Several months ago I picked up a Gaui 500X Quad Flyer kit.  I wanted an inexpensive platform for learning FPV (first person view) flying and light aerial photography.  I have several larger multicopters for professional photography.  But I wanted a small practice aircraft that would be cheap to fix.  I didn’t want to be afraid to crash it, while improving my piloting skills.  This article takes a look at the Gaui 500X for this purpose.

Gaui 500X First Impressions

When I first opened the box, I was impressed with the engineering quality of the whole package.  For example, the ESC’s are amazingly tiny.  They fit inside the quad’s arms, so both wires and ESC’s are hidden.   Very nice.  The wiring harness is an extremely small and simple design, constructed from bullet connectors soldered together.  Reliable.  There’s a very light weight blade guard to prevent blade strikes should the copter bump into something.  The arms of the quad are foldable, so you can pack the whole thing in a small canvas bag (included in some kits).  And there’s an optional lighting kit with LED’s for each motor mount.  This makes day flying easier and night flying possible.

Good Motors, Good Electronics

In case you’re wondering whether there might be overheating problems with internally mounted ESC’s, the motors and ESC’s run warm, but never hot.  Even with heavy loads (over 1600 grams but under the 2200 gram limit), I’ve never had overheating (unlike my larger multicopters).    The Scorpion motors are small, but efficient and ample for the quad plus a small camera.

Gaui 500X Motors and Electronics in an AGL Hobbies frame

 A Battle Hardened Frame

Though I like the Gaui frame, which looks quite sleek, I wanted a frame that could take a real beating.  So, I put the Gaui electronics and motors in a “Rusty Rev. 7″ frame (pictured above).   This frame was from AGLHobbies.  The Rev 7 frame has been discontinued, but the newer models are both better and cheaper.  I highly recommend these frames as a rare combination of low price and high quality.  My Rev. 7 once crashed after the quad lost power from 15 feet above my driveway.  It hit hard.  But the damage was only two bent arms (Align Trex 450 tail booms), two broken props, and a bent motor shaft.  Under $25 in repair costs for such a hard fall.  None of Rusty’s G-10 fiberglass or delrin parts broke, or even cracked.  Amazing.

Assembly

Above is a picture of the Gaui motors, electronics and wiring harness mounted in the Rev. 7 frame.

It took under two hours to assemble the Gaui parts in the Rev. 7 frame.   The complete quad copter, without the battery and without the camera, weighs a total of 988 grams.  This is well under the 2200 gram all-up limit suggested by Gaui.  It leaves about 1200 grams for battery, camera and camera mount (if used).  The GoPro camera is about 200 grams.  Batteries range quite a bit, but I use a 3S, 5000 mah, 30C battery which weighs about 400 grams.  It provides roughly 12 minutes of flight time with the camera mounted.  With an all-up weight of 1600 grams, it’s still 600 grams under Gaui’s stated lifting capacity.

Some Minor Issues

Nothing is perfect.  Here are a few minor issues I found with the Gaui hardware.

First, the prop mount is a threaded brass bushing, which is very thin. If you over tighten a propeller, the bushing will fracture.  It’s unusable when this happens.  As a quick fix, an Eflite part happens to fit the Gaui motor.  It’s part “EFLM1922 3mm Prop Adapter with Collet.”  Most hobby stores stock the Eflite part, though they may not stock the Gaui part.

Broken Gaui 500X Prop Mount

However, it’s easy to avoid snapping the prop mount.  When you tighten the propeller nut, hold the motor casing (it’s an outrunner) with your index finger and thumb ONLY.  Don’t grab the motor with your whole hand.  As you tighten the prop nut with a wrench, no matter how hard you squeeze, at some point the motor casing will slip between your fingers.  This prevents over tightening.   You may not think you’ve tightened the prop nut enough. But the prop mount grips the propeller more snugly than you may think.  I’ve never had one loosen using this method.  Nor have I broken any more prop mounts.

Safe Tightening of Prop Mount

The second minor issue is with the GU-344 Stabilizer.  This is the electronic box that contains the gyros, accelerometers and computer.  Generally, it works quite well once adjusted properly.  However, there is a trim pot on the unit that allows you to adjust the gyro gain with a screwdriver.  If the gain is too low, the quad will wobble like a drunken sailor.  If the gain is too high, the quad will oscillate a bit.  Getting the right spot is tricky because very small movements of the pot can make surprisingly large changes in gain.  Homing in on the best spot is difficult.  Don’t panic or give up.  You’ll get it with a bit of trial and error.  It took me at least 6 tries.

The last issue I noticed was with one of the 4 motors that came with my quad.  Because I was planning to shoot some video, I checked for motor vibrations with an iPhone app called “Seismometer.”  I’d strap the iPhone to an arm near the motor in question, turn on Seismometer and power up the motor.  Seismometer graphs the vibration.  It’s not as good as a vibration meter, but it does give you a rough idea of how smoothly your motors (and props) are spinning.  Using this method you can see which motors are well balanced and which are not. 3 of my Scorpion motors were well balanced and one was way out of balance.  I replaced this motor with one that was better balanced.

The Bottom Line

It’s been about 4 months since I built the Gaui 500 with the custom frame.  During that time, the hardware and electronics have performed flawlessly.  There have been crashes, of course.  But the damage is usually no more than a broken prop.  I’ve flown with both 3s and 4s batteries.  I’ve been amazed at how much weight the motors and ESC’s can lift (tried up to the 2200 gram limit) without signs of over heating.

So, if you’re looking for a small multicopter for FPV or photography with a light camera, the GAUI 500 is well worth considering.

Here’s a brief video showing how stable the Gaui is in a hover.

Click here to view the embedded video.

This is part 1 of a series of posts describing what I’ve found that works best and how you can build your own aerial platform (or what to buy if you want someone else to build it). This information can save you a year of trial-and-error experiments. Better, you can go straight to a solution that will work for you without wasting a lot of time or money.

First, here’s a picture of 6 rotor helicopter, a hexacopter, with a Skyrover camera mount and camera. I finished this one recently and it gives the best results for stills and videos that I’ve been able to get so far. In future posts, I plan to take you through every step of the build process. I’ll explain my choices and the reasons for them.

Hexacopter with Skyrover camera mount, MK electronics and D90

Here’s a view from the top.

Hexacopter with Skyrover camera mount and D90, top view

I’ve found that multicopter configurations with 4, 6, 8 or more rotors (propellers) to be the smoothest, most stable and easiest to fly. On these multicopters the rotor has a fixed pitch. So the only way to cause the multicopter to roll, pitch or yaw is by changing the speed of the motors relative to each other. Translating the joystick movements on a remote control into speed variations of the motors is done by an on-board computer. How stable the multicopter is in flight (especially in the wind) is determined largely by the computer and the quality of its software.

I’ve found that the computers made by www.Mikrokopter.de (sold in the US by www.Mikrokopter.us and by www.quadrocopter.us) to be particularly good and probably among the best available. I’ll discuss the reasons why I think they are a good choice below.

Another contributing factor to stability and smoothness in flight seems to be the number of rotors. Having looked carefully at dozens of videos, I’ve noticed that even with the best multicopters, there’s a little bit of jerkiness, or low amplitude, short duration roll movements. I call these micro-rolls. Usually, they appear as tiny and brief (barely noticeable) jerks rolling to the right or left. These movements are most often caused by sudden stick movements by the pilot. They are also caused by wind. They are subtle and most viewers don’t notice them. But professional videographers notice them. So if you have discriminating clients, you’ll want the smoothest copter you can get.

When comparing videos, my subjective judgement is that octocopters (8 rotors) show the least of these micro-rolls. If carefully flown, you can get very smooth movement with an octocopter sometimes with no visible micro-rolls in the video. It looks like the octocopter is on a pole, not flying through the air. Hexacopters can be reasonably clean too, but they seem to be a bit jerkier. Quadcopters are usually a bit worse. The reason octocopters are immune to these movements is that there is more mass at the ends of the arms. And the arms on octocopters tend to be longer to make room for the propellers. So it takes a lot more energy to roll or pitch an octocopter.

You might think that a properly stabilized camera mount would absorb micro-rolls. I haven’t seen one yet that removes them completely. Please let me know if you find one. The camera mounts tend to be better at removing slow, large amplitude movements. So, my recommendation is that if you’re building a multicopter for aerial photography, go with a 6 or preferably an 8 rotor design.

An actively stabilized camera mount can keep your camera pointed where you want it, even when the multicopter is pitching and rolling in the wind. The camera mount should also absorb vibrations caused by the motors and rotors. The Skyrover camera mount (available at MikroKopter.us) is an inexpensive mount with active stabilization for pitch and roll. It is designed to work with the Mikrokopter computer and software.

The Skyrover doesn’t have gyros or accelerometers itself for stabilization. It comes only with servos. It relies on the Mikrokopter computer (or similar) to provide the right signals to the servos. While this is a good arrangement and works reasonably well, it is possible to improve it. The main reason it’s not perfect is that the gyros, which sense motion, are on the Mikrokopter computer board. So they’re not actually sensing the camera’s motion directly. The camera is not rigidly attached to the multicopter’s frame. There are soft rubber shock absorbers connecting the camera mount to the copter. So a better arrangement would be to have the gyros mounted on the camera mount, so they are sensing the camera’s movement directly. You can set up the Skyrover to work like this (there’s a gyro+computer product called PicLoc you can use for this), but you need to provide the extra gyros and hook them up yourself. Not difficult, but you need to decide whether the effort is worth the slight improvement.

But even with the standard arrangement of the Skyrover and Mikrokopter, the Skyrover is amazing at absorbing vibrations and stabilizing the camera. I’ve shot stills at dusk at 1/125 second and got results that were consistently tack sharp. Can’t do better even on a tripod.

For video, the results are also quite good. The Skyrover mount compensates for helicopter movement quickly, smoothly (important for video) and with lots of torque for heavier cameras. The current version of the mount will accomodate a camera body that is 5.75 inches wide. This is wide enough for most small and medium size DSLR’s. Though it’s not quite wide enough for a Canon 5DII. My understanding is that a wider carriage version of the mount will be available to handle larger cameras (Spring 2011 time frame).

Equally important, the Skyrover is mostly made of aluminum. It can withstand hard landings and outright crashes, often protecting the camera from damage. At 600 grams, you can lift it easily with an octocopter or hexacopter suitably powered (we’ll explain what this means later as well). I prefer aluminum to carbon fiber. Carbon fiber can shatter in a crash. Aluminum bends. So, you can bend the aluminum back and (hopefully) keep flying.

To be continued…

What’s It Really Take?

Back in November 2009, I posted an article about choosing (radio controlled) helicopters for aerial photography. At the time, I was looking for the cheapest way to get professional quality results for both still and video. Back then,  you’d need about $3k worth of helicopter (or more). You’d also need to hire a pilot, or spend a year or more likely two years learning to fly it yourself.

Well, all that’s changed. There are new helicopter designs with ever more sophisticated electronics. The new designs are cheaper, more stable, easier to fly, stay in the air longer, and are safer and easier to repair than “conventional” helicopters. And as digital cameras get smaller and lighter, the loads placed on these helicopters are reduced. This makes it possible to use smaller, less powerful helicopters to get the job done. Sound good? Here are the details…

It’s All In The Computer

As I was building a conventional RC helicopter (an Align TRex 600 ESP) for aerial photography, I was also surfing the web to learn as much as I could about alternative ways to get cameras in the air.  One day, I discovered quadcopters (aka quadrocopters).  A quadcopter is a type of hovering aircraft that has 4 propellers and 4 motors to drive them.  The propellers are often simply bolted onto the motors.  There’s no variable pitch as with a conventional helicopter.  There often aren’t even any gears or servos in a quadcopter.  It’s just 4 motors and a computer to control them.  Very simple, very reliable.

A Good Practice Quadcopter

The first quadcopter I encountered was called a Draganflyer V.  I quickly snapped up a used one on eBay for about $150.  It’s not really powerful enough to lift a camera, but it’s great for learning to fly.  Here’s a picture.

DraganFlyer V Electric Quadcopter

Because it is small and light (475 grams without battery), the DraganFlyer is great for practicing in small spaces, even indoors.  It doesn’t usually break when you crash it.  And when it does break, it’s very cheap and easy to fix (You can get parts from rctoys.com).  It’s very stable, but not so stable that it’s too easy to fly. It will definitely help you develop some skill. But it’s way easier to fly than a collective pitch (conventional) helicopter.  The propellers are made of soft plastic and geared down (unlike most quads) so they are not spinning fast enough to hurt you if an accident happens.

The Learning Curve for Quadcopters vs. Collective Pitch Helicopters

My other practice helicopter is an E-Flite Blade 400.  It’s a conventional, collective pitch helicopter.   It’s a b***h to fly and always hours of work after a crash.  You can make some hardware and software (in the transmitter) changes to make a Blade 400 more stable and easier to fly.  But, and this is a big but, a collective pitch helicopter will never be as stable as a quadcopter.  So, quads are inherently better for photography (and beginning flyers).

Here’s the kicker.  After about 6 months of flying both quadcopters and collective pitch helicopters, I can fly a quadcopter with confidence.  This includes both hovering and flying various patterns.  Nothing fancy, but competent basic flying.  In the same amount of time, I’m still using training gear on my Blade 400.  I can hover it and move it around a little, but I don’t feel that I’ve got it fully under control. I find it very twitchy and difficult to land.  I would NOT risk a camera on it.

On the other hand, I had no hesitation strapping a Nikon D90 with a 20mm lens on my quadcopter and test flying it.  Friends thought I was crazy.  But that’s how predictably a quad flies.

Most expert flyers say it takes the average adult about 2 years to learn to fly a collective pitch helicopter with that level of confidence.  That sounds about right to me.  It only takes a few months to master a quadcopter.  That alone would be a good enough reason to use it for  aerial photography.

A Serious Quadcopter

It didn’t take me long to figure out that my time might be better spent building a quadcopter with the same lifting capacity as my TRex 600.  Around that time, I discovered the aeroquad.com site.  Aeroquad refers to a quadcopter that uses a particular computer and software for stabilization.  Here’s some pictures of a quadcopter I built using the Aeroquad computer.  (This was NOT a kit, but a home brewed design.)

Quadcopter using the Aeroquad computer and a home-brewed airframe.

Quadcopter from above with GoPro Hero HD camera mounted on leg.

As you might be able to tell, the airframe of my quad is made from aluminum tubing (actually, TRex 600 tail booms), pvc elbows and other pipe fittings from Home Depot, a plastic project box from Radio Shack, pipe insulation, nylon wire ties, plywood, wire, switches, motors, and connectors from various sources, and even practice golf balls.   The quadcopter weighs about 1.7 kilograms without a battery or camera.  The distance between motors on the same axis is 32 inches.  The total cost to build this quad was about $400 minus the radio gear.   That’s roughly $200 for the Aeroquad computer and sensors and $200 for the motors, propellars and airframe.  Radio gear varies from around $50 for a transmitter and receiver to around $300.

As it happens, the motors I used (Turnigy 2217, 20 turn 860Kv) are a bit small to match the lifting capacity of the TRex 600.   But the difference in the total cost with more powerful motors (and speed controllers) is less than $50.  I didn’t have enough experience to know that four 220 watt motors would not quite be enough power to lift a dslr and camera mount.  Something around 350 watts is more like it. Here’s a brief video demonstrating the quadcopter with a GoPro HD Video camera on board and a primitive camera mount.

Not Quite Ready for Prime Time

So what’s not to like?  There are some issues with the current release (1.71) of the Aeroquad software.  The Aeroquad computer and software that I’m using has 3 gyros to stabilitze the helicopter.  I also put 3 accelerometers on my computer,  but the software doesn’t currently use the accelerometers effectively.  The accelerometers could be used to provide a much higher degree of stability than you get with just the gyros.

In the Aeroquad software are experimental versions of heading hold, auto leveling and a high stability mode.  I have tried these features and in my opinion they are too buggy to be useful in the current release.    Since the aeroquad team is working to correct the flaws, I chose not to modify the code myself.  Fortunately, the basic gyro controls work extremely well, so the quad is easy to fly at moderate altitudes.   The heading hold, auto leveling and high stability features of the software are likely to be fixed in a near future release.   With free software, you must be willing to wait, or go in and fix it yourself.  I’ll keep you posted on progress in this area.

What’s Required For Aerial Photography?

To use a quadcopter for photography, you often need to fly at high altitudes, where you can barely see the quad.   But when a quadcopter is far away from you, it can get very difficult to tell whether it’s right side up or up side down.  It’s just a little + in the distance.  I crashed the Draganflyer on more than one occasion because I had a split second of confusion about its orientation.   That’s where an auto leveling feature is essential.  You want the quad to right itself when you either push a button or just let go of the controls. So whether you can see it or not, you know it’s flying straight and level.

Currently, the Aeroquad software does not stay level if you take your hands off the controls.  It will drift and eventually crash.  I should mention that the same is true of a conventional helicopter, except that the conventional helicopter will crash a lot faster.   So, an auto leveling feature is a MUST.

What’s Nice To Have

In the ‘nice to have’ category is the ability to set the position and altitude and have the helicopter stay hovering in place while you let go of the controls and operate the camera (also by remote control).  You need a barometer and a GPS for these two features, and the software to use them.

Will the Aeroquad platform have all these features in the near future?  I think probably yes.  The Aeroquad group recently merged with a group at DIYDrones.com. The folks at DIYDrones have an autopilot system for planes with all of the features I’ve mentioned and much more.  They are currently working with the Aeroquad folks to provide all these features for quadcopters and even conventional helicopters.  I expect a release within the next few months.  This is especially exciting because this will be open source, public domain software and non-proprietary hardware.  The benefits of this approach, not to mention the low cost, will be huge.  Stay tuned.

What If I Need A CameraShip Now?

Does having all these stability features sound like a lot to ask?  Well, there are systems that have auto leveling, GPS position hold and altitude hold.  You push a button and the helicopter stays in position within the accuracy of the GPS unit (a few meters) and the altimeter.  You can buy them ready made, or as kits.  Or you can just buy the flight control computer (see below) and build your own helicopter around it.  The fully assembled and ready to fly versions of these systems are very expensive (often around $10,000).  Manufacturers of fully assembled units include www.microdrones.com and www.draganfly.com.  Video taken from these helicopters is quite amazing.  Check out their sites.

But there is a vendor who sells a very high quality system for a moderate price.  We’ll talk about this system next.

MikroKopter quadcopters, hexacopters and octocopters

MikroKopter.de is a German company that makes quadcopters, hexacopters and octocopters (spelled with k’s in German).  The US website can be found at mikrokopter.us.  It is the brainchild of Holger Buss and Ingo Busker.  What got my attention is this amazing video where Holger demonstrates the stability and stunning capabilities of his hexacopter (6 motors). By the way, the unit he’s demonstrating can be purchased in kit form complete for about $1500.

In the video Holger uses a light weight point and shoot for demonstration, though the hexacopter can lift about a 1 kilogram payload.   Notice that throughout the demonstration, there’s always a camera mounted underneath flight controller (circular hub in the middle). You may also notice that the airframe of all the MikroKopters is very thin and light.  If you want to build a MikroKopter to lift a heavier, professional camera, you need to buy or build a stronger airframe, buy bigger motors, and use the MikroKopter computer and motor controllers in it.  I may try this.  If I do, I’ll post the results.

The cost of building a MikroKopter from a kit or using their computer ranges from about $600 US to about $2000 US depending on which model you buy and whether you provide your own airframe and motors.

MikroKopter Videos

A fellow New York photographer, Anthony Jacobs, has gotten a lot of press with his quadcopter, which is built around the MikroKopter computer.   Here’s Jacobs’ video page where you can see how stable his quadcopter is.  Nice work.

Make Your Conventional Helicopter Stable

There are several devices you can use to tame your conventional helicopter and make it more stable.   One is the Copilot II Flight Stabilizer from FMADirect.com.  Another is the Helicommand from Captron.

The Copilot II and Helicommand use completely different approaches to stabilization.  The Copilot II uses 6 thermal sensors to orient the helicopter.  The Helicommand uses a 3 axis gyro and a low res video camera.

To determine the horizon and orient the helicopter, Copilot II uses 4 sensors that look N,S,E,W, one sensor looks up and another looks straight down.  If there are no trees or buildings close to the helicopter, then Copilot II can accurately determine what’s up and down and where the horizon is.   When you let go of the controls, it can quickly right an inverted helicopter.  It can also hold the helicopter in a hover.   But, it can only maintain the helicopter in a specific orientation.  It can’t prevent drift.  So, if there’s wind, the helicopter will not stay in one place without human control.

The Helicommand, on the other hand, uses a 3-axis gyro to maintain orientation.  It also has a small video camera, pointed downward that “looks at” the ground to detect and correct for drift.  So, if there’s enough texture to the ground, the helicommand can hover in one place without drift.  Another advantage of the helicommand, is that it isn’t affected by nearby trees or buildings.  So you can use it anywhere without worrying that it’s going to mistake a tree for the horizon and flip your helicopter over.

I haven’t tested the Helicommand myself.  A short review of it can be found here.

The Copilot and the Quad

I have a Copilot II that I’m currently using on a Blade 400.  It does a very good job of stabilizing the heli when there’s no trees or houses nearby.  I plan to reinstall the Copilot on my quadcopter, to see if that helps the quad stay level when it’s a dot in the sky.  I’ll let you know how it works.

Conclusion

We’ve discussed some great options for getting an easy-enough-to-fly helicopter or quadcopter for aerial photography.

• One option is to use Aeroquad or Arducopter computers and software with a home built or kit airframe.  The estimated cost of such a project is around $500 for the quadcopter.  This option depends on software that has yet to be released, but should be available by Fall, 2010.
• Another option is to build your own airframe and use a MikroKopter computer.  This option will probably cost around $800-$1000 US depending on lifting capacity.
• Another attractive option, if you’ve got deeper pockets, is to buy a Mikrokopter kit.  They range in price from about $1200-$2000 US.  Their airframes are light so you can get flight times of around 30 minutes with a light load.

Post a comment to let us know how your project went.

This is a story that I find inspiring.  Since it’s also about photography and space, I find it even more inspiring. It’s about two MIT students who wanted to photograph the “edge of space.”

It started when Oliver Yeh had a vision. He wanted to see the curvature of the earth and the blackness of space from high up in the stratosphere.  Many of his friends thought he was crazy.  Not Justin Lee. Lee accepted the vision and made it his own.  So together they set out to accomplish the task.

The Ingenuity

Like others before them, Yeh and Lee decided to use a weather balloon  filled with helium  to lift a small camera up into the stratosphere.  They bought a 300 gram latex balloon online.  Balloons are capable of reaching altitudes of 20 miles or more. Unlike others before them, they didn’t have thousands of dollars to spend on custom electronics.  Instead, they solved difficult technical problems with inexpensive off-the-shelf solutions.  Here’s how.

From their research, they learned that temperatures in the stratosphere can reach down to -55 degrees Centigrade.  They couldn’t get their freezer to go any colder than -10 degrees Centigrade.  So they couldn’t test whether their point and shoot camera (a Canon A470) and its battery could operate at such a cold temperature.  Batteries stop working and electronics fail when they get cold enough.

What did they do?  They decided to use pocket hand warmers (the kind skiers use) to warm the camera and its battery.  The hand warmers were taped tightly against the electronic devices and batteries. They also put everything (electronic devices and hand warmers) into a styrofoam beer cooler, which became their “spacecraft.”  The hand warmers worked so well,  they were still warm when they recovered the craft after a 5 hour flight.

How did they trigger the camera?  They found a firmware hack, CHDK, that included an intervalometer.  With it, the camera could be programmed to take a picture every 5 seconds.

Another difficult problem for Yeh and Lee was how to recover the craft.  They knew the balloon would burst when it got high enough, since the weight of the atmosphere is no longer compressing the balloon.  So, they created a simple parachute from  a plastic bag.  They tested the parachute by dropping the spacecraft and parachute off a 5 story building with an egg inside.  If the egg didn’t break, then the landing was soft enough for the equipment to survive.

The most difficult problem was how to track the craft so they could find it when it came down. From their research, they discovered that the wind could drag it many miles away from the launch site.  Here’s where their ingenuity really shined.  Neither Yeh nor Lee had any electronics background.  What did they do?  They went to Radio Shack and bought a $50 cell phone with GPS. The cell phone was taped to the camera and constantly reported its GPS location via text messages.  It could also be tracked with Google Earth.

The Launch

On the day of the launch, things did not go exactly as planned. The balloon was launched on Sept. 2, 2009 at 11:45 am. They had previously checked with the University of Wyoming’s balloon trajectory website to estimate the landing site.  So, they launched from Sturbridge, Massachusetts (central Massachusetts) to make sure the balloon didn’t come down in a city or in the Atlantic. Fortunately, the winds were light.

Yeh and Lee remained at the launch site for 4 hours after the launch. During that time, they lost contact with the craft. As time progressed, they wondered: had the phone’s battery died?  Had it froze? Did the craft crash? … They thought of everything that could possibly have gone wrong.

With those thoughts, they headed back home, believing that they had lost the craft.  When they got to Lee’s apartment and checked his computer (presumably Google Earth’s cell tracking service), they found that a signal had been sent by the cell phone before it hit the ground.  The camera had come down 25 miles from the launch site and landed in a construction site near Worcester.  “We were so excited, we jumped right back into the car, and we drove out to Worcester, and we found it. That was a great moment,” Lee said.

The Results

They calculated (from the balloon’s ascent rate and time aloft) that their craft had reached an altitude of about 98,000 feet (over 18 miles).  It was so high that it took 40 minutes for it to come down by parachute.  The pictures are stunning.  Here’s the one most often published.

Oliver Yeh and Justin Lee's photograph from 98,000 feet

You can also see a video of the entire flight on their site, http://space.1337arts.com

The Cost

What is amazing about this story is that the total cost of their spacecraft was $148. You can check out their list of items and costs on their site, http://space.1337arts.com/hardware.

The Possibilities

Why am I re-telling this story?  Because it opens up incredible possibilities that were always there, but we just didn’t see them. In Oliver Yeh’s words,

“The fact that we were able to accomplish space photography on such a low budget and with minimal electronic modifications proves that it’s really possible for anyone—anyone at all—to do. Imagine how many students might be inspired if their high school science teacher took the time to give his students an out-of-this-world experience.”

What other possibilities does this suggest? Here’s just a few that come to mind.

• On similar craft, put a self-addressed, pre-paid Fedex label on the box, so that whoever finds it can just drop it in the mail.
• Use a small transmitter to radio the images and data back to you.  That way, all is not lost if the craft disappears.
• Instead of a styrofoam box, use a styrofoam glider with an electronic autopilot and GPS to bring the craft and camera back to the launch site.  This is a difficult project but not as expensive as you may think.  DiyDrones.com sells a computer with software that they call an “ArduPilot.”  A complete autopilot with GPS can be built from their parts for around $300.
• Instead of letting the balloon rise until it pops, adjust the volume of the balloon and mass of the craft so that it floats at a specified altitude for a while.  You’ll get a lot more pictures.  You may then need to find a way to pop the balloon (a timer and a servo maybe) to get it back.
• Launch a rocket from the balloon to get even more altitude.
• Add a servo to control the camera’s attitude so you can get shots looking straight down as well as out to the side of the craft.

This project also opened up new possibilites for Oliver Yeh and Justin Lee.  They are now known as aerial photographers.  They helped photograph the BP oil spill in the Gulf of Mexico using balloons and kites.  Since our government is placing restrictions on flight over the affected area, getting photos of the devastation is difficult.  Balloons, kites and RC helicopters provide ways of getting around the restrictions and recording the environmental impact for public awareness.

Ask Yourself This

Are there any photographic or scientific projects you’ve always wanted to do but assumed you couldn’t?  It’s well worth taking the time to let go of assumptions about what is possible, and really look with an open mind for creative solutions.   As my mentor and friend Harold Feinstein says, “The problem is everything you know that isn’t so.”

Let your imagination soar!

News Coverage

Here’s the major coverage I found with a Google search.

CNN Story and Video

iReport

Wired

ABC News Video

Fox News Video

With each release of Photoshop, new tools and filters are added, as you would expect from a product that is continually being refined and improved.   Sometimes the consequences of this growth are changes to the user interface that actually make Photoshop a little harder to use for those of us who were used to the previous version.

For example, in CS3 the tool palette has 24 buttons for 59 tools.  That’s a lot of tools that aren’t in plain view.  So if you are a casual user of Photoshop or you’ve just upgraded to a new version, you’ve got some hunting to do.

Exploded View of All Tools

To make it easier to find the tool you want, we’ve created a summary sheet containing all the tools in the tool palette.   There’s one for CS3 and one for CS4.  They are printable and offered free for readers of this site.  Here’s a low resolution peek at one of them.

Photoshop CS3 Expanded Tool Palette

Get Your Copy Free

Use the box below to enter your first name and email address and we will send you PDF files for both of these summary sheets at no charge.

In a previous post, I discussed difficulties I encountered while exporting MP4 files with Final Cut Express 4.0.  In this post, I will discuss difficulties I encountered while importing MP4 files with Final Cut Express. In particular, I noticed that FCE sometimes truncates mp4 clips during import.  I found that I could import 20 megabyte mp4 clips just fine.  But clips just over 30 megabytes in length were shortened.  They were truncated so that they terminated early.  I couldn’t find a setting in the user or system preferences that would allow me to import longer mp4 clips without truncation.

MPEG Streamclip to the Rescue

The solution I found is to split long mp4 clips into shorter clips, each of which can be imported into FCE successfully.  You can put the clips back together once they are inside FCE by dropping them into the same sequence.  The tool I used to split a clip is a very nice freeware conversion and editing utiliy called “MPEG Streamclip.”  You can download MPEG Streamclip at www.squared5.com.  It is available for either Macs or PCs.

Click by Click Solution

Here’s how to split a long clip into two shorter clips with MPEG Streamclip.

• Step 1.  Open your clip in MPEG Streamclip by going to ‘File –> Open Files…‘ and opening the video file you’d like to split into smaller pieces.
  

  Step 1. Open the video file you wish to split.

• Step 2. Set the starting point (the IN point) of the first subclip by placing the playback cursor at the beginning of the clip in the timeline.  The timeline is the playback position indicator at the bottom of the viewer window in MPEG Streamclip.  Now press the ‘i’ key. Alternatively click on ‘Edit –> Select In‘.
  

  Step 2. Set IN point of first subclip.

• Step 3. Set the end (the OUT point) of the first subclip by placing the playback cursor somewhere in the middle of the timeline.  Now press the ‘o‘ key. Alternatively, you can click on ‘Edit –> Select Out‘.  The subclip you have marked should now appear RED in the timeline.
  

  Step 3. Select OUT point of first subclip.

• Step 4. Save the first subclip by clicking on ‘File –> Save As‘.   In the save dialog, fill in the filename and directory where you’d like the subclip to go.  In this example, I named the subclip ‘ExampleSubClip1′.  There’s a menu for the file type at the bottom of the save dialog.  Make sure you set it to MP4, or whatever file type you’d like.  Now click the ‘Save’ button to write the subclip to disk.  This save is very quick, suggesting that it
  doesn’t recompress, but just truncates the data structure.
  

  Step 4. Save the first subclip.

• Step 5. Leaving the cursor EXACTLY where it is on the timeline, clear the in and out points for the first clip.  You can do this by typing the ‘x’ key.  Alternatively, you can click on ‘Edit –> Cancel Selection’.
• Step 6. Set the starting point of the second subclip to the current location of the cursor.  Simply type the ‘i’ key to do this.  This will automatically select the end of the video as the out point of the subclip.
• Step 7. Save the second subclip by clicking on ‘File –> Save As‘.   In the save dialog, fill in the new filename.  In this example, I named the subclip ‘ExampleSubClip2′.  You may want to double check the little file type menu at the bottom of the save window, to make sure subclips 1 and 2 are the same type.  Finally, click the ‘Save’ button.

That’s it!  You now have two clips that you can import into Final Cut Express without truncation.

I’ve been teaching Photoshop to photography students for about 12 years.  When a recent student asked for training videos to supplement the instruction, I decided it was time to make some.  Not knowing what I was getting myself into, I went out and got a copy of Final Cut Express 4.0.  And so the fun begins.

Final Cut Blurries

After reading a few sections of the manual that comes with FCE, I was able to assemble a few mp4 video clips into a sequence.  However, when I tried to export the sequence to get a final video result (also an mp4), I hit a roadblock.  The resulting video was quite blurry and in a dynamic way.   That is, text in the video would sometimes be crisp and sometimes become a blur.  It was as if someone was pouring water over freshly painted watercolor. The blurriness would flow around the image.   I assumed that this was some kind of compression artifact, so I tried using different parameter settings during the export to fix the problem. No luck.  I finally concluded that there must be a bug in FCE’s processing of mp4 files.  So, I tried a different output format.  When I chose the Quicktime Movie format (an mov file), things started to work much better.  With the right settings, I was able to get clean, crisp output.  In this post, I’ll take you through the settings that I found to work, so you can get high quality exports without a fuss.

Click-by-click Solution

My clips were shot with a 720p camera (Nikon D300s) at 24 frames per second.  My goal was to have 720p output that looked as crisp as the original clips.  Certainly no worse.  Here’s the steps that got me output that looked even better than the input.

• Step 1. Click anywhere on the canvas.  This tells FCE which sequence you want to export.
• Step 2. Click on File–>Export–>Using Quicktime Conversion. This will give you a dialog labelled ‘Save‘.
  

  Step 2: Go to File --> Export --> Using Quicktime Conversion

• Step 3. On the ‘Save‘ dialog, type in the name for your movie in the box provided, and select a destination folder.
  

  Step 3: Enter a filename in the 'Save As' box.

• Step 4. Toward the bottom of the ‘Save‘ dialog, there is an item labelled ‘Format:‘.  Choose Format: Quick Time Movie.
  

  Step 4: Select 'Quick Time Movie' format.

• Step 5. Also on the ‘Save‘ dialog toward the bottom, click the ‘Options‘ button, which will pop up a menu with different categories of settings.  This pop up menu will be labelled ‘Movie Settings‘.
• Step 6. In the ‘Movie Settings‘ dialog, you’ll want to make sure that the ‘Settings‘, ‘Filter’ and ‘Size‘ menus are set up correctly.  We’ll walk through each of these in turn.
  

  Step 6. The 'Movie Settings' menu leads to several submenus of settings.

• Step 7. From the ‘Movie Settings‘ dialog, click on ‘Settings‘ to get the ‘Standard Video Compression Settings’ menu. Here’s what I found to work for generally high quality output.
  • Compression Type:   H.264
  • Frame Rate:                Current
  • Key Frames:               Every 24 frames
  • Frame Reordering:   Checked
  • Compressor Quality: Best
  • Encoding:                   Best quality (Multi-pass)
  • Data Rate:                  Automatic
  • Optimized for:           Download

  Click ‘OK‘ when done.

  

  Step 7. Set the compression parameters here.

• Step 8. Also from the ‘Movie Settings‘ dialog, click on ‘Filter…’ for any special effects you’d like.  It’s best to experiment with a small clip, since you can’t see the effects of, say, brightening the image until after the movie is generated.  I prefer only to sharpen at this stage of processing.Sharpen: Amount of sharpening 3
  Click ‘OK‘ when done.
  

  Step 8. Set the amount of sharpening.

• Step 9. Also from the ‘Movie Settings‘ dialog, click on ‘Size‘ to set your pixel dimensions.  Since I shot my material at 1280 x 720p,  I set accordingly.
  • Dimensions:                           1280 x 720 HD
  • Preserve Aspect Ratio:         Unchecked
  • Deinterlace Source Video:   Unchecked

  Click ‘OK‘ when done.

  

  Step 9. Set the desired pixels dimensions and leave boxes unchecked.

• Step 10. With respect to the ‘Movie Settings‘ dialog, I kept the default ‘Sound‘ settings, but you may want to take a peek at this menu to see whether the defaults make sense for you.
• Step 11. Finally, on the ‘Movie Settings‘ dialog, I have set the following.
  Prepare for Internet Streaming: Checked and ‘Fast start‘ selected.
• Step 12. Hit ‘OK‘ when done with the ‘Movie Settings‘.
• Step 13. Click ‘Save‘ on the ‘Save‘ page and prepare to wait while the .mov file is generated.

That’s it. I hope these settings work as well for you as they did for me.

It’s snowing here today.  Such stormy weather can provide opportunities for beautiful and unusual shots. So, I was very eager to get out and see what I could find.  However, there’s nothing worse for your camera than getting either water or sand inside it.   So as I was getting ready to go out and take some pictures,  I looked around the kitchen for something to protect the camera.  Unfortunately, I discovered that I was out of zip-loc bags.

Zip-loc Raincoat

A large zip-loc bag can make an acceptable raincoat for a DSLR or similar camera.   The camera goes inside the bag, and the lens pokes out the opening of the bag.  You can snug the bag around the lens barrel with a rubber band, or sometimes the zip-loc itself will hold.  But what do you do if there are none around?

Cling-wrap Will Do

Camera, lens and lens hood covered with cling-wrap

The one thing I did find in the kitchen was cling-wrap.  I thought I’d give it a try.  Since my lens had no water seals (most don’t),  I wrapped the lens loosely around the lens barrel, overlapping the camera body a bit.  I made sure that the wrap was long enough to go all the way around the lens barrel and overlap itself by several inches.  This kept the wrap from unraveling in the wind.  My camera body does have water seals (some Nikon and Canon bodies have seals), so I didn’t need to completely cover the camera body itself.

I also used a lens hood to help keep snow from falling on the lens.  In such conditions, it’s a good idea to keep a microfiber cloth in your pocket to wipe the lens when it gets wet.   If you do this, use a very gentle touch and make sure there’s no grit on the microfiber cloth.  You want to absorb the water without scratching the lens.  I also put a UV filter over the lens for added protection in such cases.

Because the cling-wrap clings to itself quite well, it managed to stay on the lens even in a moderate wind.  Surprisingly, I had no trouble operating the zoom ring on the lens (the wrap was loose enough for short turns).  All in all, it worked surprisingly well.  The lens was bone dry when I got back home.

Come In Out Of The Cold, Carefully

Incidently, when you bring your camera in out of the cold, make sure you keep the camera in its camera bag with the lens cap on until it has a chance to warm up SLOWLY.   If you bring a cold camera into a warm room with nothing covering it, water will condense on and in the camera and lens.  This can cause internal parts to rust and jam, or even short the electronics.

Snow Photo

Here’s a shot I took with the make-shift wrap.

Snow Covered Tree © Tony Passera 2010

I’m always on the lookout for simple and inexpensive ways of taking pictures.  The term ‘inexpensive’ usually doesn’t go with aerial photography.  However, I recently came across two new products that are moving in the right direction. They are the HD Hero Camera from GoPro, and the FlyCamOne 3 from the Acme game company.  This is neither a preview nor a review of these two new products.  Rather, this post is a quick announcement with some notes about their capabilities.  I will write full reviews after I’ve had a chance to try them out.

Small, Light Weight and Good Image Quality

HD Hero basic camera kit.

The HD Hero is a very small, fixed focus, wide angle video and still camera designed to be worn during extreme sports. Often it is attached to a helmet and shoots everything you see as you race, jump, swim or cycle.  You turn it on and it continues to shoot continuously while you are otherwise occupied. There are optional mounts for attaching the camera to all sorts of things, including car fenders (suction cup), surfboards, handlebars, etc.  The HD Hero can shoot video at 1080p at 30 fps, and 720p at 30 or 60 fps.  It can also shoot 5 megapixel still photos automatically at 2, 5, 10, 30 and 60 second intervals via a built in timer.  The lens is very wide angle, taking in 170 degrees at the 720p setting, and 127 degrees at 1080p.  The aperture is f2.8.  Amazingly, it comes with a waterproof housing that’s good to a depth of 60 meters. The basic camera sells for $259 with battery, mount and underwater housing.

The HD Hero caught my attention for aerial photography for several reasons.  It is extremely small and light, weighing 3.3 ounces (94 grams) including battery, or 5.9 ounces (167 grams) including the waterproof housing and battery.  Its dimensions are 1.6 x 2.4 x 1.2 inches (42mm x 60mm x 30mm).  This makes it suitable for smaller UAV helicopters and airplanes.  Where the HD Hero really surprises is with its image and sound quality.  It uses a point-and-shoot size CMOS sensor with very good light sensitivity.   It shoots to an SD card with no screen, so you don’t see what you’ve got until you can get to a computer.  The videographer, Philip Bloom, has some interesting comments about his first experiences with the HD Hero camera here.

What appeals to me about the HD Hero is the possibility of shooting aerial still photos inexpensively.  Because of its size and weight, it could easily be attached to a 400 series or possibly smaller helicopter.  You could even clamp it to the top of a simple pole or mast.   Shooting real estate aerials could be done by setting its timer to take a picture every 2 seconds.  You fly it attached to your favorite helicopter.  When you want to take a picture, you’ll have to hover, making sure the camera is pointed in the right direction.  If you can hold the hover for 2 seconds, you’ve got the shot.  Since there’s no tilt or pan control, you need to point the helicopter where you want each shot.  It will definitely require a “feel” to get it right, and it won’t be for everybody.  It certainly won’t be fool-proof.  But it could be just the ticket for quick aerial stills where you want to use the smallest, most benign aircraft with very little setup to get into the air.  Let me know if you try it and it works out. Incidently, the sports videos on the GoPro site are amazing. Definitely worth a look.

Smaller and Lighter with Amazing Options

FlyCamOne 3 with downlink

Another camera suitable for very small aircraft is the FlyCamOne 3.  FlyCamOne 3 is the third generation of the popular FlyCamOne series.  What is noteworthy about the new generation is that it has some very useful modules as well as remote tilt and pan control of the camera.  The camera system is comprised of a camera module and a screen module as well as optional transmitter and receiver modules.  The separate camera and screen reduces the flight weight, since you can remove the screen for r/c model flying.  You can mount the screen back on the camera to view video that has been shot and stored in the camera’s SD memory card.  Alternatively, you can mount a FlyCamOne3 transmitter module on the camera, which gives you a real-time video downlink as you fly. The live picture is transferred at 2.4GHz to a receiver module where the removed screen of the camera  can be mounted. Within the near future there will be video goggles will give you a first person view (FPV) from your aircraft. A GPS module is also in the works. Finally, a small hand-held transmitter (433MHz) will be available to remotely tilt and pan the camera, if your r/c receiver doesn’t have extra channels for this purpose.  Typical prices are $139 for the camera module, $99 for the video downlink kit including the transmitter and receiver modules, and $29 for the 433MHz transmitter to control tilt and pan of the camera.

Some of the specs for the FlyCamOne 3 are listed here.

• Size…………………..98mm x 50mm x 15mm
• Weight……………….62 grams
• Battery……………….500mAh Lipo
• LCD…………………..shows battery, memory, mode and action
• Video…………………VGA resolution (640×480) at 28fps
• Head movement…..2 stepper motors for pan and tilt controlled via a standard receiver cable
• Memory……………..micro SD card (4MB/sec write speed).

However, before you get too excited, there are some significant limitations.  First, the transmitter outputs 10 milliwatts and uses spread spectrum technology.   10 milliwatts is fairly low power and will likely limit the downlink range to a few hundred meters.  You should also be aware that transmitting video over spread spectrum could result in the video stream breaking up when there are other flyers using 2.4 gHz nearby.  Of course, this is not a problem for the video streamed to the SD card in the camera.

But that brings us to another limitation.  The camera takes VGA resolution (640 x 480) video at 28 fps.  This, of course, is not suitable for professional purposes.  Video I’ve seen from earlier generations of the camera show the dreaded “jello cam” effect, which is objectionable when there’s a lot of vibration.  This could be more of a problem when FlyCamOne is mounted on a helicopter than on an airplane.

Limitations aside, I find the FlyCamOne 3 to be an exciting development.   The modular design with built in pan and tilt control is very well conceived.  Put in a better camera, a better lens, a bit more power in the transmitter, and you’d have an amazing package for aerial photography.  While FlyCamOne is not intended for the pro photography market, a high end implementation would be exactly what we need.

Aerial Photography: The Essential How-To Guide by Greg McNair

The Bird’s Eye View

Have you ever wondered what the world looks like from a bird’s perspective?  Have you wondered how you might get a small, or maybe not so small, camera airborn to find out?     The book Aerial Photography: The Essential How-To Guide by Greg McNair shows you how to get started with remote controlled (unmanned) aircraft capable of carrying cameras.    I recently bought a copy, which I ordered from his website www.aerialphotobook.com for $49.95.  This post is a brief review of McNair’s book.

The Dreaded AP Mosquito

There’s a rare mosquito that carries the AP virus.  Anyone bitten by such a mosquito finds himself (or herself) obsessed with Aerial Photography.  The symptoms include pouring over books and websites to gather every last bit of information on AP, spending hours on a flight simulator, and hastily buying helicopter kits and parts to assemble a cameraship.  I appear to have the AP virus, since I read Greg McNair’s book in one sitting.  Actually, I was standing.  I tore open the package and read it (it’s only 82 pages) right where I stood.  Fortunately, it was a quick, entertaining and highly informative read.   As you can see from an earlier post, I’ve been exploring remote controlled helicopters as vehicles for photography.   Though I’ve been learning a lot, McNair’s book provided a new flood of information.  It is particularly valuable because it includes so many (web) resources for every aspect of AP.

McNair’s book is meant to be a guide to starting an aerial photography business.  About half the book is devoted to the tools you’ll need to get your camera in the air.  The other half of the book is about building, marketing and maintaining an aerial photography business.

Tools of the Aerial Photography Trade

In the first half of the book he discusses tools that can be used to accomplish the task of getting an aerial view.  These include airplanes, helicopters, weather balloons/blimps, and telescoping masts.  He doesn’t discuss kites, which are another popular means of getting cameras aloft.  This is probably because kites are more dependent on the weather and harder to control than other vehicles. Thus they are less suitable for a business.

For what he does discuss, McNair’s approach is very similar to the approach I’ve taken in previous posts.  He discusses the pros and cons of each type of aircraft and lists numerous resources on the web where you can get more information.  He does not describe how to build anything.  However, he does tell you what  you’ll need to build your cameraship, and where you can get the various modules.  While this is similar in spirit to my posts, his book contains many more resources and more detailed discussions.  It is a treasure of information.  In addition to covering vehicle type (e.g. airplane vs. helicopter), he also discusses the modules or components that are needed to make up a working cameraship.  These include flight assist hardware (GPS and gyro stabilizers), camera mounts, remote camera triggers, downlinks for real-time video monitoring, and cameras. There’s quite a variety of choices for each of these modules.  For example, he discusses 8 different camera triggering devices from as many manufacturers.   Some have mechanical fingers to press the shutter, others use infrared to control the camera, while still others connect to the camera’s USB port, and yet others use the cameras existing electrical shutter release.  Similarly, for camera mounts (harnesses) he discusses about 6 different sources of mounts accommodating everything from point and shoots and palm-size video cameras to DSLR’s and professional video.   Whatever your budget or project goals, he covers something that will do the job.

The Business of Aerial Photography

The second half of the book covers a variety of topics that are important for your business, including how to start your business, marketing secrets, pricing your services, liability insurance, FAA flight restrictions, mitigating risks, and population and ground activity.   These sectoins contain information necessary to run your business profitably and safely.  Everything is based on his first-hand experience.   For example, liability insurance is harder to get for aerial photography because there are so few people doing it.  McNair’s suggestions can save you a lot of time hunting down an underwriter.  Of particular interest is his discussion about potential customers of aerial photography.  His customer list is a lot bigger and more varied than you might guess.  This too is worth the price of the book.  It may surprise you that any FAA (Federal Aviation Administration) flight restrictions apply to remote controlled, unmanned aircraft as well as manned aircraft.  You need to know about any temporary or permanent flight restrictions in the areas you plan to photograph.  For example, events that draw large crowds, such as the Superbowl, are likely to be declared no-fly-zones.  The penalties for violating a flight restriction include jail, 4 or 5 figure fines, and loss of your pilot’s license (if applicable).

Most importantly, the topic of safety is addressed many times throughout the book.  This is of greatest concern with helicopters.   A 600 series RC helicopter has a rotor diameter of 4 feet.  The rotor is a carbon fiber blade spinning at 2000 rpm.  It can kill or maim.   It should be obvious that you must not fly over or near crowds with one of these.  McNair also reminds us that helicopters tend to attract crowds.   It is vital to avoid situations where a crowd might gather.   Otherwise, you need someone on hand, other than the pilot,  who can keep people and pets out of danger.   Dogs and sometimes children may run toward the heli as it is landing or taking off, putting them in extreme danger.  If this should happen, your only choice may be to “ditch” the helicopter, rather than risk injuring someone.

The Right Tool Is Whatever Gets the Shot

At the beginning of the book McNair introduces the mantra, “The remote controlled aircraft is merely a tool for getting the shot.“  He is telling us not to be attached to any one vehicle for getting the shot.  Indeed, the risks and costs of using a helicopter motivated him to get a 55 foot telescoping mast (pole) for “aerial” real estate shots.   This makes perfect sense when you don’t need a high altitude shot, and you’ve got too many trees, telephone poles and people around for an airplane or heli.  Tethered  balloons are another safe and easy alternative.  Whatever method you choose, Aerial Photography: The Essential How-To Guide will provide information and resources to get your camera in the air.