Kilian's Robot Shop of Horrors - 03/15/2002

Click thumbnails for a larger view.

Subject: [TCRG] RSOH March 15
Date: Sun, 17 Mar 2002 12:06:52 -0600 (CST)
From: kilian@bobodyne.com (Alan Kilian)
To: tcrobots@orbis.net (Twin Cities Robots group)

Man, we got a lot done on Friday/Saturday.
Mark brought two kits he wanted help soldering together:
Infrared obstacle detector
http://www.lynxmotion.com/irpd.htm
Dual H-Bridge
http://www.lynxmotion.com/index.htm
They were pretty easy, but they were not Heathkit quality instructions. Basically they listed the part-number, type and value, and showed a picture of where they went in the board.
He also brought a Jstamp (Not J-Stamp) http://www.systronix.com/store/jstamp_order.html This is a really cool Java-running microprocessor. He just pasted in some Java code, and we had both kits running in about two hours.
Brynn has a linear CCD he wants to use in a LASER range finder, so he got out his STK300 Atmel AtMEGA103 development kit, loaded up some software, and started writing code while Jeff built a cable to go from the CCD to the development board.
The Atmel development system has a real nice GUI that lets you pick what types of internal peripherals you are going to be using (like timers, A/D, Serial I/O), and it writes a small assembly-language file with all the startup code in it for you.
This is really great. Except that it didn't work. It used the wrong syntax for including a file, the wrong include filename, one internal register name was wrong, and it used the wrong syntax when reading from an I/O port. Brynn had those fixed pretty quickly. They all caused assembly errors.
After all that, it just worked!! We could see the analog output from the CCD on the scope, and Brynn ran a screwdriver across the face of the CCD, and we could see the shadow. There is something strange about just how many CCD elements this thing has. We had to intentionally shift out too many pixels to get a reliable one-frame image.
I'm sure glad we didn't try it with that Cygnal 8051 development board because it is soo much faster than the AVR that we would have had to spend half the night just adding NOPs to get it to run slow enough ;-) (25 MHz vs 4 MHz so the AVR is 16% as fast as the 8051 in terms of single-cycle instructions)
I'm sure Jeff will get pictures up.

-Alan

Brynn's Linear CCD Project

Brynn brought this cute little 2048 element linear CCD array.	He wanted to connect it to his Atmel STK300 eval board.	So we went to the bag of tricks to find parts to connect them.	We decided to put the CCD and some connectors on a piece of perf board.
Jeff is desoldering the flexible PC board from the back of the CCD.	Cleaning up the pins after removing the CCD.	We decided it would be easier to just put SIP sockets on the end of ribbon cable. So Jeff is cutting the connectors to size. (And no one thought to say Don't do anything too stupid.)	Jeff stripping off just the right amount of insulation.
Soldering the connectors to the ribbon cable.	The final connectors on the cable.	The finished cable assembly.	The eval board and the computer for programming.
Checking the wiring on the back of the eval board.	The test setup.	The first test. HEY, IT WORKS!!!	This shows 3 complete scans. The peaks in the middle of each scan is the test screwdriver being moved across the face of the CCD.
These are the individual pixels being shifted out.	Since Brynn normally uses the pile-of-papers filing system, Alan put his data sheets in a folder.

Mark's JStamp Project

Mark Roeder brought his toys over. This is his JStamp. www.jstamp.com	He wants to put it on this Radio Shack tank. Ron traced out the driver circuit on the tank board.	Mark had two kits. This is an IR detector from Lynxmotion.	Alan preparing to do some soldering.
Soldering the IR kit.	The IR kit connected to the JStamp.	Mark is writing some JAVA test code.	We wondered if the camera could see the IR. It gives a slight purple glow. I expected it to be red.
Adjusting the frequency of the transmitter.	Testing the range.	This is the other kit connected to the JStamp. It is a dual h-bridge from Lynxmotion.

Other

We finally found the box of
goodies that were taken off
the Angie robot.

Back to Index

Back to my home page http://www.pobox.com/~jsampson

Visit Twin Cities Robotics Group

This page is currently maintained by Jeff Sampson