Pocket watch time-lapse.

A while ago, I bought a pocket watch, which has a window on its face exposing the internal workings. Because most of the gears moved too slow for the movement to be perceptible, I decided to shoot a time-lapse sequence of it. The resulting video was interesting, but pretty repetitive, so I ran the individual frames through a number of filters using ImageMagick, and blended from one filter to the next. Here is the result:

More water droplet photography.

After a couple of upgrades to my setup, I’ve taken a few more water droplet pictures. I wanted to experiment with some different backgrounds, so I started with horizontal stripes. Here are a couple of my favourites:

The main difference in my setup is that I am now using a photo-interrupter to detect the water droplets as opposed to a laser. Because the photo-interrupter is much closer to the source of the drops, it’s much easier to align correctly:

Time-lapse of fog in Vancouver.

I shot this sequence in January of this year. We had a few unusually foggy days here in Vancouver, and I wanted to see what a time-lapse sequence of the fog rolling in and out would look like. I took three sequences of shots over the course of three days, but I think that this sequence turned out much better than the other two. The setup was similar to that of the previous video I posted, however, I intentionally shot this one with a ‘cooler’ white balance because I liked the way it looked.

As with the last video, the music is a public domain recording courtesy of Musopen; in this case, of Beethoven’s moonlight sonata

First experimentation with time-lapse photography.

A couple of years ago, I picked up a Canon A620 with the intention of installing CHDK on it and taking some time-lapse sequences. After a couple of tries, I was able to shoot the following:

The camera was taking pictures all day. I believe the interval between the shots was around 20 seconds. The video starts in the morning and ends in the evening, but I cut out a large section in the middle because it was overcast and nothing interesting was happening; it’s difficult to notice where this happens because I overlapped the earlier and later sections by a second or two and blended from one to the other.

The blending in the middle and the fade-to-black at the end were done on the command line using bash scripts and ImageMagick. I encoded the final video using mencoder.

The music is a public domain recording courtesy of Musopen.

Water droplet photography.

I recently decided to build a flash trigger so that I could try my hand at high speed photography, and water droplet photography seemed to be a good starting point. The first step was to actually build the trigger box.

You’ll probably want to skip this paragraph if you’re not interested in the inner-workings of the trigger circuit. The design is fairly straightforward, it uses an Amtel ATmega8 as the brain, a couple of potentiometers to set the delay, one potentiometer to set the sensor sensitivity, and some audio jacks for the camera shutter, the flash, and the sensor. The flash and camera shutter are controlled using opto-isolators, and the sensor is fed GND, +5v on the left channel, and receives a signal on the right channel. The signal is fed into the analog comparator on the ATmega8, along with the voltage from the sensitivity potentiometer. When the voltage from the sensor becomes higher than the voltage set by the potentiometer, the adjustable delay period begins and then the flash is fired.

This is the setup I used to take the pictures in this post. I used an adjustable dripper (left over from an automated houseplant irrigation project) as the source for the droplets and adjusted the dripper until I was getting a single drop every 10 seconds or so. The delay period starts when the drop crosses a laser beam created using laser pointer.

It was a bit difficult to get the drops to hit the laser dead-on, but after a bit of trial and error I was able to get some results that I’m quite pleased with:

By holding a piece of blue bristol board in different positions when the flash fired, I was able to produce some different effects:

Fixing a bad LCD monitor.

My secondary LCD monitor, an Acer AL1916W, recently developed a number of horizontal lines which would move vertically up or down the display (I can’t remember which). It was subtle at first, but eventually became so noticeable that the monitor was very difficult to use. Because the monitor still worked in every other respect, my somewhat educated guess was that a capacitor had gone bad somewhere in the monitor. After a bit of googling, I came across this site which not only described the same symptoms that I was having, but also identified the culprits as bad filter capacitors in the power supply.

Armed with this knowledge, I disassembled my monitor and immediately noticed that there were four caps in the power supply with bulging tops, one of which looked as though it had started to leak.

I was surprised to notice that although my monitor was from a different manufacturer than the monitor in the post linked above, the power supplies appeared to be identical. I bought some replacement caps from the local supply store, but noticed after I got home that the larger two replacements were not rated for as high a temperature as the ones that had gone bad (85° vs 105°). From what I understand, high temperatures can be a factor in premature capacitor failure, so I ended up using a couple of properly rated capacitors that I happened to have lying around. They were slightly larger than the ones I removed from the board, but I was able to make them fit.

The board after the replacements:

I’m happy to report that after replacing the capacitors, the horizontal lines were gone!