Friday, July 17, 2009

Simple Resistive Load

Two days ago, a couple of fellow students (David Alvarado and David Munday) and I built a simple resistive load bank. We took a cue from Jennifer Elaan's "Old World Light Bulb Load" and used incandescent bulbs as the load. The last porcelain fixture on the end has a plug on it, so that we can daisy chain more interesting (reactive) load modules later on.

The enclosure was designed in Solidworks and cut out of 1/4" MDF by our local laser cutter. It was fitted and glued together using the standard tab-in-slot method.

The total cost of this project was about $60, but would have ended up costing <$30 if we had chosen to buy our parts from a different hardware store. (Soon, I shall post a 3D PDF of the assembly showing our wire stress-relief mount.)
video

Labels: , ,

Wednesday, July 1, 2009

Control Panel


I've wanted to design a control panel using a laser cutter for quite awhile now; This automated scanner project has finally given me the chance. I cut the parts using professor Gabriel Elkaim's 25 watt LaserProII. The material is 1/8" opaque black acrylic, purchased from a local plastics supplier. I estimate that the materials for this control panel cost me approximately $15; This cost, and the fact that the panel is completely custom justify the time I wasted (about 5 hours) designing it.

I made a mistake during the construction of the panel by gluing the faceplate down. I had meant to leave it loose so that I could actually put everything inside (Things are pretty cramped in the enclosure). The next revision of the panel will likely be twice as large, feature matte acrylic instead of glossy, and a faceplate which is screwed on rather than just slotted. By then, I should have the software for the unit mostly worked out and thus will also know how many buttons are required.

The unit was designed in solidworks. When I finish the second revision, I will post the solidworks and corel draw files to thingiverse.

Note: Better picture to come

Labels: , ,

Thursday, June 4, 2009

UCSC's OptIPortal

Today I got to play with UCSC's OptIPortal. It was great fun. I would love to get this thing to work with some Wii remote head tracking demos... Possible fun project for the summer?

Labels: , ,

Tuesday, June 2, 2009

Wolfram Alpha



Geeks are great. That is all...

Monday, June 1, 2009

Maker Faire 09

So, I finally got to go to Maker Faire. I've been following Make magazine and blog since their respective inceptions; However, because of the fact that the Faire occurs in late May (Spring quarter finals), I have not been able to attend the last 3.

The experience was pretty awesome. It would have been easy to stay for both days of the faire and not see all of the things I wanted to see, or meet all of the people I would have liked to have met.

I was able to bring our CE118 (Mechatronics) robot down and show it off at our table in the robot room. It seems that the little guy held up pretty well the whole weekend. I had to repair one of the bump sensors once (So glad somebody thought to pack a hot glue gun).

[I will post any pictures from the event if I see any around, but in the meantime, just in case you are wondering, here is a picture of our (myself, Zach Bernal, Robert Rhodes) bot]

Labels: ,

Wednesday, May 27, 2009

Squirrels under the bridge

Yesterday, when I was walking back from class with a professor of mine, we happened upon a group of baby squirrels living in an alcove underneath the concrete where the sidewalk meets the bridge abutment. I was quite shocked to find that they had decided to nest so close to such an active street. There were perhaps 8 or 10 of them out yesterday. However, today, when I came back with my camera, I only spotted four.



Cannon EOS 450D/Rebel XSi woes.

Lately, I have been fussing with the camera, trying to control the shutter release mechanism remotely via my MacBook Pro. Ideally, it would have been nice to use a single library/front-end utility to control the shutter as well as download images from the camera. But I have run into some problems...

It seems that gphoto2 has problems taking control of the camera. I get the following error:

*** Error ***
An error occurred in the io-library ('Could not claim the USB device'): Could not claim interface 0 (m). Make sure no other program or kernel module (such as sdc2xx, stv680, spca50x) is using the device and you have read/write access to the device.

*** Error (-53: 'Could not claim the USB device') ***


Rather than spending my time troubleshooting this problem (gphoto2 support documentation for the mac is virtually non-existent), I brought up my Ubuntu virtual machine and installed the latest and greatest version (2.4.6) of libgphoto2/gphoto2 from the debian unstable repository. gphoto2 (2.4.5) under Ubuntu worked like a charm; I was able to grab a summary of the Camera status, and and configure it. Unfortunately, it seems that libgphoto2, does not support many of the remotely configurable features of this camera. The only parameters I was able to change were the focus lock and storage location of remotely captured images (device RAM/SD). These limitations, however, do not really affect me because all I really need to with the camera is remotely capture images and then grab them from the cam.

Also of note: Although the gphoto2 libraries for the 450D/Rebel XSi claim that you can remotely capture images to device RAM and SD, this functionality doesnt seem to actually work. The only way that I was able to capture images remotely was by using the --capture-image-and-download command line flag.

Because of the difficulty associated with getting gphoto2 up and running with this particular camera, accompanied by the fact that none of the camera options are configurable remotely by gphoto2, I am inclined to create a serial controller for the entire scanner setup. This serial controller would not only interface to the camera's 3/32" tether port, but would also provide an interface to the VXM stepper motor controller that would allow for the operation of the rig without an attached computer. Also, the controller would likely be able to do shutter control across the entire line of Canon SLR cameras.

Labels: , , ,