The View Up Here

Random scribblings about kites, photography, machining, and anything else

Archive for May, 2014

Interchangeable KAP Bits

Posted by Tom Benedict on 29/05/2014

Now that the aimable helmet camera is finished I’m on to the next project: interchangeable KAP bits.

A kite aerial photography rig is made up of two main parts: a suspension and a camera mount. The suspension is the bit that connects the KAP rig to the kite line. The camera mount is the bit that holds the camera and lets the photographer aim it at the subject. Each of these can be as simple or as complicated as the rig builder wants them to be. In previous posts I’ve talked about a number of ways to over-complicate a camera mount. In the near future I hope to come up with some new ways to over-complicate the suspension. But in order to do the testing I want to do on the new suspension I need a way to swap one rig between a Picavet, a pendulum, a handheld pole, or what have you. This is what I came up with:

KAP Suspension Connector

I made these out of 3/8″ 6061 aluminum round. Square stock would’ve worked just as well and wouldn’t have required a lathe. The two bits are threaded to go onto a Brooxes Better Gear Guide pan axle and to take a Brooxes Picavet. Both of these use #8-32 screws. I used the same screw thread for the two screws that hold the two parts together. There’s nothing magic about the #8-32 screw. If your rig uses metric hardware you can use metric hardware for this part as well. M4 is a good size.

This probably would’ve worked with one screw, but I like backup when it comes to KAP rigs. Having a camera fall out of the air is never a good thing. Besides, by using them in opposition like this I was able to design the parts to be identical regardless of whether you’re talking about the camera end or the suspension end. The two halves are entirely interchangeable. In both cases the outboard hole is tapped #8-32, and the inboard hole is drilled for tight clearance on a #8-32 screw.

Between the two screws is another hole. I originally designed the hole to take a 1/8″ dowel pin, so these are drilled a few thou under the 0.1240″ diameter of my press-fit reamer. The idea was to ream one as a press fit and the other as a slip fit. But in the end I decided the dowel pin wasn’t really necessary, and having pins in some of the parts and not in others would’ve broken the nice symmetry I’d built into them. Which leaves the question: what to do with that hole? Easy! Safety tethers! Now there’s even more safety in case both screws fall out.

I’ll have the opportunity to try these out during World Wide KAP Week 2014. Meanwhile I’m building a slightly beefier version of this to use on the end of a 25′ painter’s pole. Once that’s done I’ll be able to use the same KAP rig on a Picavet, a pendulum, or a pole, just by popping out two screws.

– Tom

P.S. I typically write a post over the course of several days, editing at the beginning of a writing session and writing during the second half. In the middle of this post’s write/edit cycle I was approached at work about the possibility of doing some still photos using a cable cam. Only problem: we don’t have a cable cam at work. Hey, no problem! The suspension on my KAP rig is now interchangeable! So swapping it over to whatever cable cam system we build is a no-brainer. YAY!

Posted in Engineering, Kite Aerial Photography, Photography | Leave a Comment »

Testing the Helmet Cam

Posted by Tom Benedict on 27/05/2014

In my previous post I described the design(ish) and build of a helmet cam with a reflex sight. Over the last few days I had the chance to take it out and use it on two separate occasions for two very different purposes. Here’s how it went:

Outing #1: Documenting Kite Aerial Photography

One of the reasons I built the helmet cam is that I want to document how I do kite aerial photography. There seems to be a lot of confusion when it comes to things like attaching the camera to the kite line, how the kite goes up the line (hint: it doesn’t), and why KAPers insist on using single line kites instead of two-line or four-line kites. A good set of still photos or even a short video could answer most of these questions. I’ve tried photographing the process of doing KAP in the past, but without a second person to operate the camera it’s just not possible. Enter the hat cam.

To try this I took it with me on a recent outing to the anchialine ponds near the old village of Wainanalii. The ponds are out in the middle of a lava flow, so the only way to get there is on foot. I parked my car off the side of the Queen Kaahumanu Highway and walked in.

King's Trail

A short way in off the road you run into the King’s Highway. This is a trail system commissioned by King Kalakaua in the 1870s. Considering when it was built it’s a remarkable piece of engineering. It runs straight and flat for most of its length, and it spans much of the north Kona coast. I took it from where I parked my car to the turnoff for the anchialine ponds.

As I walked down the King’s Highway I became painfully aware that I had a kiawe thorn stuck straight through my shoe and into my foot. Every time I stepped on a rock my foot was stabbed. Right about the time I reached the turn-off to the ponds, I remembered I always pack pliers in my KAP bag. YAY! I pulled out the thorn and kept going.

Trail to Wainanalii Pond

The trail to the ponds is a little more rough and ready than the King’s Highway. It’s in reasonable shape, but there’s no doubt you’re walking on rocks. It only takes a couple of minutes of hiking the trail before you reach the ponds.

Wainanalii Pond (Golden Pools)

This set of anchialine ponds are sometimes called the Golden Pools for obvious reasons. The color comes from the plants growing on the rocks in the water. As pretty as it is from the ground, it’s even more striking from the air.

Hand Launch

The ponds are in the middle of a relatively fresh lava field so the rocks are sharp enough to cut kite line and shred kite cloth. Since the entire island is volcanic, the same can be said of a number of places I fly. I long ago learned to hand-launch my kites. Long-line launches are simply out of the question.

Clean Wind

Once the kite was up it flew beautifully. The wind has been squirrelly the past few years, but it seems that the end of the drought we’ve been having has brought back the steady winds I grew used to when I first started doing KAP here in Hawaii back in 2007. I let out about 150′ of line before hanging the rig on the line.

Hanging the Rig

The reason most KAPers hang their camera well below their kites is that the more line there is between the camera and the kite, the less the motion of the kite can influence it. Even a relatively stable kite will cause some camera wobble if the camera is suspended directly below the kite bridle. But with even 50′ of separation that motion can be reduced to a slight sway.

The rig is attached firmly to the kite line using snap hooks. The hooks don’t move on the line, so the camera is fixed to that point. All that’s required to raise the camera is to let out more kite line. To bring the camera back down it’s a simple matter of reeling the line back in.

Up and Running

After letting out some line to give the camera some altitude, it was time to do some kite aerial photography. The rig I’m using has two axes of motion: pan and tilt. Some while back I added a video downlink, but it’s not necessary to use it. I did KAP for years aiming by eye and produced some good photos this way. When I added the video link to this rig I did it in such a way that I could bypass it any time I wanted to.

That being said, the viewfinder is handy for tricky subjects or for tricky situations like the one I encountered that day. The sky was blah overcast which caused nasty reflections in the water. I stuck a polarizer on the end of my lens to cut the reflections, but because of the overcast conditions it was very sensitive to orientation. The viewfinder helped me tune the filter’s orientation at each pointing before tripping the shutter. Here are the results:

Golden Pools 1

Golden Pools 2

Except for the two KAP photos at the end all of these were done using the helmet cam. I had both the Gopro and the A2200 on intervalometer mode. Unfortunately I’d left the A2200 set for videos rather than stills, so all I got from it was a series of three second videos made every five seconds. GAAH! All of the helmet cam still photos were from the Gopro. I’m not partial to the Gopro’s fisheye effect when making stills, so I used PTLens to de-fisheye the frames. They were then cropped to a 3:2 ratio in Photoshop. Note to self: double check the mode of the camera next time!

Outing #2: Slope Soaring Videography

The other reason I built the helmet cam was to make slope soaring videos. Slope soaring is one of my favorite ways to fly RC aircraft. It requires no motor, no aero-tow, no high-start, no nothing. You just toss your plane off of a high place and use the wind to create lift. It’s about as close to pure flying as you can get. It also doesn’t cost an arm and a leg to do, which is great for someone like me. All it takes is a plane, a slope, and some wind. Honestly, it’s about as close as an RC airplane can come to a kite. (Hmmmm! Now I know why I like it so much!)

The day after the KAP session at the anchialine ponds I took my Zagi 5C and my helmet cam out to the cinder cone near Kua Bay. This has been a popular slope soaring site on the Big Island for many many years. It’s a good place to fly, but folks who are more used to grassier slopes may find the landing zone a little harsh.

I wanted to test two things: First, I wanted to make videos with both the Gopro and the A2200. Second, I wanted to make a set of stills to see how the fields of view compared. For this test I had the A2200 zoomed all the way in.

I should probably preface the rest of this by saying that with the exception of one Bixler flight, I haven’t flown a plane in months. I certainly haven’t flown slope. The wind was coming from an oddball quarter, so the lift zone wasn’t where I was used to finding it. All of this combined to make for challenging flying. Throw the helmet cam into the mix and I had a hard time splitting my attention between flying the plane and aiming the camera. I eventually did get the hang of it, but in the beginning I made a lot of short flights that ended with abrupt landings.

The reflex sight made it very easy to keep the camera pointed at the plane. When making stills I found it helped to set the shutter sound as loud as I could. The camera was running a five second intervalometer, so every five seconds I heard a loud “kachunk” sound, letting me know it had taken a picture. After a while my flying fell into a cadence of one-two-three-four-click! one-two-three-four-click! At around four seconds I found myself rolling the plane so I could see the top or bottom of the wing. This was great for photography, but it exacerbated my tendency to fly off the lift. Still, the pictures were nice.

Zagi 5C In Flight

You don’t even want to see the corresponding Gopro shot from that point in time. The plane is just a dot out toward the water. The long reach of the A2200 really brought the plane in close, and the reflex sight made it easy to keep it in the frame.

Videos were a little tougher. As a test I ran the Gopro in wide mode, the A2200 zoomed all the way in, an 808 #16D keychain camera on the Zagi, and my T2i on the ground on a tripod. The T2i footage was rubbish, but the rest came out ok. I also recorded audio using a portable field recorder. My audio work isn’t all that hot, but the video came out better than I expected.

I ran into two big problems: The first was that even a tiny lapse in concentration was enough for the plane to move out toward the edge of the field in the A2200. When I really focused, it worked better. I tried putting the dot near the plane rather than directly on it so that I could shift it around in the frame. This didn’t work out too well. The field of view of the A2200 was so small, even tiny shifts were enough for it to slide out toward the edge of the frame. But with judicious editing most of those lapses could have been covered. I didn’t cover them because I wanted to see the faults with the method as well as its strengths.

The second problem I ran into was that keeping the horizon level was non-trivial. I never realized how much we roll our heads when we turn our necks. It’s astounding how much the horizon bobbled a round in the video! By the end, though, my rolly-polly horizon was getting better.

I think both of these problems would have been minimized if I hadn’t been the one flying the airplane. The helmet cam certainly lets you document your own flights. But I think its real strength may be in videoing the flights of others.

– Tom

Posted in Engineering, Kite Aerial Photography, Photography, RC Airplanes | 1 Comment »

Building the Helmet Cam

Posted by Tom Benedict on 19/05/2014

The biggest problem with most hat cams or helmet cams is that there’s no way to aim them accurately. A couple of posts back I mentioned a project to attach a reflex sight to a helmet camera to give me some way to keep the camera aimed at the subject. I ordered a reflex sight and a Picatinny / Weaver rail to mount it on along with a Gopro Frame mount. All the bits and pieces have arrived, so I started designing. Then I realized this is really a scrap box project: the design and build phases are really one and the same. So I started over with the bare bits I wanted to stick on the helmet: a Gopro, an A2200, and a reflex sight, and got busy.

Two Cameras and a Reflex Sight

Reflex Sight:

The reflex sight has nearly zero magnification and projects a red dot into your field of vision as if you had a laser pointed out at the landscape. The idea is that once the sight is dialed in, the center of the frame of the camera will lie where the red dot points.

Reflex Sight Dot

The dot is projected out at infinity, which explains the reason why I couldn’t get the dot and the body of the sight in focus at the same time. (Dang limited depth of field!) Even though this one was made for a firearm, reflex sights like this have been used as wide field finder scopes on amateur telescopes for decades. Adapting one to this application seemed reasonable.

Camera Mounts:

I toyed with the idea of having the two cameras share a common mount, but wound up scrapping the idea. The field of view of the Gopro is wide enough that I can get the two reasonably aligned even if the Gopro is on a separate mount. Since my Gopro came with a bunch of helmet mounts I sacrificed one to the project and stuck it to my bicycle helmet. Done.

Gopro Helmet Mount

Since I’m planning to use the A2200 zoomed in to about 35-50mm equivalent, its pointing is more critical. The reflex sight has screws for adjusting azimuth and elevation, so the mount for the A2200 is fixed – no ballhead, no adjustments, just a hard stop at the back to keep it from rotating on its tripod screw. The reflex sight and its mount need to maintain registration with the A2200 mount, so I built them as a single unit that can be mounted to my bicycle helmet.

Reflex Sight Mount:

I’m left-eye dominant. Most of the time when building aiming devices you want to use your dominant eye. But since the idea here is to also have one unobstructed eye to… well… to keep an eye on things, I also wanted to leave my dominant eye free. In short, I really wasn’t sure which eye to put the sight in front of. So I made it ambidextrous. To do this I just needed to have another rail on the other side of the camera mount so the sight can be mounted on either one. In anticipation of this I bought two lengths of rail. As it turns out each one was just over twice as long as I needed, so I cut one in half and kept the other as a spare.

Reflex Sight on Rails

Since the camera will be mounted above my eye line rather than below, I decided to mount the sight upside-down. This is the opposite of how these are typically mounted, but in this case it made sense. This keeps everything nice and compact, minimizes parallax between the sight and the camera, and simplifies attaching the sight to the camera mount.

Helmet Mount:

A quick word about the fate of my bicycle helmet: It’s doomed never to be used on a bicycle again. In order to mount all this hardware to the helmet I wound up screwing and gluing a strip of aluminum to the front of the helmet. From a crash standpoint this basically means I’ve got lots of sharp metal points aiming at my head. A crash with this thing on would be an instant frontal lobotomy. So from this day forth it’s strictly a camera mount.

Reflex Helmet Mount

The helmet has a vent hole straight down its centerline so I had to use a wider strip of aluminum than I wanted to. I wound up using a strip 1.7″ wide with screws spaced 1″ apart down the sides. The strip was bent across a form (the corner of the bandsaw table) and glued using white Gorilla Glue. It’s not coming off.

The edge of the strip that protrudes down in front of my face has  double row of holes spaced 1″ apart. The holes are spaced 0.2″ apart vertically to give me some range of height adjustment for the camera mount. I put four corresponding threaded holes in the camera mount spaced 0.3″ apart. Depending on the combination of holes I use, I can change the height of the whole camera mount in 0.1″ increments. It turned out I didn’t need this kind of adjustability, but it’s there.

Putting It All Together:

A lot of the “where should things go” decisions on this were made by standing in front of a mirror with all the various parts in hand. I know now my eyes are slightly less than 3″ apart, they’re shifty, and I’m not to be trusted. No! Wait! Disregard that. (They really are less than 3″ apart, though. Gotta love calipers.) I now know my helmet stops several inches above my eye line. So much so that I had to change my original idea for the mount to add more “drop” to it. Once I got those issues sorted out, though, it was just a matter of bolting it all together.

Whole Shebang

Of course no design choice, no machining technique, could keep this from making me look like an utter dork once I strapped it to my head.

Dork in a Headdress

But hey, if it gets good photos and video footage, it’s worth it.

– Tom

Posted in Photography, RC Airplanes | 5 Comments »

Coming Up for Air

Posted by Tom Benedict on 13/05/2014

It’s been a rough three months.

I have no clue if I’ve written about any of this, so bear with me if this covers stuff I’ve already talked about. Some months back my wife was diagnosed with a coarctation of the aorta. This is a birth defect that’s typically caught in the teen years. Left uncorrected it usually results in permanent damage to the internal organs, but by virtue of her active lifestyle, her diet, or by sheer providence, Rydra’s internal organs have taken no damage at all. Even better news, it’s a treatable condition. So about two months ago she went in for graft surgery.

Because of complications that only came to light the day of the surgery, the graft never happened. Plan B was to install a stent, but at that point no one knew how much her aorta could be dilated, and stents for aortas aren’t exactly off-the-shelf items. So they went with plan C and performed a balloon angioplasty during which her aorta was dilated to 12mm. (A normal aorta runs from 22 to 26mm in diameter. Prior to surgery hers was constricted to 3mm.) A couple of days after the procedure we flew home so she could recover and her doctors could see about getting an appropriately sized stent.

Recovery was supposed to take a couple of weeks, but she never really recovered. For the first few days she couldn’t digest anything except oatmeal and cream of wheat cereal. That improved a little as the days went by, but not by much. Her legs had their full strength, but no real stamina. After a few weeks she could walk to the end of the driveway and back, but that was it. Clearly something was not right.

Her surgeon ordered a CTA of her abdomen to see what was going on. The evening after the CTA he called to tell her he’d scheduled a CTA of her chest for the next morning. Something was wrong with her aorta: it was splitting. Just downstream of the coarctation, the inner lining of her aorta had torn. As blood flowed into the tear it inflated the lining, separating it from the aorta wall further and further down toward her abdomen. As it inflated it blocked the flow of blood through the rest of her aorta. This was restricting flow to her legs and internal organs. He booked the OR, and we booked plane tickets back to the hospital.

This time the surgeons installed a stent across the constricted area of her aorta and down over the tear. The idea was to cut off the entrance to this false path so that blood could no longer flow into it. Over time her body would reabsorb the blood in that false path, restoring her aorta to normal function. But complication is her middle name. The operation called for a small incision over her femoral artery so the surgeons could insert their tools. But because of the coarctation, her femoral arteries are atrophied. The surgeon made a longer incision, then a longer one, then a longer one, trying to find a large enough section of artery to insert the tools. She came out of the OR with a seven inch incision leading up from her leg to her abdomen. But they got the stent in place!

She was discharged less than a week later. We flew home for the second time – me a little more rattled, her feeling a little more beat up. But hey, she was well on the road to recovery, right? Right… Less than 24 hours after we got home one of the things I’d been terrified of happened: a hard knot started to form over her femoral artery. This is a sign of potential internal bleeding. One of the points that was drilled into the class when I did my EMR training is that internal bleeding is bad bad bad. Internal bleeding from an artery is worse worse worse. And internal bleeding from the aorta or femoral arteries can leave only a few minutes to respond, depending on the severity of the injury. We called her surgeon, packed a bag in case she had to be transported back to the hospital, and drove to the ER.

As it turns out it was a combination of a seroma – a buildup of fluid – and a hematoma – a buildup of blood – both associated with the incision site rather than the femoral artery itself. We spent most of the night in the ER, but when she was released it was to go home, not to be air lifted to a larger hospital. We both sighed a big sigh of relief. But it delayed her recovery. The pressure of the seroma made it almost impossible for her to sit down, stand up, or get in and out of bed by herself. I wound up taking an extra week off from work to assist her as the seroma was slowly reabsorbed.

Rydra made steady progress over the course of the last two weeks, and on Saturday she gave me the green light to step out of the house for some R&R. There was no wind for flying kites, so I grabbed my recently repaired Bixler 2, my battery bag, and my transmitter, and headed into town. After that long a break I probably should’ve started with a couple of hours on a simulator, but I didn’t. I just stuck a battery in, pre-flighted the plane, and tossed it. I was rusty, and the flying was unremarkable. But it was like seeing sunlight for the first time in months.

I’m back at work as of today. I went home for lunch and spent most of my lunch hour going for a walk around the neighborhood with Rydra. Already she is better than she was before the second operation, and it’s only been two weeks since the ER visit. Things are finally really looking up.

– Tom

Posted in Uncategorized | 1 Comment »