Sumobot Rev. D board issue
Hi All,
I have run upon a small issue with a couple of Rev D. Sumobot boards. A little background...
I have 8 sumobots, six of them being Rev C, and 2 of them being·Rev D. The Rev D's are older Rev D's, as they still had the female SIP header as opposed to the newer machine pins. I teach sumo classes regularly and am very familiar with the bot.
A talked a buddy of mine into getting his boys a pair of sumobots a couple of years ago. He finally got around to assembling and playing with them, and was complaining that the object detection was very poor. I assumed that he just had high expectations, as the Sumo vision system works but is not as accurate as folks seem to hope for or expect.
Anyway, he came up to visit this weekend, and I checked out his vision issue. Indeed, the vision was very, very poor. I busted out the scope, and discovered a steady 1kHz beat on the IR vision sensors. Soldering a .1uF cap directly to the power pins of the sensor eliminated the issue, and now they work as well as any.
There is nothing else hooked up to the board, just the batteries and the IR sensors. Even the IR led's are not plugged in. I initially thought it must be hash from the FREQOUT command, but not so. If I put it in an endless null loop (Bla:, goto bla) it does not show up, or if I just put an END or STOP at the top of the program. If however I have it in a loop as above and put·a PAUSE statement (bla:, pause 500, goto bla) it does it. It's processor noise, somehow.
The Rev D's I have do NOT display this issue, but BOTH of his do. As I said above, while they are both Rev D, mine are from an older production run. I tried a couple of Rev C's, and found no problems.
Anyway, I am curious if this is a known issue, and if so what is the problem. As well, if anyone is experiencing troublw with Sumobot vision, I would reccomend adding the decoupling cap. Makes a night and day difference on these two bots. His bots had essentially no vision and were getting false triggers and running al over the place. No wonder he wasn't happy!
Jonathan
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.madlabs.info - Home of the Hydrogen Fuel Cell Robot
I have run upon a small issue with a couple of Rev D. Sumobot boards. A little background...
I have 8 sumobots, six of them being Rev C, and 2 of them being·Rev D. The Rev D's are older Rev D's, as they still had the female SIP header as opposed to the newer machine pins. I teach sumo classes regularly and am very familiar with the bot.
A talked a buddy of mine into getting his boys a pair of sumobots a couple of years ago. He finally got around to assembling and playing with them, and was complaining that the object detection was very poor. I assumed that he just had high expectations, as the Sumo vision system works but is not as accurate as folks seem to hope for or expect.
Anyway, he came up to visit this weekend, and I checked out his vision issue. Indeed, the vision was very, very poor. I busted out the scope, and discovered a steady 1kHz beat on the IR vision sensors. Soldering a .1uF cap directly to the power pins of the sensor eliminated the issue, and now they work as well as any.
There is nothing else hooked up to the board, just the batteries and the IR sensors. Even the IR led's are not plugged in. I initially thought it must be hash from the FREQOUT command, but not so. If I put it in an endless null loop (Bla:, goto bla) it does not show up, or if I just put an END or STOP at the top of the program. If however I have it in a loop as above and put·a PAUSE statement (bla:, pause 500, goto bla) it does it. It's processor noise, somehow.
The Rev D's I have do NOT display this issue, but BOTH of his do. As I said above, while they are both Rev D, mine are from an older production run. I tried a couple of Rev C's, and found no problems.
Anyway, I am curious if this is a known issue, and if so what is the problem. As well, if anyone is experiencing troublw with Sumobot vision, I would reccomend adding the decoupling cap. Makes a night and day difference on these two bots. His bots had essentially no vision and were getting false triggers and running al over the place. No wonder he wasn't happy!
Jonathan
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.madlabs.info - Home of the Hydrogen Fuel Cell Robot
Comments
We are trying to run a sumobot competition and we have 50 rev d sumo bots. and 2 rev c. the rev d's randomly reset themselves.
Dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Dave Andreae
Parallax Tech Support
Post Edited (Dave Andreae (Parallax)) : 3/11/2009 6:48:20 PM GMT
I'm assuming that your advice is about the reset, not the vison issue. Correct?
Tim,
Do your rev D's have the machine pin header? If they do, is there any problem with the vision, and can you scope the output?
Jonathan
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.madlabs.info - Home of the Hydrogen Fuel Cell Robot
We are still testing the differences between the Rev D and Rev C. Currently it looks like the rev D's have more noise on the board. We have placed the caps across the IR detectors. They seem to help.
Dave
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Dave Andreae
Parallax Tech Support·