botputer

Make magazine has to be one of the best publications ever - and they’ve created a great weblog as well. Anne got me a subscription and I’ve only received two issues, but after the first issue, I knew that I could never go back.

Immediately, I knew that I needed to build some of the things in the magazine … but for what purpose? Issue 8 gave me the inspiration with the Ornithopter: my 9-year-old nephew, Raine. I started thinking how cool a lot of these projects were and that at his age, I would have LOVED if one of my uncles had built some of these things with me - in addition to the wrestling and hard-tackle football we played. And Raine is at the age where he is interested in how things work and why. And many of the Make projects cover a number of subjects, which would be good for him.

It was with this mindset that I fell in love with the Pummer - my incarnations of which will here-to-fore be known as botplants, or biobots.

At first, the Pummer seems to be a simple beast: during the day, it charges the batteries and and night it blinks an LED. Simple as it may seem, with only one Integrated Circuit (IC) chip, Bill and I spent some time figuring out how the circuit from the article even worked - and, since it clearly must work, WHY it worked. After a number of consultations … and a number of trips to different places to get parts (like Ax-Man and RadioShack), today I was finally ready to take the plunge and try out the circuit. I chose to build the circuit ahead of time because I can imagine how frustrating it would be for it to not work after finally building it with Raine - and how uninterested he would be by the end of it.

Connecting everything up on my breadboard took less than an hour — being my first foray into much along the lines of electronics, I had no idea how long it would take me to follow the schematic in the article and bring it to fruition. However, after making only a couple rookie mistakes that I figured out by myself (like hooking up the diode in the wrong direction … and, similarly, hooking up the LED in the wrong direction), I got the circuit to work (without the solar cell - because it hasn’t arrived yet). Looking at the circuit, it shouldn’t be difficult to hook up the solar cell and have everything work now that the rest is working.

Below are pictures of the breadboard version, and a video of the LED blinking for proof.

The whole breadboard:

The left-hand-side with the small capacitors and a handful of resistors:

The right-hand-side with the big capacitor, a few resistors, and the LED (caught while blinking):

Another view of the LED blinking:

And here’s the video (feel free to ignore the Calexico playing in the background):
  

All-in-all a pretty good day. We’ll see how things go when I need to pull it all together again while getting Raine to do most of the work.

A few nights ago (after some sushi, a martini, and some beer), I double checked all the soldering and all the wiring on the cvs cam. Solder joints looked good (no joints seemed to be crossing). Wiring looked good (meaning I didn’t criss-cross colors or anything stupid).

So, I plugged the usb cable into the camera again without changing anything. This time, rather than freaking out my computer (and me), it worked as expected - meaning the camera was recognized as a “Saturn” device and I was asked to install drivers, which I politely declined.

Next up … installing the drivers to see if I can get my crappy test video off the camera.

For Win2k, I found a few versions of the drivers to download: 0.1.8.0, 0.1.10.0, and 0.1.10.1. Even though 0.1.8.0 was recommended, being a reasonably computer-savvy user, I thought the newest version would be best, however, to be safe, I also downloaded the recommended version. Having both 0.1.10.1 and 0.1.8.0, I began - one wonders at this point, why anyone would install a version 0.1.x of anything (and 4 version points? Interesting and scary.) … but hindsight is always 20-20.

I Installed v10 [0.1.10.1, that is] and all was fine. I started the test application and everything seemed happy. However, based on the screenshot from this post, I expected to see some information about the Manufacturer (== “Pure Digital”). This was a little bit important to me because I have a couple other usb devices connected to my computer (including a usb wifi device for internet connectivity). I thought that maybe this had been suppressed in the newer version. So - not wanting to mess anything up by choosing the wrong device - I uninstalled v10 and installed v8 [0.1.8.0, that is] - rather than just unplugging the other devices and running the test app again (this is an important point to note because I had never considered this a possibility until just now while writing about my escapade).

Installing v8 seemed to go fine, but gave me a number of weird messages after the fact. I’ll spare you the long list of crazy things that happened (especially since I can no longer read the notes I kept). However, the short version is that I was warned about a usb device that was unsafely unplugged, followed by a couple “USB found hardware” messages. I was asked to restart, which I politely declined. I plugged in the cvs cam, which showed up as a “Saturn” device. Not perfect, but things seemed to be coming around - after a little frustration … and kicking myself for being greedy and wanting the Manufacturer information displayed. I start the test program … my computer reboots. Uhhh, crap. After it starts up, I run the test program again … my computer reboots again. Here’s where I start worrying about being greedy and wishing I had left v10 installed.

It may interest the reader that at this point none of my usb devices are working - including my internet connection. Checking out the device manager, I see that all my usb devices have an unfriendly icon featuring an exclamation point engulfed in a yellow circle. Clicking for more detail (for any of the devices), the device status says: “Your registry might be corrupted.” Excellent.

At this point, I think that maybe it was a bad idea to install v8, so I uninstall it - ahh, the beautiful naivete that can sometimes consume an otherwise savvy - but currently hopeful - computer user.

I then re-install v10, run the test program … nothing … no usb devices found. I restart the computer, run the test program … nothing … no usb devices found … AGAIN. Here we jump back a few paragraphs and find out that - oddly enough - all my usb devices are disabled in the device manager, so I enable them. It’s actually at this point in my story that I find “Code 19″: “Your registry might be corrupted.” I looked at the CD that came with the motherboard and it wasn’t very helpful. It pointed out that the usb drivers were supplied by Microsoft. So I tried to update the drivers through the device manager … no change. At this point, I’m looking to re-install Windows to hopefully fix the registry … or whatever is the problem.

Clearly, this leaves me more than a little disheartened. Thankfully - or so I tell myself - I have an ethernet card in my computer and can get by for now by running cable down the hallway (under the rug) into the bedroom where my hub is located. So I have internet access again.

A couple weeks have now passed without much movement on this issue.

UPDATE: Success! My USB devices are working again. Digging around in the Device Manager again, I (finally) followed the instructions for troubleshooting my USB Controller devices. I was directed to open Regedt32 to check that the System user had full access to some key and all its children (I don’t think this made any difference, but I did it anyway). There was also a link to the Hardware Troubleshooter, which I thankfully followed.

I chose “I’m having a problem with a Universal Serial Bus (USB) device” and followed the directions to restart in Safe Mode and uninstall all my USB devices - it only mentioned the troublesome USB device I was trying to fix, but all my USB controllers were messed up, so I decided to uninstall all the USB Root Hubs and USB Controllers. I restarted the computer and all my USB controllers were displayed in the Device Manager, happy as could be. Plugging in my USB wifi device and digital camera proved that they were indeed working correctly again. Apparently, I need to pay more attention to Windows Help pages in the first place.

I ordered a few USB female-female connectors (USB A to USB mini B) to use for the usb hack and received them Friday, 2006.10.27. I chose these connectors because my digital camera uses a usb mini b cable. Since I always leave the cable for it connected to my computer, I thought it would be nice to be able to use the same cable for the cvs video camera (and the connector is nice and small, which I thought would be good for hacking it into the camera - that and I’m a freak).

After getting home from work last Friday to find the connectors had arrived, I immediately broke into the plastic case of one to see what it was like inside.

Sunday night, I finally got around to trying to solder things together. I removed the USB A female connector from my connector and yanked the camera apart. Things seemed to be going okay so far.

Next, I found short wires (harvested from some extra cat-5), tinned the wires, and figured out which pins on the connector should go to which pins on the camera. Then I tried to solder the wires to the pins on from the USB A connector. After spending plenty of time getting frustrated and wondering why I didn’t just give up and watch some TV, I finally removed the pins and just soldered the wires right to the board.

After that foray into electronics soldering, Anne was kind enough to help out on the camera side - she’s got mad soldering skills from all her stained glass work.

I also used my dremel to cut a hole in the side of the camera where I wanted the usb port to reside. I cut the hole and everything seemed to fit - except that I was not nearly as accurate with the dremel as I had hoped to be.

— add image —

Before putting the camera back together, I thought I would connect the camera to my computer to make sure everything seemed to be going right - or at least consistent with my expectations from what I had previously read. That’s when things didn’t seem to go as well.

When I plugged the usb cable into the camera, the camera started up - so far so good. I expected to be asked to install drivers for the new unknown usb device, but instead, my computer immediately gave me a windows warning that I had unsafely removed a usb device (my external usb wifi device) … which I had not done - quite the opposite, in fact. A little concerning, but whatever. When I tried to click Ok, I realized that my mouse was no longer working - but my mouse is not - I repeat NOT - a usb mouse. A little more concerning. Unplugging the camera seemed to make everyone happy again, but I rebooted the computer just to be safe.

By this time, it was starting to get a little late and I was overly tired from Friday and Saturday night Halloween parties, so I decided to put off the fight for another night. My plan now is to double check my wiring (including checking to see that each of the wires has a good connection) and try again to make sure the same thing happens. Then I’ll try to install the drivers to see if that changes anything. If I still have problems, I’m planning to use a different usb connector (and maybe have Anne do the solder work), in case I ruined something with my crappy soldering.

Wish me luck.

A quick recap of events up to the launch of botputer.com:

Order robostix-wifi pack, including Gumstix, RoboStix, Wifistix

• ordered: 2006.08.20
• shipped: 2006.08.24
• received: 2006.08.31

Somewhere between those dates

• Buy CVS One-time use digital video camera
• Assemble Gumstix (multiple times)

• Realize that I cannot connect to Gumstix over Wifistix after trying many different things from different forums including:
• adding resistors to the robostix [graciously supplied by Bill]
• supplying the power on the robostix
• supplying the power on the wifistix
• and even entirely removing the robostix from the equation - so only the wifistix was connected to the gumstix)

Then order tweener board and serial null-modem cable so I can connect to the Gumstix via HyperTerm.

• ordered: 2006.09.25
• shipped: 2006.09.25
• received: 2006.10.02

2006.10.08: Finally connect to Gumstix with tweener board and cable. Realize that I cannot connect over wifi because my gumstix shipped with an old build (really old) - one that does not start the wifi card on startup, so I get the latest filesystem image and replace mine. Now I can connect over wifi, as originally expected - with the robostix attached, and it doesn’t currently seem to matter if the power is connected to the robostix or the wifistix (but based on the forums, I expect that I will eventually need to supply power to both).

2006.10.10: Bill supplies me with an extra digital multimeter and a mini grabber patch wire he has around his house.

2006.10.18: Pick up breadboard, LEDs, hookup wire, and resistors from RadioShack.

2006.10.24: Bill supplies me with another mini grabber patch wire and some connector housings. I come to the realization that Bill has basically become my parts supplier, but he has yet to accept payment of any sort.

2006.10.26: After playing around with my new heap of parts and looking at what it will take to try get the robostix to do the flashing LED sample, I buy botputer.com in a drunken frenzy. As though it’s still 1998, I quickly snatch up the .org and .net domains for no reason. Before the drunken domain buying spree, I had some weird trouble with the gumstix acting funny when trying to flash the robostix code for Simple-Flasher. I realize that I probably need to do more research (mostly here, here and here) and promptly give up for the evening.

2006.10.29: Setup wordpress and modify Hello World entry. Then 1998 calls and wants their stupid decision back regarding registering all three domains.

That brings us to the end of October.

Doo Doo Doo Doo …
botputer. is. alive!
Doo Doo Doo Doo …