G-load on our airplanes

I'm sure this have been hased and re-hased, but a quick search of the archives didn't turn up anything.

Like a lot of others here, I've flown full-size aircarft in aerobatics. One of the thinsg we concern ourselves with is the g-load during a manuever. While airframe intergrity is a concern, usuallu the airframe is up to more stress than the pilot is. The pak loads experienced by the top unlimited pilots exceed 9 G for very brief periods, but the aircraft themslevs don't have enough power to maintain that G-load for very long due to induced drag. Typical loads are usuauly +/- 4-6 G.

I began to wonder what kind of loads our airplanes undergo, knowing that it must be pretty high compared to full-sized aircraft. I used a lap time of 5.85 s for a 63 ft radius circle. This yields a speed of 25.77 m/s (about 59 MPH). I then plugged in the acceleration required to make a 1.5 m radius turn at this speed, and get 442.65 m/s^2. Divide that my 9.8 m/s^ to get G-load. I get 45.17 G. I doubt that our airplane's structures can survive that many G. If I'm right, then it's not possible to fly the prescibed pattern in the book (again, I'm sure this ground has been plowed over hundreds of times). Of course, the speed is probably rapidly decreased during hard corners, so the radius of the turn isn;t constant because the speed isn't constant, but this is a first-order guess.

Does anyone have any data about the loads on our planes when flying the pattern? If I guess at 15 G, then I get a minimum radius of about 4.5 m (about 14.7 ft). If I allow 20 G, I get a radius of about 11 ft. Still more than twice the prescibed 5 ft.

Does anyone know how many G we put on these models?

Kim

I did an article in Stunt News some years ago which basically said, "a stunt ship flies level with a coefficent of lift at about .1. A typical 00 series airfoil can generate lift up to somewhere around 1.8 thus there are only about 18 g's of lift available and backing into the centr. force formula you get a turn radius of about 15 feet which just so happpens to coincide with what others have others found to be the real turning radius od our machines. Percieved turn radius is of course another whole subject.
I will now watch for the flak.

I certainly didn't intend to start a war over this...

I heard from some of the RC Pylon guys that they put an accelerometer on one of the FormulaI airplanes. The readings they got were somewhere above 30g's in the turns.

I can see this as these models fly about 3 to 4 times as fast as our stunters, hence they would be flying at a much lowere C sub L in level flight and be able to pull many more G's before the airfoil runs out of lift.

>I heard from some of the RC Pylon guys that they put an
>accelerometer on one of the FormulaI airplanes. The readings
>they got were somewhere above 30g's in the turns.

I find this a bit dubious. I'm sure this is what you were told, but a *susatined* 30g requires a maximum coefficient of lift that's quite high. I also question if these planes have enough power to sustain a 30g load. I don't dobth that they may have occasional peak loads of 30g, but a sustained 30g seems doubtful to me. I am, however, admittedly ignorant about what speeds and turn radii these planes fly. on the back of my envelope, I figured that if th eplane is going, say, 200 mph, a 30g turn yields a radius of about 71 ft (first order v^2/r stuff, assuming constant speed). A 90 deg turn with this radius at 200 mph would take about 0.38 s to complete at constant speed (about 1.5 s for a full 360). I haven't watched pylon racing in a long, long time. Do they really turn that sharply?

As I stated in the message above the max G loading in the turns was 30+. This did not mean that it was a sustained G loading. F1's are approaching speeds above 200mph for a really good one. They very seldom fly wings level and most fly banked all the way round the course using elevator to pull them through the turns.

Those pylon racers are really cool. My wife suggested that I take up Formula One. I explained to her that the airplanes took as long to build as CL stunt planes and lasted as long as combat planes. "It's not for sissies," she said.

>I did an article in Stunt News some years ago which
>basically said, "a stunt ship flies level with a coefficent
>of lift at about .1. A typical 00 series airfoil can
>generate lift up to somewhere around 1.8 thus there are only
>about 18 g's of lift available and backing into the centr.
>force formula you get a turn radius of about 15 feet which
>just so happpens to coincide with what others have others
>found to be the real turning radius od our machines.

That's pretty consistent with Wild Bill's measurements, as well.

Kim,

As Brett mentioned in here, Bill Netzeband came up with some numbers.

Background: (Wild Bill's background and numerous publications in our magazines is far too extensive to even suggest, here. Stipulate that it is excellent?)

Bill doubts any CL Stunt model will exceed 35g in a corner.

Don -- IMHO, we go beyond simple aerodynamics, here. I have a "smattering" of a basis along this line plus a long amateur (in the best sense) interest in both the practical and theoretical sides.

Igor's stop-motion photography suggests (to me, at least) that the models reach interestingly large Angles of Attack at places in corners. Surely the airfoil section has passed stall AoA? Fuselage rotation and actual (skidding) path shown in the photos do not track 'clean' (unstalled) motion.

All this said, it would be interesting to see what an accelerometer -recording both g and time axes - would show in a top performing model flown by a top level flier. (I'm thinking "rotating-drum" or "moving paper" nature of trace -- we should be able to miniaturize something like that these days. "We"(?) -- way beyond MY competence!)

It's V^2/gR. For American units, because we're all Americans, use V in ft./second, loop radius R in feet, and g = 32.2 ft./sec^2.

>Igor's stop-motion photography suggests (to me, at least)
>that the models reach interestingly large Angles of Attack
>at places in corners. Surely the airfoil section has passed
>stall AoA? Fuselage rotation and actual (skidding) path
>shown in the photos do not track 'clean' (unstalled)
>motion.

Should bear in mind that there is apocryphal evidence that very high rates of change of AoA greatly alter the maximum Cl. It's mentioned somwhere in Abbott and Van Doenhoff, but I couldn't find it last time I looked. Positive rate of AoA was observed to *greatly* increase the C/L for a given AoA. You might guess that it also greatly altered the stall AoA compared to the static data. I don't think I have ever seen anything definitive on the topic, however.

We did this exercise for CL Comabt planes and came up with something over 30 G's. This was based on the radius of the smalled loop the plane would make and the speed.

Whether it's "instantanous" or sustained, doesn't matter if breaking the wing is the issue.

At sea level, the air pressure is 2,116 pounds per square foot. A BIG combat job has 3 sqaute feet of wing ares. This the total air pressure on the wing is 6,348 pounds.

The question (for Howard Rush and other aero engineers) is how much pressure differntial can a wing generate?

By working backwards, we figureed that a 2-pound airplane needs 60 pounds of lift to pull 30 G's. Thus the required pressure differential is less than 1%.

We just go fly combat planes and see if they break, of course. I did actually do a static test on the Son of Snort spar. I hung some PVC tubes from the spar and poured sand in them to load it. I quit when I got to 145 mph or so worth of load. I didn't want to clean up all the sand when it spilled in my basement.

Dover, 1959 edition, p143, refers to NACA TR 618, which doesn't seem to be available?

http://naca.larc.nasa.gov/reports/1938/naca-report-618/

Those guys must have had fun.

Lou, that picture shows AoA in corner 7 degrees (unfortunately not very exact). It is far under stall AoA.

I did that picture to proof my spreadsheet saying that THIS model is able to do ~3.5m (11-12ft) radius at 20G. The theoretical AoA had to be 6.4 deg and lift coefficient 1.8 (in that circular path at elevator deflection 30 deg). It is even before separation on flaps - to suppress "bumps" on polar if you remember.

Actually that AoA is limited at full deflection of elevator just on edge of stall AoA (16 deg), so it will not stall also in straight flight at low speed. (landing or so)

But you are right with that AoA over static data. One thing is pitching rate as Brett mentioned - that pitching rate "helps" to make circulation making lift - or - you can imagine it easier like airfoil having different AoA at leading edge, another (higher) at middle and another (even higher) at TE - as chord is rotating relatively to incoming airflow. The result is that you do not have original symmetric airfoil, instead of that you have little chamber. I did analyze of my airfoil in circular flow modeled as 3.5m radius and it really makes significant difference (positive ).

Another thing is, that separation at high AoA needs some time. The airflow must "fold" and "fill" place with low pressure and so make separated vortex. So if you have instant but only short deflection in time, it has no time to fully form the vortex and it will last more degrees than static AoA for long time.

But to original question, it is relatively light model with strong foam wing, so those 20g (carrying only fuselage) do not make any structural problems.

HOWARD RUSH....told me last weekend the Paul Walkers plane was pulling 29g's. I aksed and he fingered it out for me. Thanks Howard.

For Howard Rush. Paul's planes, or I assume any of the top competition flyers planes are pulling 29 g's over the time of the onset of the increase load to the time the load reverts back to normal g. Approximately, how long is say; a hard competiton corner from initiation to return to normal flight?
Thanks in advance.

Hi

There was a thread on turning radius and some strobe photos indicated turn radius of around 10 to 15 feet though it looks much tighter. You can do the "G" calc. Don't be too quick to to dismiss 45g. Look at the old F3B 2 lap speed run for R/C gliders. Given the times, lift capacity of the wing and the fact they turn an inverted half loop on full up, the cals show 40g + and that's loaded with ballast with much longer and thinner wings than C/L stunt.

Given the distribution of lift force across the span is close to eliptical, the mean moment centre of lift is at only about 1/3 of the semispan. The moment stress is not as great as you might at first think.

not sure how accurate this is but once along time ago Mike Moreland told me that someone did a study on fast combat planes and found that in their tightest corners they were pulling 80 g's sounds realistic to me.. BEN

80 is what I remember.

I just took a speed and a loop radius that Dan or Scott suggested and figured the acceleration in g's. I didn't do anything sophisticated.

Wow. This has been eye-opening. I never considered the non-linear effect of rapid pitch changes or that it must take time for the flwo to separate. Both of these effects alond could really crank up the g loads for brief periods.

And, I'll take back what I said about sustained high g loads. Maybe our airplanes really can maintain some huge loads for extended periods. 80 g. Sheesh...

Kim Elmore

P.S.: Ben, say ""Hi" to your Dad for me! I'll try to make it out this weekend if you guys are around. Gotta take our Cessna 140 in for it's annual in$pection...

Along the lines of the rapidity of applied and released G's, have you ever noticed how after you dorked your landing gear pretty badly backwards, maybe close to but not enough to puncture the wing, how you can't bend it back to it's original position while still on the plane, at least not without serious damage to the plane... You have to take it off, put it in a vise and apply forces that would no doubt shatter any style of balsa, plywood, laminated or whatever.

I think this really illustrates the even load /speed distribution theory. Kind of like the tales you hear when a tornado whips a piece of something relativley fragile like a straw though a telephone pole.

Evidently there is a density issue though. (Yeah, maybe Eric is just plain dense!) But really, Jump in the water from 10ft, you make a splash, but jump in the water from 1000ft you make a splat! The water can't get out of the way fast enough. We are after all mostly water ourselves.

Well anywho, if the 45G figure is correct, that means a 4lb airplane sustains 180lb's for a split second?!!! Yikes, I better stock up on the carbon fiber and epoxy!

>Wow. This has been eye-opening. I never considered the
>non-linear effect of rapid pitch changes or that it must
>take time for the flwo to separate. Both of these effects
>alond could really crank up the g loads for brief periods.

It *is* just an anectodal observation. The in-flight studies indicate 13-14 foot radii, which is perfectly consistent with static values of Cl. With no accelerometer data, hard to say if it "spikes" anywhere in process. But it couldn't be for long - because if it was, the radius would be tighter.

I mentioned it, but that' doesn't mean I completely buy it.

>And, I'll take back what I said about sustained high g
>loads. Maybe our airplanes really can maintain some huge
>loads for extended periods.

Depending on what you mean by extended, I guess. A corner only takes about .25-.3 seconds, total, and you'd have to expect the transient effects (if any) to be far, far shorter. If I was going to design to a G loading, I'd assume it was about 15G's, and add whatever margin seemed wise. That's still pretty impressive to me - that means 60+ lb of lift, and maybe 25-30 lbs load on the center section of the wing. If you want to hold the wingtips and put 25 lbs of lead on the fuse, I think you'd be impressed with the dihedral that generated! You can see it in flight if your eyes are fast enough.

I think you'd be impressed with the
>dihedral that generated! You can see it in flight if your
>eyes are fast enough.
>
> Brett

Yes, I have seen it coming out of the wingover, only on a very few planes, it happens so fast. I have also seen it on the stab.

Here's one of those areas where some simple instrumentation on a stunter would answer nagging questions. Analog Devices ADXL type accelerometer, PIC controller on a small battery. Two axes, and we could get that much discussed line tension measurement as well..

[photo not recovered: 3e69bb3870f12ad5.jpg]

"Ideas are like rabbits. You get a couple, learn how to handle them, and pretty soon you have a dozen." -John Steinbeck