Archive for March, 2014

Today’s science lesson

Mentioned at the lunch table: “I should put this (nearly empty) bottle of rooster sauce in the salad spinner to make it come out.”

The kids response: “How would that help?”

“It would work like a centrifuge”

“What’s a centrifuge?”

“but the salad spinner would probably be too small. The sauce might collect in the bottom as well. A bike wheel would work though…”

Well alrighty then. A couple minutes later, the rooster sauce was taped in the rear wheel of my bike. Clearances checked, shifted into high gear. And Spiiiinnnn. Noticably unbalanced, but success! All of the sauce in the bottle wound up in the neck. All of the sauce on the outside of the spout ended up on the wheel. Kids were amused.

What we didn’t talk about was the need to dynamically balance the wheel, possibly with a nearly empty mustard bottle.

Movie Link

No comments

Pillow compilation failure on OSX 10.9.2

For those of you installing Pillow/PIL on the latest version of Mavericks, 10.9.2, using XCode 5.1, it’s going to fail with this error:

 clang: error: unknown argument: '-mno-fused-madd' 

There is a thread with a workaround (Unable to install on OS X 10.9.2 due to unsupported compiler flags:-mno-fused-madd) on GitHub. It would be nice if Google linked it a little more prominently.

No comments

Pi day

It’s Pi day. My poor little raspberry pi is hanging out, finally connected to the controller board for the chicken coop door. It’s only waiting on a final enclosure and some programming. And then the poor pi will be shown to the coop and left there to sit. And raise the door. And lower the door. It’s probably going to sound like the bored computer from Hitchhiker’s Guide to the Galaxy.

No comments

More hardware hacking

One of the joys of this house is that the appliances are all more than 15 years old. That used to not be a problem, but appliances being what they are these days, it means that they’re starting to be just a little bit tempermental.

The oven part of the range failed before – the element went in a most sparkly and disturbing way. Thankfully that didn’t fry anything else, especially the controller. At the time I thought that the controller was an expensive part, probably about half as expensive as just replacing the whole range. But the only part required was the element, and it’s easily and cheaply available.

This morning, part of the controller broke. The up arrow button just didn’t trigger. Timer, useless. Temperature on the oven – good for less than 350.

So, I pulled it apart, looking for obvious signs of failure and a part number. There were many numbers on the part, but none of the turned up the right part on the google. Thankfully, the failure was obvious, and it was just the little plastic piece that connects the membrane to the button on the curcuit board. Just. The plastic button is part of a frame with all the other buttons, and that is part of the clock module. Which is only available (if that’s the word) as a unit from the factory. Which has stopped making them. In the day, they were about $300, maybe a little less than I had estimated a few years back. So, now, this bit of plastic is essentially the value of a new range.

This button presser is currently glued back into place along the hinge line to the frame. I’m skeptical about the long term durability here, since glue on a flex line is kind of iffy. But if that fails, there are other things that could do the job. A 3d printed piece might be able to snap into the hole that’s left, provide some springiness, and a post to press the button. It might even be an upgrade.

Someday, when parts like this go out of production, I hope that the 3d models will be vailable, so that people can print their own. It might take a good bit of time to do the whole face plate, but it wouldn’t be that bad. Materials might be tricky – this feels like ABS, not the standard thermoplastic. But that’s a detail. An individual button would be way faster, and probably just as effective.

No comments

Refurbishing a power adapter

A week for hardware hacking of one form or another. The old laptop’s power cord got shredded in a keyboard tray slider. One day, it’s fine, the next, a cut in the cable, then a discolored cable around the cut, and then, no power to the laptop. I’m not really into paying for a $70 adapter for a computer that’s approaching 8 years old.

Amazon’s got magsafe cables for cheap (<$10) and prime, so armed with the parts and an instructable, I gave it a shot. The worst thing that could happen is sparks, magic smoke, and a even more dead power adapter.

I didn’t have a whole lot of success using pliers to open the casing – they did a bit, but a pair of putty knives, one thick, one thin made a lot more progress. (interestingly enough, I bought one of the putty knives to open the case of my old mac mini to install ram. It’s still running, most of the time.)

It’s not that bad of a repair, it’s just a litte tight getting into it without taking both halves of the shell off and all of the shielding. Once you have access to the solder joints, it’s a quick hit from the iron to desolder the old wires, and another touch of heat to slide in the new ones.

Too late in the process I realized that I didn’t have any glue to resecure the shell, but that’s what electrical tape is for. It’s also a convenient mark to let me know not to trust this adapter without supervision till it’s proven itself for a while.

(images to come, once I get my image processing and posting pipeline back in action following the mavericks upgrade. System python considered harmful)

No comments


Not actually remembering the blog password is a good way to not be posting much. the one place that it’s conveniently saved is on the iPad, which isn’t really ideal for posting.

No comments