Well, it looks like the joysticks are going to be in tomorrow, so I’ll be able to control the movement of the maze table a little easier. Today I found an online maze generator and whipped up a sample. Evelyn used her Snag-It magic to get the thing to print larger (I needed 15″ x 15″). Here’s a photo:

On the right is the 15″ x 15″ board, with a 4-sheet maze printout glued to it. This 15″ square board will get mounted to the 12″ tilting base. Having this “board on a board” allows me to easily swap out playing surfaces.

On the left, you see a Letter size piece of paper glued to a box. That was my test piece, as I’m using construction adhesive to affix the maze “wall”. We were going to use little rocks as the wall, but that was going to take way too long, and may have blocked the marble from moving around. For the prototype, we’re using some bright pipe cleaners, about 12″ long each. They are just tall enough (when glued to the surface) to stop the marble from falling off.

<rant>

Why, oh WordPress 2.5 must you create JavaScript errors when trying to “Add Media” to posts? Why did you prevent photos from being added, due to stupid debug code? Why doesn’t your Post editing RTF box stay resided (large) like the old version did? Why must I always enlarge it again? Do most of your users access WordPress on an Atari 800XL? Why must your editing windows and themes be so narrow? Are there really that many people running 800×600? Why do you go for glitze and break the most common features, and yet still don’t have super-easy linking (YouTube, anyone?). What about spell check? You have your fancy Video Upload, and yet most people don’t host their videos. Why must all of these sites go “widget” crazy and yet break the things I need, you know, like FUCKING POST EDITING. Christ almighty. Web 2.0 can kiss my ass. Let’s be done with this experiment and move on.

</rant>

Don’t get me wrong, I love WordPress, and 2.5 is nice. But, come on. It’s embarrassing to break the most common features (creatings posts, adding images to posts) so that you can spend more time creating a widget engine that I’ll never use. Sigh.

Update: I just upgraded to version 2.5.1. Same problems trying to insert photos into existing (published) posts. Nothing more annoying than trying to use the SIMPLE FEATURES that I use NINETY PERCENT OF THE TIME and they DON’T WORK. Ugh. Web 2.0, suck it.

We’re big fans of popcorn, and order many pounds of the most excellent Crown Jewel Popcorn on a regular basis. We use a stove top popper, because fat is flavor, and it makes a better bowl of popcorn. For a while we’ve used the original Whirley Pop. I love the machine, but I hate that they are using plastic gears for their mechanism. Ours has split once, we wired it together, and it just split again. Into the garbage.

So Evelyn spotted another stovetop popper at Target during her last run. This one is made in China (unlike the US made Whirley) and seems pretty decent–and it’s $10 cheaper:

The nice thing is that it has steel gears, unlike the breaky plastic ones. The bad thing was noticed the moment I started to stir: It binds. The bottom mechanism was binding against the bottom, and the handle gear wasn’t working to great either. Another piece of trash? No. This could be handled the old fashioned way: The Paperclip.

Doing my best MacGyver impression, I grabbed two paperclips, bent them a bit, and jammed them below the gears. Basically, you want the paperclip to act like a bit of spring metal, pushing “up” the stir gear, and pressing “out” the handle gear. Two paper clips, bent just the right way, makes the gears mesh perfectly. It’s now a dream to use, though the lid kinda sucks. Maybe it needs another paperclip…

This project must be dedicated to my Dad, and you’ll see why. Here it is:

What you see is a carbon air filter made out of a Dollar Store “crate” (about 8″ x 7″ x 6″). It’s used to pull soldering fumes aware from the builder, clean out some of the gunk, and exhaust cleaner air. The black part is the carbon filter.

Here’s a view with the filter taken off. Those of the Nelson Clan will understand why this is dedicated to my Dad:

Fans! My Dad loves fans, and this has four of them. The above shot is kinda cool, because you can see them spinning (a full-sized photo is in the gallery).

The unit is designed to suck air in through the front (through the filter), then it goes through the fans, and is exhausted out the back and the sides. I used pieces of foamboard, cut to fit tightly, to make a 2″ space between the fans and the filter. This helps prevent the fans from getting ”bogged down” and acts as a baffle.

Here’s a photo of the switch. It’s nice when you have a round switch that fits perfectly into your dollar store crate:

In the above photo, you can see the exhaust holes on the side (all four sides). There are also holes in the back. This allows you to operate the unit on any side, or sitting upright, depending on your needs.

The power supply is a left-over unit from a dead external hard drive (thanks, Mike). It give 12V @ 3A. You need a LOT of power to drive fans. My bench supply started getting sad at the 8 fan count.

I designed the unit to support 8 fans total, four toward the rear, and four up front (right behind the filter). So far, I’ve only done the rear four fans, because I want to see if the other four are necessary. We tested the draw with some incense and it was sucking fumes from over 18″ away. Not bad!

Why so many fans? Well, I was originally going to do this project with one nice fan. But the filter is tough on the guys. If you put the filter too close, it bogs down the fan. So, I decided to beat the problem with quantity, and ordered a bunch (10) of 80mm fans from Goldmine Electronics. They were dirt cheap ($1.60) each–in fact, those 4 fans cost less than the one nice ball bearing one I was previously testing with. More fans mean more surface area. From my experiments, it seems to give a nice draw, doesn’t kill the fans, and has a good (6″ x 8″) filter surface area.

The baffle & spacing between the fans and the filter is important. The space seems to help the fans intake better, and they don’t sound anywhere near as labored as when the filter is closer. The filter sits about 2 ” from the fans. For the filter material, I bought a package of carbon filter from Lowes. It was about $12 and is a big 16″ x 48″ piece, so I’ll have filters for a while.

In total, the cost was: $1 crate, $8 fans, $1 foamboard, $0 recycled power supply & jack, plus $2 for filter (cut up). That’s $10 for the unit and <$2 for filters. Given that new units start in the $50 range, that’s not too bad.

It won’t filter out nasty crap, but it will at least clean up those solder fumes. And it’s small enough to fit on a desk. I was going to make this hose-based, with the unit on the floor, but this came out nice, so we’re leaving it on the workbench.

In Puerto Rico, no less! As I mentioned previously, PR is handing out federal “rebate” checks, just like in the states. Well, today Evelyn’s mom received a check in the mail, addressed to me, for $600. I gotta give them credit, they actually got those out pretty quick. I still haven’t got my US Federal one.

So, the question is, should I have been stimulated twice? If not, do I keep the $600 from PR or the $1200 from the Feds? Either way, the entire amount is going to my quarterly estimated tax payment. Time to call the IRS.

We’ve been hearing a lot about food costs recently, and how they are going up and up. From a subjective standpoint, it does “feel” like prices are rising, but it’s harder to tell just looking at the totals. One large item can throw things off.

As you know, I keep a very detailed food shopping list. It helps Evelyn navigate the store faster, and helps us to identify real sales from bogus ones. Well, I was going through the list, updating “average” prices and came up with some interesting numbers:

Heavy Cream – Was $1.76, now $1.99. Up 12%.
Baked Beans – Was $0.94, now $0.97. Up 3%
Refried Beans – Was $0.67, now $0.72. Up 7%
Potato Packets – Was $0.94, now $0.99. Up 5%

That’s an average increase of 6.75% for these couple of small items. I don’t know about the rest of you, but I didn’t get a 6.75% raise last quarter. Or the quarter before that. Or the quarter before that. It would be one thing if these increases were “annual”, but I’m seeing +5% jumps every quarter. 

With food and utilities going up-up-up at this rate, it’s no surprise that the consumer has less (if any) discretionary income. I’m feeling extremely fortunate that my work keeps me close to home, and that I live in a city where I can walk and bike to reach most everything I need.

I have a hunch that we’re going to see more and more people migrate away from the far suburbs and toward the cities again. In the past, the working class poor moved out for lower home prices. But now fuel costs have wiped out any of those savings.

And frankly, there are plenty of areas in most major cities that can support a rebirth and influx of people. There are huge tracts of “wastelands” in places like Chicago, just minutes from downtown. Gentrification is often given a bad time, but I think the majority of people are going to need to live in the cities again. It’s the only way they can afford to live. It takes good urban planning, though. Something I rarely (if ever) see.

Here’s a video of a couple of fireflies wired and mounted to some of the plastic branches we’ll be using. This assembly, along with the little circuit board, will be placed inside a standard sized mason jar. There will be five “bugs” per jar, and they give a nice night-light-like glow. The board gets covered up with some Spanish moss, and perhaps some pebbles.

Evelyn did a really nice job wiring them. She used 30 gauge black and green wire-wrap wire, which is pretty thin, but not as hard to work with as some of the tiny “magnet” wire I’ve seen others use. It blends in very nicely with the plants. The effect is more impressive in person, and looks pretty neat in the jar. You get a nice glow and a feeling of a little ecosystem in the jar.

The switch and battery will be up-top, near the screw top. Just unscrew and flip a switch. No need to dig down into the “plants”. It will run off a 9V battery. With these LEDs and the CMOS chips, this could run for a long time without a battery change.

Here’s a quick video of the Maze table. I only have it tilting along one axis. Evelyn sped up the video a bit, since I was running it on 9V at the time. The actual piece will use 12V and runs a bit faster than what you see here. (That’s a container of Loctite super glue sitting on top, to help illustrate the movement.)

One of the big challenges will be putting the right material on the board. A plain board, with a marble, is very “slick” and hard to control. Though, part of the control problem may be because I’m just toggling switches back and forth, versus using a joytick (which will give me 8 directions). We’ll see.

Now that the playing field board is attached to the mirror assembly, it’s time to re-mount the whole thing onto the base. Here’s what it looks like, side view:

So, from top to bottom, you have: The foam board, where the maze will be. The tilting assembly. And, finally, the base. You can see why I only did a “dry fit” of the motor to the base. Otherwise it would have bee too difficult to screw on the top part.

Here’s an angled view where you can see how the things are bolted to the base:

In the photo above you can see the bolts/nuts holding the board to the motor assembly along with the bolts/nuts holding the whole thing to the wooden base. You can also spot some of the test wires that I have in place.

Here is another view of the device, along with Evelyn’s breakout board switch (DPDT). I used it to test each motor. (There’s only one test switch, so I could only test one motor at a time.)

When I was originally coming up with this idea, I was thinking that this would be a computer controlled maze game. Something were a computer would guide a marble through a maze (similar to the “mouse in a maze” computer programs I wrote on my old Atari 800XL).

However, I think making this a game of skill would be more fun. So, I decided to have the user (human) control the tilt of the board, using a joystick. It will all work something like this:

• The board will have a maze built up on it, much like the wooden labyrinth game.
• The center of the maze will be the starting point.
• Around the four corners will be finish / destination points. They will be lit by LEDs.
• The user starts a new game. The microcontroller will choose one of four ending points and light it up.
• The user then must tilt the board, using a joystick, to move the marble toward the ending point (finish line) within a time limit.
• If the user reaches the goal within the time limit, they get a point, the MCU picks a new destination, and they repeat the process.
• After a successful round, the amount of time for the next round decreases. For example, they may have 30 seconds to complete round 1, but only 28 for round 2, and so on.
• Obstacles could be added. For example, if the user knocks into one of them they lose time off the clock. Or it could be a mechanical effect. Think of pinball game bumpers.

The beauty of the game is that the playing field can be independent from the base structure. The game is the same: get a ball from point A to point B. What the user can change is the playing field. In my prototype I’ll do something simple, like the wooden maze. But you don’t have to do that. You could make a playing field of anything you want. Plastic army guys, blocking the way. Mutiple marbles on the field. Different materials on the base (which would affect the speed of the ball), and so on. I’ve made my base square, but there’s no reason the playing field couldn’t be round, or a triangle, or whatever.

So that’s it for now. I’m waiting for my joystick parts to come in, hopefully later this week. In the meantime, I’ll start building out the maze with foamboard, and work on the LED/sensor setup. (I need something to tell me that the ball has arrived at the destination.)

Next up, the playing field. For the prototype I’m going with a piece of decent foam board. I say “decent” because some of the dollar store stuff warps easily. Since I need a flat surface, I used a piece of the better “shiny” board we pick up at a craft store. If this were the finished product, I’d probably use PVC board, or something like that. But, for the prototype, cheap is good.

I decided to go with a 12 inch square. It’s big enough to make a maze on, yet small enough to deal with. It’s also a good ratio for the base I’m using. In the above photo, I’ve cut out the 12 inch square out of foam board, and have drawn two lines between the corners. Why the lines? Look here:

The mirror assembly has it’s four mounting holes, and they are laid out in a square. My board is a square, so there is a ratio between one and the other. Instead of messing around with calculating where the piece should be mounted, I simply have to line up the lines within the “crosshairs” of each mounting hole. If the lines are right in the middle of each hole, you know you have dead center.

(As an aside, I wish they taught this sort of thing in school, in addition to the “pure math”. Sure, kids may learn at what time two trains meet when one leaves the station at 2:34 and the other at 1:15, but they can’t use that skill for anything else. There was even a recent study to prove this. In my opinion, teaching kids real world skills is far more important, and becomes the pathway into “higher” math. Because it helps give them a real-world grasp of the problems. Yeah, I know, there’s that “pure math” that exists only in equations, but 99% of us will never have a use for that. Get kids started with real world hands on problems, then get into the theory. As usual, we teach everything backwards.)

Okay, enough ranting, back to construction: I used a little finger drill, similar to a gimlet, to make a pilot hole into the foam. I used the drill because my pen didn’t reach. Then I used a larger, though still manual, drill to make the hole bigger. Don’t use a power tool here. There’s no need, and you’ll destroy the board.

Next, I installed the screws. These are 1 inch long and are a bit fatter than the #6′s I used earlier. They are just long enough to go through the board, the plastic, and then have a nut. They are wider to better fit the plastic holes in the mirror assembly.

Finally, here is the mirror assembly mounted to the screws. I’ll hold it in place with nuts. I had a photo of that, but it came out blurry.

When you are choosing mounting screws, make sure to keep in mind the table material thickness. You want those screws to just pop out enough to tighten a nut onto. If you go any longer they might bind as the assembly tilts. In the above photo, you can see how close that right screw gets to the black base. If your screws are too long, just put some washers between the board and the mirror assembly.

Also, everything here should be hand tightened. If you use a driver, you risk breaking the fragile foam board. If you’re worried about the nuts coming loose, just put a bit of Loctite on them, or maybe a thin lock washer. Since this is a prototype, I’m not doing that, in case I need to take things apart again.

We’re getting close. Next time you’ll see how the platform, mirror assembly, and base go together.