Aerial photography -- remote wing mount

[Peter Ralph]

yes I understand the concept of safety wiring - in fact I made the instructional vid used by the FAA. As long as you understand that without FAA inspection/approval what you are doing is illegal - sounds like you are good to go.

Will the camera be in some sort of box? 100 mile an hour winds are going to push particulate into every nook and cranny.

I would definitely advise not orienting the cam in the direction of travel - backwards would be best - but even an oblique angle will keep the lens clean for a lot longer. Using a filter will make any dirt on the lens a lot more noticeable. Using a lenshood/mattebox would help with contrast but I doubt it would hold up to wind shear without major mods.

best of luck

[Chuck Fadely]

From my experience shooting aerials with the XHA1, you definitely want the image stabilization OFF. Footage with it on is very "jumpy". It holds for half a second then leaps off to a different place.

[Vince Halushka]

Forget the plane.
Get one of these.
Or use a helmet cam inputting to a deck,

You have me curious what your filming like that.

[Tim Ribich]

Eric,

We've been experimenting with something like you are describing. Currently testing a strut mounted rig.

As I expect is the case with any aerial video, vibration and movements are the the biggest challenges. Out of the box, this particular system was very secure (and reinforced further by us) but as a result offered little isolation from the air frame.

Obviously the A1 with it's far superior manual control offers numerous advantages over the little Optura X1 we were using, but I'm not quite ready to hang an AI out there yet. [g]

One thing we're looking at was mentioned here:

Bob: If it is possible to add tightly wedged dense foam in around the camera body inside of the "U" section and tape it in, this might be worthwhile as it will further dampen case ...

We have some things to try, but if anyone has any sources for such material please post here.

I've been doing photography and aerial photography for years, but the video is relatively new. Much of the advice I've received is seemingly contradictory and I think simply the result of applying "normal" video principals to aerial video without actually having the benefit of first hand experience. Perhaps what appears as different advice is even more attributable to the fact that I'm still learning this stuff and don't fully understand the points being made.

For example:

Chris: 1. Yes, you'll want OIS on in this case, to dampen any vibration transmitted through the mount.

Chuck: From my experience shooting aerials with the XHA1, you definitely want the image stabilization OFF.

I've seen many discussions about IS for aerial stills-- from both "use it" and "lose it" perspectives. Personally, I can see little to no difference between the two.

Shutter speed:
Like yourself, I had to rethink my understanding of shutter speed to understand how it applies to video. Here too, there seems to be a difference of opinion: (?)

Trevor:
SLR Shutter speeds on your camcorder will result in a staccato movement on your movie. Choose your frame rate and stick with default.

Bob: If you want to try deshaking the footage in post a faster shutter speed helps. Without it the motion leaves odd trails behind when the footage is stabilized.

Perhaps both of the above are true? I definitely see the staccato movement Trevor refers to. Need to do some more experimenting obviously.

I put up some small thumbnail video of one of our first tests. Just send email to diffbeat AT gmail.com and the link will be in the "vacation message".

Pretty crude, but one small step.

[Eric Muehling]

Tim, thanks. Yes, I also noticed the contradictory opinions in this thread RE: image stabilization and shutter speed. I plan to use IS on, and 1/60 shutter, but I'll summarize all my settings in a final post along with a clip.

Also ... I plan to point the camera downward so the horizon isn't in the center. The terrain is my interest, and I don't want too much sky in the image influencing the exposure. I'm not sure what my downward angle will be (perhaps 15-20 degrees). I want a little sky, but not much.

[Bill Pryor]

I would definitely use the OS; it helps a lot with that type of vibration.

[Bob Hart]

I don't think anyone is invalid in their comments. All answers address specific issues.

Perhaps a look at things point-by-point. view these comments with a jaundiced eye as I am no expert.



THE AIRCRAFT.


Engine and related airframe vibration, a consequence of extracting power from a piston engine. Frequencies at level cruise assuming engine rpm of 2250, may be in the ballpark of :-

2250 divided by 60 for Hz for crank cycles and out-of-balance on prop.

4500 divided by 60 for Hz for two bladed prop and slipstream buffet on wheels, struts, inner wing, fuselage and tail surfaces. This is a hairy beast to assess as there is the tiime taken for the airflow to travel down the fuselage.

What hits the wheels and struts will be most evident and also co-incident and is the source of much of the signature sound of a particular aircraft type as it passes overhead.

1125 divided by 60 for Hz for variations in the power cycle due to stronger cylinder versus weaker cylinder.

There may be multiples of these induced in the airframe but good design attempts to eliminate them.

Close to the outboard strut to wing junction, this might be expected to appear as a tiny combined vertical and twisting movement across the image plane at the tripod mount, amplified by resonances in the camcorder structure which will convert them to angular deviations off the optical centre axis.

Provided the vibrational frequencies do not co-incide with or phase in and out through head drum rpm or camera frame rates, I would not expect much of a problem. Issues will likely manifest as pixellation dropouts (drum rpm or tape transport) and periodic line doubling artifact in interlace footage.


Airframe compliance.

All airframes bend and twist, all a part of the natural dynamics of flight. A camera mounted in a fixed upper position on a wingtip, viewing across to the other wingtip, would likely observe small bending movements of the wings during turbulence penetratration or heavy pitch movements, a desireable view if not re-assuring to nervous fliers.


Airframe movement.

The direction an aircraft is pointed constantly changes with control inputs and turbluence, all part of the natural dynamics of flight.


THE CAMERA.

Modern small consumer video camcorders tend to be structurally comprised of molded plastic casework with metal re-inforcements added where needed. They tend to have flex or compliance built-in.

This can confer some impact protection but can come at expense of movements of the optical axis relative to the baseplate area.

This compliance may permit rapid periodic movements of the optical axis relative to the baseplate if the camera structure moves in harmony with eingine vibrations.

The lens area of the camera represents a relatively heavy resistive mass and the tripod mounting point sits well behind it and well beyond its centre axis, therefore bending of the camera structure between the mount and the lens body can be expected.

The imager chip can be expected to be firmly attached relative to the lens focal plane but bending deviation of the lens centre axis relative to the optical centre of the imager chip is not uncommon. This causes the image to move laterally on the imager.

Lenses, especially zooms, contain moving groups of elements within. Rapid vibrations can be expected to move them. This might be destructive over time.


INTERACTIONS BETWEEN CAMERA AND AIRFRAME.

There is a phenonema known as flutter which can destructively affect an aircraft. This is defined as the ability of a particular part of the airframe being able to move and extract energy from the passing airflow.

In normal R & D, this is vigorously tested for and eliminated as far as is possible.

The camera represents an object which does not pass smoothly through the airflow. It is possible for the combination of the camera mass, its alteration of airflow over the aircraft and flex within the camera structure itself or the effects of its mass in causing flex of a part of the airframe, to extract destructive erergy from the passing airflow especially if the camera becomes partially loosened by mechanical vibration.

Locking wires do not necessary confer protection. They can prevent a fastener from coming off but cannot maintain tension if the structures are altered by vibrational wear on contact surfaces. Plastic is a poor performer in this respect.

The onset of destrructive movement would be rapid with little or no warning.


RELEVENCE OF SUGGESTIONS.

Engine and related airframe vibration. - Optical image stablisation cannot respond in time. Digital image stabilisation is unlikely to. Camera shutter speed set higher is unlikely to assist.

Solutions - mount camera firmly with lens area of camera supported. Set lens field-of-view to its widest.

Airframe compliance. - Optical image stabilisation and digital image stablisation cannot respond in time. Additionally, much of the movement can be expected to be a minute rollling movement which image stablisation is not intended to deal with.

For rapid repetiion, high camera shutter speeds are unlikely to assist. For slower movements such as wing flex in across the wing views, a higher camera shutter speed will assist apparent image clarity.

Solutions - mount camera firmly with lens of camera supported. Set lens fied-of-view to its widest.

Airframe movement. - These motions are normally fairly slow and limited. They are a natural to the flight experience and should be faithfully reproduced, epsecially if part of the airframe is in the shot.

When the aircraft is maintained on a constant heading and the view directly forward or rearward, optical and digital image stablisation will most definitely help, but during course or attitudinal changes, a periodic stutter of the image will be introduced whiich will not be desirable.

A higher shutter speed will assist apparent image clarity but may cause an undesirable stutter effect at corners where human peripheral image motion blur might normally be apparent.

I would favour the higher shutter speed, say 1/100th to 1/150th sec but no higher unless a sharply defined strobe effect on the propeller blades is wanted.

If the view is directly to the side, optical image stabilisation may be beneficial but digital image stabilisation may cause stutter in the image.

Solutions - If the airframe is to be in shot, optical and digital image stabilisation should be off, as unnatural movements of the visible airframe may be introduced if it is a very insignificant part of the image. At worst however it will appear as if the camera has been hand-held.

If no portion of the airframe is to be in the shot and direction of view is to be forward or rearward, then image stabilisation may be benefical.

Set lens field-of-view to its widest.



ADDITIONAL NOTES.

Rolling shutter. - If the cam is a CMOS imager type, the rolling shutter issue will become evident in the propellor disk or any shadows it casts.

at very low altitudes, say 100ft, image shear in the lower portion of the forwards and rearwards views may occur and downward views of the flight shadow on the passing terrain may be unusable.

Automatic exposure controls. - If automatic exposure controls are chosen, set the response times to their slowest to avoid the camera attempting to rapidly hunt for best exposure levels whilst the propellor shadow falls across the lens.

Lens controls. Select manual focus. My personal preference is to cycle the zoom all the way to the tele and then back to the wide view, to cycle the manual focus to close-up then back to infinity. Hopefully this locates the moving lens groups into positions hard against end-stops where maybe a bit of grease will dampen movements due to high-frequency vibration.


Finally, except for aviators and the techically inclined, in-flight and ground-to-air vision without humans interacting becomes very boring.

[Jack Walker]

What effect would there be by using a fast shutter speed then slowing the video down by 50% or so in the editor.

Unless there were action in the frame, wouldn't this help to smooth the vibration in the video and also eliminate the jerky look from a fast shutter? I don't know, just asking.

Also, when they shoot from one plane to another doing aerobatics, how is the camera held... just handheld by someone sitting in an open cockpit?

[Bob Hart]

For the technical interest of those who take cameras into the air, a short clip derived from experiments with a compact home-made stabiliser, a sort of much shortened version of a steadicam. There was no float in the lateral twisting plane only in yaw and pitch movements.

www.brightcove.com/channel.jsp?channel=570698206

[Steve Yager]

Hey Eric. No one has answered your question about shutter speed yet. No, it's not the same as photo. The normal shutter speed for video is double that of your frame rate. So, if you're shooting in 24f, then a normal shutter speed will be 1/48. It won't be blurry, video is much different than taking stills. If you're shooting 60i, then shoot 1/120. 24f will look more like film and 60i will look much more like video.

[Tim Ribich]

Wondering what would be an example of an application where high shutter speed in video (in excess of the ratio above) would be utilized?

Obviously with still photography high shutter speeds are used to either freeze moving subjects or to allow for a wider aperture and a desired depth of field effect.

[Steve Yager]

Originally Posted by Tim Ribich

Wondering what would be an example of an application where high shutter speed in video (in excess of the ratio above) would be utilized?

Obviously with still photography high shutter speeds are used to either freeze moving subjects or to allow for a wider aperture and a desired depth of field effect.

It's been called the gladiator effect. Essentially if you have a super fast shutter speed it will make all the motion overly crisp and give you the "gladiator effect". There is no motion blur. So, fight scenes, action scenes, intense emotional scenes could all use this effect when appropriate. Be careful though, its getting a little cliche.

[Don Jones]

As a pilot and user of the A-1 airborne, I would suggest flying your shoot either early in the morning or late in the afternoon when the wind is more likely to be calm. If you fly and video any other time while the air is bumpy, you will be wasting your time. This camera does not like to be moved. I have used 24fps with a 48 shutter speed with good results. Props spin and don't show any stoppage. As far as hanging your pride and joy outside an airplane, I never would except in a closed container of some sort. Get your pilot friend to take you up and try to hold the cam outside the cockpit and get a feel for the slipstream at the speeds you intend to fly. I bet that you will call this off after you sample the forces that your camera is about to be subjected to. You may prove me wrong and if you make the attempt, I surely hope that I am.

Best of luck,

Don

[Bob Grant]

Just to explain a little further about shutter speeds and this only applies if doing image stabilization in post, if your camera is mounted on gyros you can ignore this.
If the plane / camera makes any unwanted movement that you'd like to remove and there's plenty of good software to do this ranging in price from free upwards then the standard shutter speed will impart a blur to the image in direction of travel. This in the original footage appears normal and at slow frame rates helps mask the perceived stutter, faster shutter speeds expose the stutter from shooting a low number of frames per second. At times fast shutter speeds and films low fps are used deliberately to impart a certain 'feel' to the footage. This isn't the intent here though.

Consider a simple unwanted vertical oscillation of the aircraft, could be from buffeting or rotor wash in a helicopter. Camera moves up and down, each frame will have matching up and down blur. Fine so far.

Now we use our stabilization software to remove the vertical movement. Unfortunately each frame still has the blur and watching the sequence of frames as per normal we see a blur trail that moves up and down. Shooting with a faster shutter speed will avoid this problem but at say 24fps certainly create stuttery footage if there's fast motion involved. However it all depends on what you're shooting here. If it's dog fights between two planes then you probably don't want to remove any of the unwanted camera movement anyway, you'd endup with a pretty unnatural look to the shot, your audience knows if the camera is in a fighter plane being flung around the sky it's not going to be a silky smooth ride. But if you're shooting aerial landscapes then usually the wanted motion of the camera is quite low and the impact of faster shutter speeds become less relevant.

In my seat of my pants, Z1 out the window of a Longranger, shooting at 100th of a second 50i there's no sign of stutter. I was shooting as wide as I could and we were at 500' so in the shots of us approaching the city motion is very slow, the 180deg turn over the bridge and Opera House are also pretty slow. When I did try to get tighter shots of places of interest the footage just became useless anyway although very fast shutter speeds might have allowed me to extract some stills however it was a very overcast day and there just wasn't enough light to really push the shutter speed up to 1/1000th.

[James R. Wilson Sr.]

Dear Eric,

You'll have no problem with airframe harmonics. If your wing and struts are vibrating noticeably you got much bigger problems than shaky video. The folks who warned of legal/FAA liability were right on the money. You make a modification to an airframe, which is what you're doing, the aircraft's insurance underwriters will have a field day if anything would happen. In fact, they'll have a field day if something happens that isn't remotely related to mounting the camera to the airframe. Lot's of people do it, but you should know the risks and so should the pilot/owner. I spend about 300 hours a year shooting from everything form helicopters to F-16's, still and video.

My guess is you're going to proceed with your shoot and hope you've covered everything. I understand. Just make very sure your camera mount doesn't compromise the aircraft structure or control cables. In the Cub the aileron cables run up the trailing edge of the rear of the rear streamlined strut. Your camera should be mounted as close to a hard attach point as possible, meaning where the struts attach to the fuselage (unlikely that will get the perspective you need) or the top attach point at the wing. While aircraft structures are over built, you can't assume that hanging a 5 lb camera plus mount, placing a 60 to 100 mph wind against it's profile and then pulling G's, changes the equation considerably. With a fairly mild dive and pull it's fairly easy to put 2 G's on the airframe. That means your camera and mount weight doubles, plus wind force, so you've got 12 to 18 lbs hanging on that streamlined strut. One possible solution to some of your weight problem is use an HV-20 instead.

I hope this helps, sure don't want to be the bearer of bad news, but I've done all this 50 to 60 times annually for the last 26 years. A fun adventure can turn into a tragedy in the blink of an eye.

Sincerely,

Jim Wilson
www.Jimwilsonphotography.com