http://ozreport.com/3.136
Weaklinks
Mark "Forger" Stucky - 1999/12/21
Many months ago I wrote to you with the idea of trying to do some hang glider aerotow testing, the intent of which was to define the actual loads encountered under differing conditions of tugs (low and high power), gliders (beginner, intermediate, and advanced), and pilot weights (single and tandem). Due to the magic of your Straub Report, instant interest was gathered and Malcolm at Wallaby Ranch was quick to call, leaving a message that he would be glad to sponsor the testing.
The brains behind the effort was Jim Murray, a NASA engineer who specializes in flight dynamics and is a true-life "Maguiver" with a reputation of being able to instrument a gnat's knee. Early in the Eclipse (aerotowed F-106) program, in which I was the test pilot, the computer simulation revealed the existence of an oscillatory tension mode in the towrope. The computer predicted something like a 12,000-pound steady-state tension value but overlaid on top of it was a continuous cycling value of several thousand pounds. In some cases this "bungee" mode would grow unstable and eventually exceeding the 24,000-pound weaklink. The level of bungee present was dependent upon the two aircraft, the stability characteristics of the tethered pair, the towrope attachment points, and the towrope itself.
Like the Spectra line used in many hang glider tow operations, the exceptionally strong Vectran towrope we were planning on using had low stretch characteristics. This meant low shock absorption and increased chances of encountering the bungee mode. At the other extreme a nylon towrope would have been too springy and it too could result in dramatic (traumatic?) bungee oscillations. The computer predicted a certain level of stretch would give the best tow characteristics. For our initial flights we planned on adding a 50 foot section of nylon strapping in the middle of the 1000 foot length of the 3/4" diameter Vectran rope.
There was some skepticism about the mere existence of this bungee mode. The Germans had towed unconventional aircraft during the war years -- large troop-carrying transport aircraft, even multiple aircraft were towed. They also towed the swept wing Me-103 Komet, the first rocket-powered fighter. Pilots hated towing the Komet and a USAF test pilot who got the lucky straw to tow a captured Komet described the tow as the scariest experience of his life. Even NASA's predecessor, NACA had towed a propeller-less P-51 Mustang in an aborted attempt to compare it's real world L/D to what had been obtained through wind tunnel testing. The steel tow cable broke wrapping around the aircraft, interfering with control, and resulting in a crash.
In all these tests there was never any mention of any bungee mode - did it really exist or was it some computer artifact? The answer was to run the simulation using conventional glider and tow aircraft numbers. The simulation indicated the bungee mode existed in normal everyday towing of sailplanes. Some of the old-time sailplane pilots expressed doubt over the simulation because over their years of towing experience they hadn't noticed any bungee mode. One said, "I've never felt no stinking bungee" (or words to that effect).
So Murray made up a couple of battery powered instrumentation packages, each about the size of a lunch box. We put one in a rented Pawnee tug plane and one in a rented Grob sailplane. The one at the front of the towrope read tow tension (using a solid state metal link at the attach point). The package in the Grob read longitudinal acceleration.
We launched in early morning conditions and the tug looked for level flight in smooth air. We flew at a couple different speeds and tow positions but most of the data was gathered at 55 mph, which was published L/D max for the Grob.
The data showed the bungee mode was very evident and I swear I could feel it. It was always present to a minor extent but was easily excited by turbulence or maneuvering, in which case it took several cycles and perhaps twenty to thirty seconds to reduce it back down to it's normal small oscillations. Probably the greatest excitation of the bungee occurred during the takeoff roll, most likely due to bumps in the dirt runway.
So what about the issue of the bungee mode and its effect on the Eclipse program? We found that as predicted, there was a stable region on low tow where the bungee was minimized and where the F-106 was extremely easy to fly on tow. Outside of that stable region the bungee became more of a factor and the F-106 became more and more of a handful to fly. In fact, in a conventional high tow position it was quite unstable and if I wasn't extremely careful the weak link would fail within several seconds.
Without doing any dedicated tests with hang gliders I can only guess but I think it is reasonable to expect the bungee mode is present in hang glider towing. In fact, I think we've all felt it while platform towing, the surging of tension that occurs when the drum is slowly unwinding at the end of the tow. I attributed the pulsing in tension to the difference in the static and dynamic friction coefficients of the disk brake. While this may partly be true, the cycle itself could be caused by the bungee mode of the towrope.
So what does this mean to hang glider towing and weak links? It means that a weak link that is the perfect value on a spectra towline would be the wrong value on a polypropylene rope. It means a weak link that is perfect on a 150 foot towline could be less-than-perfect on a 200 foot length. It means that a weak link that is perfect on a large-diameter wheeled dolly on a concrete runway could be too weak on a rough runway or a less absorbing dolly. It means a weak link that works with a lightweight tug won't be right for a high-power, high mass tug. It means the towrope attachment point can be critical and the effect may be exacerbated if not in the proper tow position or if flying tandem.
It means that towing may be easier and weak links less prone to breaking if a small amount of shock absorption was added to low-stretch towlines. Perhaps a few feet of nylon rope on the end next to the pilot would be sufficient. I remember the smoothest tow I ever had was on a stationary hydraulic winch in Canada. I attributed the smoothness to the hydraulics but perhaps a contributing factor was the twenty feet of 1/2" nylon rope that was added to the end of the towline so it would hang down below the inside wingtip during turns on a step tow.
One last point to make is the breaking strength of rope is very dependent on the radius of any knot or bend in it. A weak link that is looped around a metal ring will fail at a higher value than one looped around a narrow loop of nylon.
Obviously, the correct weak link depends on many variables and identifying what works best would take a bit flight research (perhaps just a single day worth of smooth air flights). This did not occur because several things happened since I first wrote to you. First, Jim Murray was shipped off to the east coast to work a temporary assignment on the "Mars Flyer" -- a remote aircraft designed to fly in the atmosphere of Mars on the centennial anniversary of the Wright brothers first flight. Secondly, I decided to leave what on the top surface was my dream job as a NASA research pilot to pursue a job with the airlines. There were many reasons for this decision, not the least of which was NASA's continuing aeronautical budget cuts, emphasis on unpiloted aircraft, and their seemingly inability to get things done.
The NASA administrator's "Faster, Better, Cheaper" mantra has become a joke in the industry, reminding me of Jack Nicholson's presidential proclamation in the movie "Mars Attacks" when, in the midst of mass destruction, he gets on national TV and says something along the lines of, "I know I promised you these three things but hey, two out of three ain't bad." Unfortunately, with NASA's current record the quote would be more along the lines of "hey, none out of three ain't bad."
Until we ever do a real hang glider aerotow research project we can only make semi-educated guesses on the bungee mode and its effect on the towing of hang gliders. The intent of this writing was to point out some of the issues and to apologize for my failure to follow through with the research that I hinted at so long ago. A number of pilots sent emails to me at NASA asking me about the status of the project and encouraging me to pursue it. Unfortunately, when I went to retrieve all of those archived messages in my last week at NASA I found I had already been locked me out of the email system so I can't answer those emails individually.
Someday I may be able to get together with Murray and do the research. In the meantime, if you are ever flying the "friendly skies of United" look for me in the right seat of a Boeing 737 (especially if you are flying any of the west coast "Shuttle" routes).
1999 US Team
Many months ago I wrote to you with the idea of trying to do some hang glider aerotow testing, the intent of which was to define the actual loads encountered under differing conditions of tugs (low and high power), gliders (beginner, intermediate, and advanced), and pilot weights (single and tandem).
Yeah. Davis - the point man for the scientific advancement of hang gliding. Excellent choice.
Due to the magic of your Straub Report...
Sewer.
...instant interest was gathered and Malcolm at Wallaby Ranch was quick to call, leaving a message that he would be glad to sponsor the testing.
Yeah, good ol' brake lever velcroed to the downtube, bent pin secondary, 130 pound Greenspot Malcolm. Another excellent choice.
The brains behind the effort was Jim Murray, a NASA engineer who specializes in flight dynamics and is a true-life "Maguiver"...
MacGyver.
...with a reputation of being able to instrument a gnat's knee. Early in the Eclipse (aerotowed F-106) program, in which I was the test pilot, the computer simulation revealed the existence of an oscillatory tension mode in the towrope. The computer predicted something like a 12,000-pound steady-state tension value but overlaid on top of it was a continuous cycling value of several thousand pounds. In some cases this "bungee" mode would grow unstable and eventually exceeding the 24,000-pound weaklink.
- The maximum takeoff weight is 41,831 pounds.
- So you're towing at 0.29 Gs.
- And your weak link is 0.57 Gs.
- Why?
- Were you breaking any 24 thousand pound weak links?
The level of bungee present was dependent upon the two aircraft, the stability characteristics of the tethered pair, the towrope attachment points, and the towrope itself.
Like the Spectra line used in many hang glider tow operations, the exceptionally strong Vectran towrope we were planning on using had low stretch characteristics.
Good idea.
This meant low shock absorption and increased chances of encountering the bungee mode.
YOU'RE TOWING THROUGH *AIR*. AIR has EXCELLENT shock absorption. Ever notice the way when you're towing with nothing but Spectra or Vectran you're not feeling any shocks?
At the other extreme a nylon towrope would have been too springy and it too could result in dramatic (traumatic?) bungee oscillations.
Yeah, that's why you wanna go to the other extreme.
The computer predicted a certain level of stretch would give the best tow characteristics.
The computer was programmed by a moron.
For our initial flights we planned on adding a 50 foot section of nylon strapping in the middle of the 1000 foot length of the 3/4" diameter Vectran rope.
Idiots.
There was some skepticism about the mere existence of this bungee mode. The Germans had towed unconventional aircraft during the war years -- large troop-carrying transport aircraft, even multiple aircraft were towed. They also towed the swept wing Me-103 Komet, the first rocket-powered fighter. Pilots hated towing the Komet and a USAF test pilot who got the lucky straw to tow a captured Komet described the tow as the scariest experience of his life.
Shoulda tried going up with a Wallaby primary, Bailey secondary, and standard aerotow weak link.
Even NASA's predecessor, NACA had towed a propeller-less P-51 Mustang in an aborted attempt to compare it's real world L/D to what had been obtained through wind tunnel testing. The steel tow cable broke wrapping around the aircraft, interfering with control, and resulting in a crash.
Maybe it would've been a good idea to use a:
- four G cable
- two G weak link on the front end
- one and a half G weak link on the back end
Shitheads.
In all these tests there was never any mention of any bungee mode - did it really exist or was it some computer artifact? The answer was to run the simulation using conventional glider and tow aircraft numbers. The simulation indicated the bungee mode existed in normal everyday towing of sailplanes. Some of the old-time sailplane pilots expressed doubt over the simulation because over their years of towing experience they hadn't noticed any bungee mode. One said, "I've never felt no stinking bungee" (or words to that effect).
But let's find this problem anyway so we can solve it.
So Murray made up a couple of battery powered instrumentation packages, each about the size of a lunch box. We put one in a rented Pawnee tug plane and one in a rented Grob sailplane. The one at the front of the towrope read tow tension (using a solid state metal link at the attach point). The package in the Grob read longitudinal acceleration.
We launched in early morning conditions and the tug looked for level flight in smooth air. We flew at a couple different speeds and tow positions but most of the data was gathered at 55 mph, which was published L/D max for the Grob.
The data showed the bungee mode was very evident and I swear I could feel it.
What were you using for a towline?
It was always present to a minor extent but was easily excited by turbulence or maneuvering, in which case it took several cycles and perhaps twenty to thirty seconds to reduce it back down to it's normal small oscillations.
What were you using for a towline?
Probably the greatest excitation of the bungee occurred during the takeoff roll, most likely due to bumps in the dirt runway.
What were you using for a towline?
So what about the issue of the bungee mode and its effect on the Eclipse program? We found that as predicted, there was a stable region on low tow where the bungee was minimized and where the F-106 was extremely easy to fly on tow. Outside of that stable region the bungee became more of a factor and the F-106 became more and more of a handful to fly. In fact, in a conventional high tow position it was quite unstable and if I wasn't extremely careful the weak link would fail within several seconds.
- Oh. You put fifty feet of nylon into the tow configuration and you detected a bungee effect. Who'da thunk?
- Oh. You got surges as a result of putting fifty feet of nylon into the tow configuration, got out of the optimal position, and blew a weak link 0.23 Gs short of the FAA's legal minimum and 1.43 Gs short of the FAA's legal maximum for sailplanes. Who'da thunk?
- What results did you get when you towed WITHOUT the nylon?
Without doing any dedicated tests with hang gliders I can only guess but I think it is reasonable to expect the bungee mode is present in hang glider towing.
With the total fucking morons who put polypro into the system? Yeah. Pretty good bet.
In fact, I think we've all felt it while platform towing, the surging of tension that occurs when the drum is slowly unwinding at the end of the tow. I attributed the pulsing in tension to the difference in the static and dynamic friction coefficients of the disk brake. While this may partly be true, the cycle itself could be caused by the bungee mode of the towrope.
It could be. But since nobody's feeling it when they're aerotowing with Spectra it's probably not.
So what does this mean to hang glider towing and weak links?
I have no idea. Let's consult Drs. Colletti and Tillman. With all their experience with all the factors and variability that exist in the big, real world and backgrounds in formal research it's really hard to believe that we wouldn't get some really good answers.
It means that a weak link that is the perfect value on a spectra towline would be the wrong value on a polypropylene rope.
Idiot.
It means a weak link that is perfect on a 150 foot towline could be less-than-perfect on a 200 foot length.
Idiot.
It means that a weak link that is perfect on a large-diameter wheeled dolly on a concrete runway could be too weak on a rough runway or a less absorbing dolly.
Idiot.
It means a weak link that works with a lightweight tug won't be right for a high-power, high mass tug.
Idiot.
It means the towrope attachment point can be critical and the effect may be exacerbated if not in the proper tow position or if flying tandem.
Here are a few thoughts...
- Align the towline with the tug's thrust line.
- Tow the glider two point.
- Stay in position behind the tug.
- Use one and a half G weak links.
It means that towing may be easier and weak links less prone to breaking if a small amount of shock absorption was added to low-stretch towlines.
Look dude...
- Use a fuckin' weak link one and a half times fuckin' max glider weight. They don't break until the tow's over anyway.
- KEEP ELASTIC MATERIALS THE HELL OUT OF THE SYSTEM.
Perhaps a few feet of nylon rope on the end next to the pilot would be sufficient.
Idiot.
I remember the smoothest tow I ever had was on a stationary hydraulic winch in Canada.
Yeah. Mike Deadeye Robertson, no doubt.
I attributed the smoothness to the hydraulics but perhaps a contributing factor was the twenty feet of 1/2" nylon rope that was added to the end of the towline so it would hang down below the inside wingtip during turns on a step tow.
A living - though depth perception challenged - testament to the benefits of shock absorbing towlines and ideal weak links.
One last point to make is the breaking strength of rope is very dependent on the radius of any knot or bend in it.
Yes.
A weak link that is looped around a metal ring will fail at a higher value than one looped around a narrow loop of nylon.
You've tested that, right?
They virtually ALWAYS blow at a knot - and that takes the metal ring versus narrow loop of NYLON out of the equation.
Obviously, the correct weak link depends on many variables and identifying what works best would take a bit flight research (perhaps just a single day worth of smooth air flights).
No no no no. What you really need is practical information and knowledge gleaned from the real world of aerotowing, developed over decades and hundreds of thousands of tows by experts in the field. This is the only possible way to determine that a loop of 130 pound Greenspot on a solo one or two point bridle is the correct weak link which works best for most pilots and gliders and is usually near the USHPA recommendation of a nominal one G weak link for most pilots.
The CORRECT weak link depends on ONE variable - the MAXIMUM CERTIFIED OPERATING WEIGHT of the glider. Read the fuckin' manual.
The NASA administrator's "Faster, Better, Cheaper" mantra has become a joke in the industry, reminding me of Jack Nicholson's presidential proclamation in the movie "Mars Attacks" when, in the midst of mass destruction, he gets on national TV and says something along the lines of, "I know I promised you these three things but hey, two out of three ain't bad." Unfortunately, with NASA's current record the quote would be more along the lines of "hey, none out of three ain't bad."
This is EXACTLY why hang glider aerotowing is perpetually up to its neck in this bullshit. Go back and read some Hewett.
- Don't wanna spring for a good tension regulating payout winch? Use a nylon towline and a tension gauge.
- Don't feel like engineering a safe release system? Use a half to one G weak link and put somebody on the back of the truck with a machete.
Until we ever do a real hang glider aerotow research project we can only make semi-educated guesses on the bungee mode and its effect on the towing of hang gliders.
A REAL hang glider aerotow research project? Why? We've got several decades of experience and hundreds of thousands of tows conducted by numerous aerotow operators across the county on which to base our decisions and have a really good de facto weak link standard - a proven system that works.
The intent of this writing was to point out some of the issues and to apologize for my failure to follow through with the research that I hinted at so long ago. A number of pilots sent emails to me at NASA asking me about the status of the project and encouraging me to pursue it.
Funny how much apathy and hostility there was when I did the in flight testing that you never got around to and published a solid set of standards - which were essentially a reinvention and refinement of sailplaning's wheel.
Do you mean to tell me they wrote an article that wasn't insipid and self-congratulatory to the extreme? I've found their entire series on aerotowing to come off rather poorly to say the least. A sad waste of such exalted and highly qualified medical professionals. How do I know this? Well they won't stop patting each other on the back about how great they both are. Pardon me while I puke. 