Or play, whatever way I look at what I’m doing now as my full time occupation. Yes, it’s nearly official: I have just one more project to finish, and I’ll be retired from my old day job.

This entry is not really about woodworking (although there is some involved), but how I made this video:

The heart of it a 6.5 second clip that took a full day to create. Admittedly, I took some wrong turns along the way, but it was my first time trying to do a stop motion video as detailed as this.

I had just finished making a pair of crossed gears, and recording how I did it to make a video of the project, when I started thinking about the best way to demonstrate how the gears work. Just standing there, turning them by hand is ok, but lacks impact.
I thought this would be a perfect use of stop motion animation. Of course, by the time this occurred to me, it was 11:15 at night, and I really had to fight the urge to go out to the shop and get started on it.
The next day I could begin and got at it fairly early. The first thing to do was to decide the angle that the gears would be shown from and that’s when I thought it would be nice if the block that I have the gears on rotated 90 degrees while the gears themselves turn. Taking it a step farther, I decided that instead of making the block turn, I’d swing the camera around it in an arc.
Next, I had to figure out how far away the camera should be from the gears and still fit correctly in a 16×9 frame size for the video. My camera takes pictures in a 3×2 format, so I had to take several, then crop them down to 16×9 to find the correct distance. This was the winning entry:

This gave me enough room on top and bottom of the gears with the image cropped out to the edges.

With that critical dimension in hand, I could look through my scrap pile for a piece of melamine that was big enough, and I found a 16″ square of 5/8″ thick for the base. I cut an arm from a scrap of 1/2″ and drove a 1″ finish nail in as the pivot point (arrow):

I used the arm to draw an arc around the base, then removed it. I would need to scale this arc over the 90 degrees so that I could swing the arm a relatively precise amount for each photo. At first, I thought about using my protractor and the framing square to mark out each degree, but this turned out to be tedious and difficult to line up accurately.

Then I figured I’d use the protractor to lay out the 10s, and use a spacer that is the correct thickness to divide those into single degrees:

It turned out that the space is exactly the thickness of 1/4″ plywood.

I knew that there was a chance that I might lose track of the arm position, so I numbered every second degree on the scale:

Moving from zero degrees to 90, the arm runs out of space on the base, so I had to make a second scale that lines up with the other edge of the arm. In hindsight, I really should have used a larger piece of melamine for the base.

Next, I made a place for the gears to sit that would be above the pivot. I screwed a strip of 1/4″ plywood to the bottom of the block that holds the gears:

Then screwed that to a piece of 2×4, and clamped the whole thing down to the table saw.

The plywood projects past the end of the 2×4, leaving space for the base and arm to fit underneath:

This allows me to have the pivot point for the arm directly below the gears, so that the camera will be the same distance away throughout the arc.

To mount the camera, I used a scrap of 3/4″ plywood and a short 1/4″-20 bolt. The bolt had to be cut, since I didn’t have one the right length:

With the camera in place, another problem presented itself: the gears were now higher and I had to angle the camera up slightly to get them back in the frame. I used a block under the front to prop it up, then took some pictures to check (again!).
The lens I’m using is a 10-20mm wide angle set at 10mm. Getting auto focus to work at this close range was tricky, but I couldn’t really see well enough through the viewfinder to manually focus. Once it was focused, I switched to manual focus to lock it in.

With the exact location for the camera determined, I used hot melt glue to hold it to the arm. My glue gun doesn’t get used much, but is very handy to have around. This is the only one I’ve ever owned, and I bought it new nearly thirty years ago. Still going strong:

To avoid it snagging on something while swinging the camera, I removed the strap. It’s doubtful that I’ll put it back and I don’t know why I didn’t take it off earlier, since I rarely use it and it is often a nuisance.

With the weight of the camera on the arm, yet another problem cropped up: the base was not big enough to properly support the arm, and it would tetter off the edge when it neared 90 degrees. I had to add another piece of melamine to the base, but didn’t have any 5/8″ left, so I used 1/2″ and shimmed it up with the framing square and a carving knife:

With it shimmed exactly flush with the bottom of the arm (arrow), I hot glued it it in place.

To help stabilize the arm as much as possible, I stacked a bunch of 1″ washers on as ballast:

Lining it up for the first photo.

I wound up taking more than 800 photos, through four separate series before I finally got it right.

The first two series were done with the overhead lights on. This photo is an example of how that looked:

Series number one was a total of 180 pictures, with one taken for every gear movement and the arm advanced one degree every two pictures. The problem with this was that I turned the gears exactly one tooth, thinking that it would be easier to keep track of. When I strung the photos together and watched it, the gears didn’t really appear to be moving – I would have to look hard at the wood grain to see any motion.

This is similar to a strobe effect, where a spinning object can appear motionless. To make the motion more obvious, I would have to turn the gears slightly more or less than one tooth.
I also noticed that the panning motion was a bit too jerky, so I decided to move the arm by 1/2 degree increments for each photo. One degree is only 1/4″ wide, fifteen inches from the gears, but it’s a whole lot more at the other side of the room. So, back out to the shop for series number two.

This time, the gears really looked like they were turning, but the background panning was still a bit jerky. I decided to up the number of photos to 270, or three per degree of rotation. I also wanted to try it with the overhead lights off, with just a single lamp directly above the gears. Back out to the shop for series number three.
The setup:

To subdivide each degree, I put two dots between, approximately equal. Pinpoint precision is not important for this, it just has to be close:

Equally important to the process is a remote for the camera. Physically pushing the shutter button can move the camera enough to make a visible difference.

Speaking of moving, I was about 220 frames into the third series when I accidentally jarred the base, moving the whole rig by a small amount. I immediately tried to correct it, and finished the series with my fingers crossed that it would not be noticeable. Nope! Back out for number four…

The fourth time was a charm, and I ended up with 276 perfect frames, overshooting the 90 degrees by two degrees, just for the heck of it. Stringing the frames together was easy enough: I loaded them into Power Director full-size and made the duration of each frame 0.01 seconds, the smallest possible. I then rendered this sequence at 2K, 30p and 40 Mbps. Reloading the clip back into Power Director and cropping it down created the vertical pan. I increased the speed by 1.3, since the clip was running a bit slow, and added the titles and music. The second sequence in the video is just the first clip reversed.

Even though the results of my day’s work can be measured in seconds, I must say that I have had much less productive days, and certainly ones that were less fun. I found the whole process extremely interesting, although doing it for a larger project would undoubtedly stress the limits of my patience. It’s something that I wanted to try – a serious stop motion sequence – and I’m completely satisfied with the outcome.