The concept for the Cessna 310 was established before I joined the company, and there had been a wind tunnel test completed as well. Having the latter was a little odd as it was the only model of the Cessna airplanes of the period of the early to mid 1950s that was tunnel tested, till those for the four engine 620 and, of course, the T-37 jet trainer.
The 310 was also the only one of these 1950s models with two prototypes, and we reverted to just one prototype for the 620, but the Air Force required three of them for the more extensive testing of the higher speed T-37. But the defining experience herein was with the first 310 prototype, which was intended for flight testing leading to certification and was accordingly Spartan–only two very functional seats (they had to accommodate parachutes), with no upholstering and a bare metal floor and exposed main spar in the cabin, and only a couple of stripes for a paint job.
The 310 was to be the company’s first twin engine, retractable-gear design since the UC-78/T-50 Bamboo Bomber of WWII, and the first Cessna ever with a tricycle gear. It was expected to be, and was, the first all metal, high performance twin using the then new fuel-injected horizontally-opposed engines and having all of the fuel contained in wing tip tanks.
Less visible was that it was to be an extension of the company’s single-engine line in that it would fit in a typical T-hangar of the day and be operated out of grass fields. Nothing I will discuss was directly affected by the grass field requirement, but the T-hangar sizing dictated the fuselage length, the wing span, including the tip tanks, and perhaps to a lesser extent the span of the horizontal tail.
That limited the possible solutions to our greatest problem in certification–inadequate longitudinal stability. And that problem was caused by the tip tanks and the flat nacelles which were made possible by the elimination of the carburetor.
I call the following discussion defining, rather than designing, the 310 because only one major change was made before certification, and some of us were convinced that change was necessary and had to come (but I didn’t know it would be on such an expedited basis when it did).
We did however then spend some time evaluating especially one design feature: the wing tip tanks, to understand their contribution to the airplane’s characteristics, and decide whether the tanks should be removed from the configuration for overall improvement of the plane’s performance.
But my first exposure to the 310 was well before its first flight and was to assess alternative ways of determining its drag coefficient to assure ourselves it would be high performance in cruise. The analysis confirmed to my satisfaction that it was going to be fast, showing a cruise speed of over 200 mph, which seemed great to me for the early 1950s. So speed performance wasn’t a big concern any more, though you surely didn’t want to compromise it in any way.
Then I inherited the cleanup of the analysis of the wind tunnel tests. One of the other selections for the configuration was a split flap, to be sure of drag when you needed it on this very clean configuration, such as during power-on approaches. This choice was a departure from flaps selected for other Cessna models of the era.
The tunnel test indicated not only that the split flap gave good incremental lift, but produced so much drag it was almost dangerous. Thus we limited the deflection of the flap from the 75 degrees tested in the tunnel to 60 degrees on the prototype and, based on flight confirmation, limited it further to 45 degrees on the initial production airplanes.
The other thing the tunnel showed was that the configuration didn’t have enough static longitudinal stability, which confirmed an analysis done earlier by some other aerodynamicist. Nevertheless the prototype configuration first flown was entirely that of the less than stable tunnel model.
Now I’m going to disclose something I’ve never mentioned before, even to my Flight Test, Aerodynamics and Preliminary Design colleagues.
In sorting through the accumulated engineering information on the plane, I came across a planform drawing superimposing the 310 on a Beech Bonanza, with the 310 horizontal appearing to be the Bonanza V-tail flattened out to be the 310’s horizontal tail (the vertical tail was of course not depicted in this planform display).
I thought to myself that if you were going to emulate a design for speed, that Bonanza was the ideal choice, but not for sizing a horizontal tail on a twin engine model. The drawing did not have a title block, was not signed or dated, and I don’t know who created it or what part it had in defining the early 310. You can tell I’m suspicious of it, though.
The prototype was first flown, solo, in January 1953 by our flight test chief. The second flight was solo, too, but I got to be the first passenger on a 310 on its third flight. I then again got selected to be the project aerodynamicist, flight test engineer, and flight test observer–specifically for certification–and was teamed with the pilot who flew the same mission with me on the 180.
He was a WWII fighter pilot who had to get a multi-engine rating for this new duty. After one of his first familiarization flights in our airplane, and before we had started the flight test program, he came storming into the group area and said to me, “You’re trying to kill me” and related that in trying to bank the airplane the aileron response was so slow that he increased the input and then couldn’t stop it from rolling too far and got into what he considered a dangerous attitude.
I thought, but didn’t say, “Why do you blame me for a basic characteristic of the design of the airplane you’re flying? And besides, if I really wanted to kill you I would have done a more thorough job.”
He had encountered one of the difficulties with large tip tanks, when full of fuel, that caused such resisting inertia to rolling and banking the airplane that you had to make a disproportionate input to get it started and then anticipate and reverse it well before you got to the desired bank angle. He didn’t have that problem on fighter planes, or the 180. This was a characteristic that was easily resolved with instruction and familiarization, and as he soon admitted it really wasn’t all that bad.
So, we started the certification process. I’m going to skip over some of the mundane problems (like gear retraction, prop synchronization and vapor lock), and get to the serious ones in my main area of interest. The flight tests showed we couldn’t pass the longitudinal stability requirement. Once this was understood I was given just three days, including doing some additional flights if I wanted, to determine a new area and configuration for the horizontal tail – or we would miss the planned date (seemingly pretty far in the future, so what was the hurry?) for the ceremonial introduction of the production 310 airplane.
I did have some additional flights done to compare to prior analysis and wind tunnel results, and didn’t want to be so safe in adding area that we would give up much speed. I met the deadline, we flew the new tail, and the longitudinal stability requirement was passed, and still with a cruise speed for the airplane above 200 mph.
Note that my assignment was not to solve the longitudinal stability problem, but to size the horizontal tail to overcome it. That was because the problem was caused by the large aerodynamic tip tanks that acted as a lifting body ahead of the wing, and the clean nacelles associated with the fuel injected engines.
The nacelles were elongated in front of the wing and moved the (destabilizing) propellers forward compared to where they might have been on a more conventional nacelle. Obviously those trim, streamline nacelles and hefty but aerodynamic tip tanks were sacrosanct to the original concept of the airplane and couldn’t be changed.
But with the tip tanks came inherent characteristics, some excellent, some good and some bad (but none too evil). Primarily, if you can get all, or nearly all, of the fuel in them and place it far from the passengers they provide a seemingly undeniable safety advantage.
The other contributions depend on their aerodynamic properties and their concentration of mass. Aerodynamically they are a contributor to drag, but those on the 310 were also oval in cross-section so as to create some endplate effect, which might improve climb rate as well as aileron efficiency. And as described above, tip tanks can detract from longitudinal stability. Their mass and related inertia also affect apparent roll performance, again as described above, but adequate lateral control procedures for them can be learned and used (aided by the improvement they can provide in aileron effectiveness). Less apparent is the contribution of their mass to the roll-yaw oscillation known familiarly as Dutch roll. Also to be noted is that their mass contribution lessens as fuel is consumed.
I thought these offsetting features of tip tanks, some good and some bad, ought to be quantified so we would better understand what we might be giving up for their safety advantage. I did this analytically, but convinced the decision makers that we ought to confirm them with brief and concise flight tests (they took four days). This was after we completed the company certification preparation flights, but before the FAA/CAA had done their official certification confirmation, so I planned the tests and wrote the related engineering report, but was not the observer on the flights.
By my plan we flew the prototype with the tip tanks on, full and empty, along with a tanks-off configuration with substitute conventional wing tips. These last two conditions were flown by putting a temporary fuel drum/tank behind the crew inside the cabin and placing it on the airplane centerline there.
Let me dispose of the characteristic of tip tanks I couldn’t resolve in my own mind, that of the annoying Dutch roll oscillation. It is induced by gusts in real (non- experimental/test) flights, but in tests is started by decisively yawing the airplane with no bank, returning the controls to neutral and holding them there, and reviewing the decreasing oscillations in roll and yaw angles as the airplane recovers to undisturbed (near straight and level) flight.
High inertia along the roll and yaw axes, as full tip tanks provide, by theory would result in a longer time and larger number of oscillations to smooth out, and that was certainly proven true on the 310. For the academic in me that was satisfying. But I must not have done the analysis with tanks off or empty, as I was completely surprised in these conditions by how fast the yawed airplane bounded toward and past neutral, and then an equally quick reversal, lowering the time and number of oscillations to recover, but subjecting passengers to high lateral accelerations in doing so. Maybe not so satisfying for the people on board in real flights.
This aspect of the tip tank evaluation was inconclusive to me as I had no information on whether longer exposure to lower accelerations was better than shorter exposure to higher accelerations. Interestingly, the pilot and observer of these test flights said they found the (high acceleration) motion with the tip tanks off less objectionable. But since we sought smooth air to get good data in flight testing, thus never experiencing Dutch roll in practice, I had no personal opinion about the possible difference.
Of course the standard 310 with tanks would have been subject to both end conditions tested anyway as (full) fuel was consumed during a real flight. So I concluded that Dutch roll characteristics should not be a determining factor in the tip tank controversy. I feel somewhat justified in that I have since never heard of objections to Dutch roll types of motion from users of the 310. (That doesn’t mean that there haven’t been any.)
We did learn something about the 310’s native lateral control capability in tests by which nominal roll rate was determined by rolling ninety degrees from one bank angle to its opposite. The roll rate with tanks empty or off was about 50% faster than with full tanks, indicating the steady maximum rate was never achieved in a 90 degree roll with tanks full. But that great performance with tanks empty showed how good the ailerons really were and how they could provide the necessary control of roll and bank under conditions with tanks with plenty of fuel in them.
Other criteria were more clear cut. There was a very slight improvement in climb rate with tip tanks on (less than one tenth of 1%). In opposition level high speed increased by seven miles an hour with the tip tanks off. Also with the tip tanks off the longitudinal stability improved by a factor of four, by one measure, meaning that without the tanks the horizontal tail area could have been substantially reduced, possibly increasing the level high speed by another 3 or 4 mph, or for a total of over 10 mph (that is 5%, and likely more, of the nominal cruise speed, felt possible since we had added one-third more horizontal tail area to pass the stability test earlier in the flight program).
The airplane’s range would have gone up by that same percentage as well. So basically we gave up some speed and range, and a little on lateral control, too, for the safety afforded with the tip tanks.
Incidentally, for reasons never specifically investigated, the second prototype was a little faster than the first, and the production models were a little faster than the second prototype–so the consumer’s 310 (available at the originally scheduled roll-out!) really was, and is, a high-performance airplane.
So, with all this activity we had first provided the necessary longitudinal stability, then quantified the somewhat offsetting impact of the tip tanks on 310 performance and flight characteristics. With the safety aspects still dominating, they were, as everyone knows, retained. Maybe that decision was preordained, but I felt we needed to confirm the magnitudes of the trade-offs they presented for our own knowledge and satisfaction.
But then came the report of one of the first serious accidents with a 310, in which the airplane was losing altitude and clipped the top of a mountain it was being flown over. The plane skittered down the other side of the mountain, with trees shearing off the wings, the tip tanks rupturing and the fuselage coming to rest further down the slope. Some fuel from the ruptured tanks flowed down the mountain and some of that went into the fuselage. But nothing ignited (with the fuel always distant from any heat or sparks created) and there were no fatalities. It was concluded that the safety aspects of tip tanks, even in this quirky situation, really were worth the compromises they required.