut bolt checks out (in bearing load), with a conservative safety factor. (Well, maybe. See * below.)

The next failure mode to analyze is the shear loading scenario. Here we are theorizing the force required for the strut attach bolt to shear the bracket material along two lines as indicated by the wavy lines:

Shear load

The closer the hole is to the edge of the part, the less shear load it will support. So although Mr. Collie determined that these bracket holes were stronger in shear load than in bearing load, it was important for me to calculate the shear strength to see if that was still true with my improperly located holes. Here is the formula for analyzing shear load:

As designed (.375" edge-to-center):
.375" (L) x .100" (bracket thickness) x 2 = .075 in^2 (area)
.075 x 55,000 psi** = 4125 lbs shear strength
4125/2171.5 = 1.9 safety factor (front strut)

As fabricated (.295" edge-to-center):
.295" x .100" x 2 = .059 in^2
.059 x 55,000 psi = 3245 lbs shear strength
3245/2171.5 = 1.5 safety factor (front strut)

(**4130 ultimate shear strength = 55,000 psi)

So, the conclusions I draw from this exercise are:

1) The bearing strength of the lift strut bolt holes was undiminished by the reduced dimension from the edge of the bracket (see * below), and
2) The shear strength was diminished from a safety factor of 1.9 to 1.5.

If I'm interpreting Mr. Collie's strut fittings diagram properly, the weakest link in the front lift strut chain is the bearing load of the aluminum strut tube where it is bolted to the the end fittings. He shows a bearing load safety factor of 1.38 at this point, so don't reduce your wall thickness here with deep countersinks on the bolt holes! In fact, I wouldn't chamfer them at all, unless very lightly.

Bottom line: In theory, my improperly-drilled brackets should still have an adequate safety factor*. I personally, however, am not willing to trust my life -- and especially the life of a passenger -- to brackets that appear so borderline. In the airplane industry, critical holes are generally not less than 1 1/2 diameters from the edge of the part (FAA guidelines specify no less than 2 diameters even for rivets). I know from my experience with the first airplane I built and flew that if I am not confident in the safety of every part, I won't be able to relax and enjoy flying the airplane. I don't want to worry about the lift strut brackets (or any other part) every time I hit a nasty bump in the air, so I'm replacing them with properly-drilled ones.

*This just in, after the above was originally published:

My aero engineer friend Scott Bledsoe of Proprietary Designs, Inc. reviewed my strut bracket calculations and sent the following reply:

Having looked at (your) pictures on the web, I think new tabs are definitely the way to go. If you do encounter any difficulty with the prospect of replacing the tabs, though, it does look like you may have room to weld (or mechanically fasten) a doubler tab onto the original tab and back drill the holes. Just a thought in case you need a plan B.

Your analysis is good, but you should note that the bearing strength of a material, as listed in a handbook like MIL-HDBK-5 *assumes* a certain edge distance. Usually this is 2 or greater. So in your calculation with a hole at 1.5 ED, the bearing you used may not be right. That's nit-picky, but the important point is that "bearing strength" is not a unique number like "tensile" or "shear" strength. There may be an additional reduction when the hole is in a short-ED corner, like you have. That's where things get murky and it's best to play it safe as you have done.

So, my conclusion that the bearing strength of the bracket was undiminished by the improperly drilled holes might be flawed. This adds even more justification for playing it safe and replacing the flawed lift strut brackets.


Engine news

Thanks to the June 2002 Kitplanes Engine Beat article, I'm now considering putting a Hirth F-30 on my Tandem. The comparison in the table below to the Rotax 582 should point out the pertinent reasons why!

Hirth F-30 (fan-cooled version shown)


  Rotax 582 Hirth F-30
RPM (for below) 6000 4500
Horsepower 65 85
(90 with F.I.)
Torque ( 55 78
Displacement (cc) 581 1042
No. of cylinders 2 4
Weight (lbs) 110 - 128* 125 (free air)
Fuel consumption (gph) 7.2 4.8**
TBO (hours) 250 1200

* Depending on who you ask. Probably 110 - 111 lbs with C box and electric starter, fully wet.
** 3.7 gph with F.I.? See letter below.


Holy moly! Those are some intriguing numbers. I sent a query to Recreational Power, the Northeast US and Canada Hirth dealer, asking them to comment on using the F-30 for my glider tug application. Here's the reply I received:

Hello Doug,

For your application I would recommend the F-30 85Hp free air engine. Total weight (125Lbs) will come in about 3 pounds less than a 582 with E-box, water and radiators (128Lbs total). Those long full power climbs with the F30 will be at 5500 Rpm with 20 more Hp than the screaming 582 turning 6600. If you add the fuel injection option it will put out 90Hp still at 5500 RPM and burn less fuel than the 582. I run an injected F-30 in my plane and average about 3.7 gal per hr. The difference in sound climbing out at 1100 RPM less than the 582 is considerable. With the extra thrust the climbs should be faster, and as you will not be working the F-30 as hard as the 582 I would think the F-30 would last a lot longer.

The rules that apply to Rotax air cooled engines do not apply to Hirth. Rotax uses steel sleeve cylinders, Hirth uses Nikisil cylinders which are not susceptible to cold seizure or shock cooling. The only way to seize a Hirth is if it is run too lean, cooling air is blocked or lack of oil. So a water cooled engine is not necessarily better for towing. I think an engine with substantially more torque is the better choice.

We have F-30's in stock, If I can answer anymore questions give me a call 800 583-3306 10 to 5 eastern time.

Matt Dandar - RPE


Too good to be true? Dunno. I'm well aware of the negative buzz about Hirth engines that was going around the internet a couple of years ago, but I've also been made aware that there are plenty of users out there who have only good things to say about them. And with all that power, I wonder if I would really need to use over 4,500 rpm, even for climbing out with a glider in tow. Even at 4,500 rpm the F-30 will produce 5 more HP than the 582 at 6,000 rpm, and will produce 41% more torque to boot! See graph below.

HIrth F-30 power/torque curves


When asked where I could see an F-30 in person, Matt replied:

Hello Doug,

Yes you can use (publish) my reply.

I don't know of an F-30 in the KC area, There are 2 in Iowa. 1 in southern Iowa on a gyro and another in Central Iowa on Golden Circle Airs factory trainer. Mike Solono the factory pilot for Air Command Gyros has around 500Hrs. on his trainer. You could contact him at Air Command to talk to him. A Gyro trainer is about the closest thing that would compare to the heavy duty cycles your plane will endure. As the F-30 does well on Gyro trainers, then your tow plane will not be a problem.

Matt - RPE


I'm not very close to being able to afford an engine at this time, but I'll keep you posted on this new development. Tandem Rocket, watch out!


If you can't afford to do it right,
be darn sure you can afford to do it wrong.