What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Wind Shear Video

Vac

Well Known Member
Benefactor
We were flying some AOA software tests last week and had a camera installed to record the standby airspeed indicator along with the normal forward view. My home base is a tree canyon close to the Gulf Coast 5 miles east of Eglin AFB in the Florida Panhandle. Winds were light on the surface (80’ MSL), variable around 090 degrees or so at takeoff, but I encountered over 30kts at pattern altitude. As I climbed through about 3500’ MSL, winds dropped off to light and variable. This abrupt shift in velocity in such a small altitude band meant the chance for wind shear would be high during RTB, and sure enough, it was: https://youtu.be/0RiB6_8bx3c

The important thing to note is how rapidly the bottom dropped out in the transition and flare. Stable is a relative term under these conditions in a light plane. I was flying a slight fast approach (high rate beeps of the tone) and transitioning to ONSPEED during transition to landing. The AOA tone provided good SA throughout, but it was sporty. If 15 kts of rapid shift is “severe” wind shear, this event was close. If shear results in a rapid loss of airspeed, it’s called “performance decreasing” and if it causes a rapid increase in airspeed, it’s called “performance increasing.” The recent Van’s EAA webinaire addressed this and I’d recommend it to anyone that hasn’t watched it. If you’d like more info on our AOA work, visit our site at http://www.flyonspeed.org

Fly safe,

Vac
 
Last edited:
We were flying some AOA software tests last week and had a camera installed to record the standby airspeed indicator along with the normal forward view. My home base is a tree canyon close to the Gulf Coast 5 miles east of Eglin AFB in the Florida Panhandle. Winds were light on the surface (80? MSL), variable around 090 degrees or so at takeoff, but I encountered over 30kts at pattern altitude. As I climbed through about 3500? MSL, winds dropped off to light and variable. This abrupt shift in velocity in such a small altitude band meant the chance for wind shear would be high during RTB, and sure enough, it was: https://youtu.be/0RiB6_8bx3c

The important thing to note is how rapidly the bottom dropped out in the transition and flare. Stable is a relative term under these conditions in a light plane. I was flying a slight fast approach (high rate beeps of the tone) and transitioning to ONSPEED during transition to landing. The AOA tone provided good SA throughout, but it was sporty. If 15 kts of rapid shift is ?severe? wind shear, this event was close. If shear results in a rapid loss of airspeed, it?s called ?performance decreasing? and if it causes a rapid increase in airspeed, it?s called ?performance increasing.? The recent Van?s EAA webinaire addressed this and I?d recommend it to anyone that hasn?t watched it. If you?d like more info on our AOA work, visit our site at http://www.flyonspeed.org

Fly safe,

Vac

Yet another reason to use the AOA tones....instead of looking at airspeed when landing.
 
Mickey,

Was 090 variable, light at the tree line, near as I can tell. You can see the transition from the crab to the right during final as I descend out of the heavy air. Could have shifted slightly southeast at the surface, no doubt. Interesting ride through the different layers. Eglin reported 050 at 9 a few minutes prior. Daily sea breeze was starting to fill in and winds were beginning to clock southeast out on the water just to the south of both airports.

Cheers,

Vac
 
Looks like I am completely missing it in the video.

I was once in a wind sheer on short final and the experience was completely different than seen in this video OR that at least I can identify it.
 
Looks like I am completely missing it in the video.

I was once in a wind sheer on short final and the experience was completely different than seen in this video OR that at least I can identify it.

Did you see the airspeed instantly drop? Hard to do that in a plane without wind shear or an arresting hook, unless you hit something head on !
 
I had a similar situation in my -9 a number of years ago during a BFR.

On speed, slight breeze over the trees, as soon as we dropped below the trees I lost all lift. My AoA went from plenty of lift to screaming in my ear.

Full throttle and I left it with a level pitch attitude. I dropped about 30 feet vertically, contacted the ground on the mains, no prop strike and the tailwheel didn't touch. (Thank goodness I don't fly a tricycle gear, or the results would have been much different.) I then bounced back up almost vertical 20 feet, according the witnesses. I had just enough time to firewall the throttle and the power came up just as we touched. I had never seen that before!

I flexed the right gear enough to crack the gear leg fairing.

I didn't call it "wind shear" but I'm sure that is what it is because I went from a headwind to no wind and the bottom just dropped out; plenty of lift to no lift in the blink of an eye. I thought for sure we were dead. The CFI in the right seat was really impressed we were able to fly away. During the debrief, we discussed this at length and he really didn't have anything to say other than being impressed by my quick reflexes as I yelled "STALL" as I shoved the throttle home before he realized what was going on. I honestly think I reacted to the feeling in the stick before I reacted to the tone in my ear.
 
Last edited:
That's the quality of the 9 we had. It was a handful in gusty windshear type conditions. I had a similar experience with ours. We were halfway out of control; I was moving the stick what seemed like stop to stop trying to maintain proper attitude. We touched down hard; I was concerned the nosegear wouldn't hold. The 9 was a good airplane for us but I didn't like flying it in rough air.
 
... The 9 was a good airplane for us but I didn't like flying it in rough air.

A little drift... I have learned to slow way down in turbulent air and just let the plane roll with the bumps rather than hammer through them. I also turn the AP off on those conditions as it makes for a much smoother ride.

Still LOVE my -9!
 
Did you see the airspeed instantly drop? Hard to do that in a plane without wind shear or an arresting hook, unless you hit something head on !
NO, watched it multiple times and don't see the airspeed drop, only gradually going down till the turn on the ground.
 
quick drop

NO, watched it multiple times and don't see the airspeed drop, only gradually going down till the turn on the ground.
Looking at the ASI in the bottom right corner, there was a quick drop from about 80 to 60. The drop from 70 to 60 was very quick.
 
That last sudden drop in airspeed from 70 to 60 MPH when the bottom fell out, looks like it coincided with your final flare to a 3-point attitude. That may have helped ‘cushion the blow’, and also points out the great forgiving nature of the RV4 “land-o-magic” landing gear when in that 3-point attitude. It’s the nicest landing RV model there is. I also noticed the sudden tone change when this last negative performance wind shear was encountered. I’m not sure the tone helped in this situation because it happened so quickly.
 
Last edited:
N941WR;1431762 I flexed the right gear enough to crack the gear leg fairing. [/QUOTE said:
I thought only I had done that, and it was an installation issue. I have made some mods to allow more movement.

BAVAFA - -I had to watch twice to catch it right in final flair.
 
Normal Landing Comparison

Scott, agree. What tone did let me know what my energy state was as soon as the event occurred, which helps to reinforce the requirement to hold the aircraft in the proper landing attitude.

One other point to ponder is that a go-around would have also been perfectly acceptable. One interesting airline statistic is the small number of go-arounds after an unstable approach or bounced landing--I think this demonstrates that as pilots, the instinct to land when in close proximity to the ground is pretty strong, and no doubt when my wheels contacted the ground I shifted from "fly" mode to "land" mode. Also, if you watch the video carefully, there is a portion in the landing roll where the tail comes up slightly. This is partly due to the nature of our runway, and partly due to my right elbow failing to properly maintain the tail-low attitude during deceleration.

There is usually a "tree canyon" effect at our field, but maintaining an ONSPEED condition during the transition below the tree line and into the flare usually provides sufficient cushion to transition to the ambient conditions at the runway surface. Not so in this case. I was fortunate that the mains were close to the ground when airspeed dropped and AOA increased dramatically.

I thought it might be interesting to compare the video at the start of this thread with a normal landing under ideal conditions: https://youtu.be/gB3Vxkabo3c. Note the difference in how the standby ASI behaves throughout, as well as the steady transition (trend) in the tone pattern.

Cheers,

Vac
 
Last edited:
Here's a tail cam video of a wind-shear landing I made recently. The shear wasn't front-to-back, but mostly side-to-side. I came in hot to account for the uncertainty of the gusts, but almost overdid it. I floated for a long way and used most of the runway to get off. In the flare, my nose was high and I couldn't see down the runway so I drifted toward the left side (bad thing). My AoA was going off so I knew I would set down soon regardless :eek: In the video, the ground speed is displayed and the wind was 6 Kts, varying from dead ahead to 45 degrees from the right right. Enjoy my ordeal...

https://www.dropbox.com/s/m4yimpe14i97gkw/2020-05-15 GMU Gusty Landing.mp4?dl=0
 
Data Plot

We were recently working on some data for the ONSPEED AOA project and decided to look at the May wind shear encounter as a natural lab experiment. My RV-4 has two independent AOA systems in addition to the Dynon EFIS and a calibrated test boom on the left wing.

I re-edited the video I previously posted and uploaded it to YouTube in high resolution. You can view it here https://youtu.be/_pic-kISjAY

This figure plots airspeed and AOA for the 30" of base/final prior to the wind shear event. As you can see looking at the plot and in the video if you watch the standby airspeed indicator, it was a bumpy ride down final:

3c039a_7cc3bb8e1f2b437382e9207cfb4f2a2c~mv2.jpg


The next figure shows the final 15" prior to the bottom dropping out. Note the difference between the boom and the ONSPEED computed alpha. Generally, the low inertia vane leads the high inertia wing/airplane; but as soon as the actual event occurs, roles reverse and the pressure solution "out peaks" the boom (i.e., more accurately captures the aerodynamics). Frankly, we don't fully understand these dynamics yet (learning in progress!); but figure that a picture is worth 1000 words:

3c039a_2f8e3ecc49e24749a03bd8ca240cd8b1~mv2.jpg


The next figure shows the actual severe wind shear event itself (with "severe" wind shear defined as a wind change greater than 15 knots). If we average the EFIS and ONSPEED system computed indicated airspeed, the wind shears 15.395 knots in 1.3 seconds during the landing transition (likely an effect the the tree canyon that surrounds my home runway). The NACA 23013.5 airfoil on the mighty RV-4 fully stalls at 20 degrees geometric angle of attack at 1G. Compare that to the ONSPEED computed AOA at the peak of the grey curve and yellow spike. It was truly a "glove save" event. The boom measured a peak alpha of 17.24 degrees; but as I stated earlier, we don't fully understand these dynamics yet, and as the video shows, the airplane was definitely stalled during the event. It was **** fortunate the wheels where close to the turf when this occurred. The audio stall warning is also evident in the video:

3c039a_d76174ed3cda41a2989891568ddafb9e~mv2.jpg


All's well that ends well; but in hindsight adding full power and initiating a go-around would have been a smarter option under these circumstances. Being in the proper attitude was the thing that saved my bacon when the wheels touched; and if it's not evident, I'm working hard to fly the airplane throughout the landing roll to get safely to a stop. This is a classic example of a salvaged unstable landing, i.e., I pressed too hard. Thanks again to Van for engineering one of the best airplanes I've had the pleasure to fly: lots of maneuverability, power and aerodynamic margin to compensate for a momentary lapse of reason (shout out to Pink Floyd).

Things are happening quite fast during the actual event; and it's somewhat difficult to assess the stall warning provided by the AOA system. Here's a simple 4G acellerated stall (i.e., stall occurs at 4G's) that demonstrates how the system keeps up with dynamic maneuvering. Unfortunately I left the overload warning system (the "G Limit" voice warning you hear in the video) in training mode, so it's activating at 2.5 G's. I should have reset that prior to the maneuver (no excuse): https://youtu.be/K4POvXIu1QA

Fly safe,

Vac
FlyONSPEED.org
 
Last edited:
Interesting ...

Vac,
I greatly appreciate the fine work you guys are doing. With only constructive intent, a little devils advocate (speculation):

Why are you convinced the “ground truth” is closer to the V2 output than the boom probe? Are you really sure the wing experienced aerodynamic stall (flow separation) as suggested by the high AOA reading, or just that there was a significant uncommanded (downward) vertical acceleration? It would seem to me a significant downdraft could both cause the Nz without a spike in AOA due to the change in direction of the relative wind.

Perhaps such a downdraft could have resulted in the (relative) pressure spike in the AOA probe (if one port is shaded from above relative to the other). The subsequent 2 peaks could perhaps be (damped) pressure transients in the pneumatics of the AOA system ... ie ringing pressure pulses from unsteady flow. This could be tested on the ground to explore the dynamics of that system (characteristic frequency for example). Similarly the inertial characteristics of the vane could also be explored on the ground (perhaps confirming that the vane probe was simply numb to the entire hair raising affair).

Of course your team has much more experience, knowledge, and data available, so perhaps these ideas have already been validly ruled out.

In any case, excellent job bringing the event to a successful ending ... no one hurt, no bent metal, and significant learnings. That’s a great glove save any day. Thanks for sharing.

Peter
 
Thank You!

Peter,

Excellent thoughts. We post here and on our web site with the intent that we'll work in a fish bowl; and let the folks smarter than us contribute to the project through critique.

I'm frankly NOT convinced, the V2 solution is more accurate that than the boom, that's speculation only. We consider the boom as a "ground truth" reference...we just aren't sure of the inertial effects and aren't 100% confident in our dynamic correction algorithm we're applying to correct for upwash, G and pitch rate (not to mention aerodynamic effect which we are currently assuming to be negligible due to the low airspeed we're working at--perhaps a bad assumption). The error tables assume the boom is correct; which, if I understand you correctly, would correlate with a down draft event. The stall assessment is based on my Mark I Butt Device, which isn't recorded in the data stream ;) The airplane stopped flying momentarily; and things worked out due to proximity of the landing gear with the ground and reasonable pitch control that kept the airplane in the landing attitude. I was fortunate.

By means of comparison, here's a plot of the accelerated stall data from the video in the previous post. I'm listening to the yellow line in the plot:
3c039a_b764400b5fe64ed9a9e8e32dc534b44a~mv2.jpg


Contrast that with a 1G stall:
3c039a_79b3fb31cc05492fa10fc72c68ad3254~mv2.jpg


We are using wind up turns, high G stalls and vertical transient response tests to define performance during maneuvering--if you (or any other readers) have any thoughts on that methodology or boom dynamic correction; we can definitely use the insight!

Thanks again,

Vac
 
Last edited:
Video was great as usual and I liked where I could see where that stall, which one would expect to be greater, didn't stop the turn and you lost around 500' in the recovery. I know it wasn't your concern at the moment, but it's nice to see what happens after a stall like that and not bringing the wings level immediately. Thanks !
 
Hi Carlos,

Yeah, had to finish up the 360 to RTB...I was beating the undercast in the video back to initial and didn't want to divert.

To get the airplane to stall at 4G's, I add full power and start with a "break" turn (unload, roll, set, pull) where "set" means getting the stick back in the center laterally before applying max G. In this case, I'm shooting for just under 6. A good rate to pull is to add 2 G's per second. Airspeed is going to bleed off rapidly, and with it G. The trick is to get to the aerodynamic limit fast enough to force a stall. The way that you know you've stalled is that the nose stops tracking across the horizon, and in my airplane, there is a distinct "thump" sensation. Just like a regular stall, all you have to do to recover is ease the stick forward slightly and the airplane is flying again. As you can see in the video, you can continue to bat turn if you want.

Even though it's within limits, I don't do this on a regular basis--only for flight test or if I'm teaching in an RV. Our airplanes are made out of metal, which remembers abuse! At test weight, that 5 1/2G break imparts a load of almost 6900 pounds on the structure, and when the airplane stalls, its effective weight is 5000 pounds. I've flown enough bent airplanes in the military to know that there is a price to pay when you are constantly operating around the top of the flight envelope.

Good to hear from you,

Vac
 
Last edited:
You might try stalls at a range of G loads (1G, 2G... 6G) and see what you see. I've done stalls at 0.9 G in the RV-9A, and I've also done accelerated stalls at 1.2 G in straight and level flight. (1.2G occurs at 110% of 1 G stall airspeed, set the airspeed and pull back briskly but smoothly.)

With the relatively low G limit of the RV-9A and the even lower G limit of my surgically reinforced spine, I'm not in a position to go explore those phenomena myself. Besides, your plane is better equipped to measure the phenomena in detail.

Don't know what you'll see, but you might find something interesting.

Ed
 
Hi Carlos,

Yeah, had to finish up the 360 to RTB...I was beating the undercast in the video back to initial and didn't want to divert.

To get the airplane to stall at 4G's, I add full power and start with a "break" turn (unload, roll, set, pull) where "set" means getting the stick back in the center laterally before applying max G. In this case, I'm shooting for just under 6. A good rate to pull is to add 2 G's per second. Airspeed is going to bleed off rapidly, and with it G. The trick is to get to the aerodynamic limit fast enough to force a stall. The way that you know you've stalled is that the nose stops tracking across the horizon, and in my airplane, there is a distinct "thump" sensation. Just like a regular stall, all you have to do to recover is ease the stick forward slightly and the airplane is flying again. As you can see in the video, you can continue to bat turn if you want.

Even though it's within limits, I don't do this on a regular basis--only for flight test or if I'm teaching in an RV. Our airplanes are made out of metal, which remembers abuse! At test weight, that 5 1/2G break imparts a load of almost 6900 pounds on the structure, and when the airplane stalls, its effective weight is 5000 pounds. I've flown enough bent airplanes in the military to know that there is a price to pay when you are constantly operating around the top of the flight envelope.

Good to hear from you,

Vac

Thank you. On my TO-DO list once my back injury gets better and I can actually fly my RV8 again, and once the pandemic is under control where it's also safe, is to test my OnSpeed installation then go to Florida to meet you in person and fly with you so I can pick up some more RV tips. Be safe !
 
Peter,

Excellent thoughts. We post here and on our web site with the intent that we'll work in a fish bowl; and let the folks smarter than us contribute to the project through critique.

I'm frankly NOT convinced, the V2 solution is more accurate that than the boom, that's speculation only. We consider the boom as a "ground truth" reference...we just aren't sure of the inertial effects and aren't 100% confident in our dynamic correction algorithm we're applying to correct for upwash, G and pitch rate (not to mention aerodynamic effect which we are currently assuming to be negligible due to the low airspeed we're working at--perhaps a bad assumption). The error tables assume the boom is correct; which, if I understand you correctly, would correlate with a down draft event. The stall assessment is based on my Mark I Butt Device, which isn't recorded in the data stream ;) The airplane stopped flying momentarily; and things worked out due to proximity of the landing gear with the ground and reasonable pitch control that kept the airplane in the landing attitude. I was fortunate.

By means of comparison, here's a plot of the accelerated stall data from the video in the previous post. I'm listening to the yellow line in the plot:
3c039a_b764400b5fe64ed9a9e8e32dc534b44a~mv2.jpg


Contrast that with a 1G stall:
3c039a_79b3fb31cc05492fa10fc72c68ad3254~mv2.jpg


We are using wind up turns, high G stalls and vertical transient response tests to define performance during maneuvering--if you (or any other readers) have any thoughts on that methodology or boom dynamic correction; we can definitely use the insight!

Thanks again,

Vac

More data is always better. :)

The most significant observation I can make from the 5G and 1G stall data is a pronounced positive AOA transient of several degrees just at the point of stall (flow separation). I observe that such a transient is not in evidence in the “wind shear event.” But perhaps this is just confirmation bias colored by my earlier speculation.

If this transient is repeatable, I think it provides a useful benchmark for the V2 system. That we observe a significant signal change from V2 contemporaneously (though opposite in direction) suggests it too is accurately capturing the transient flow re-arrangement at the point of stall. That’s good to know.

Accurate AOA estimation at up to +5G is an ambitious (and worthy) goal. Are you incorporating AHRS data with the AOA probe pressure to generate the real time AOA estimate (and tone) ? Or is the “extra” data not needed dynamically and only used for calibration?

Peter
 
Please send a link to this thread to the next chump who tries to tell you that winds don’t affect your airspeed. I got into a great debate about this here on VAF sometime back and all the guys who learned their flight mechanics at the airport coffee shop came out of the woodwork. My years of programming flight models in airline training simulators and matching it with flight data was no match for the coffeeshop who thought they saw something in a book once. It was painful and in the end I gave up.

Flying through a wind profile as you climb or descend or change your horizontal position impacts airspeed magnitude and direction i.e. your airspeed, aoa and sideslip can all be affected. Anyone who works in the field of simulation, navigation or autopilots or auto throttles understands this very well.

And this is another reason why a certain percentage of good landing is luck :)
 
Please send a link to this thread to the next chump who tries to tell you that winds don’t affect your airspeed. ....Anyone who works in the field of simulation, navigation or autopilots or auto throttles understands this very well.

Anyone with a Private Pilot certificate should know that wind affects your airspeed.
 
Peter,

I agree with your assessment of the stall signature and that it's lacking per se in the wind shear plot; thus I suspect your initial assessment of a down draft event is correct. The airplane likely did "quit flying" as I perceived; but likely due to rapid, momentary change in relative wind (airspeed). If you watch the landing roll out carefully, it actually starts flying again (you'll hear a decrease in AOA and see a momentary change in pitch as the tail rises slightly) before fully decelerating. I suspect due to a combination of ground and gust effects.

Yes, we are looking at integrating the inertial reference unit into the AOA (and sideslip) computations, but we are somewhat limited with our AHRS capability since we aren't using GPS or a magnometer (trying to keep it cheap and simple), thus we are limited to a six degree of freedom solution. Our first hurdle is automating calibration, and we are close. The distinct stall signature is a portion of that logic.

Scott,

Couldn't agree more, on both points. This landing was a good natural experiment. Nice to have the data and video to share.

v/r,

Vac
 
Back
Top