Hi to all,
I've almost successfully built a spokepov, everything works perfectly, I uploaded the firmware and some images, mounted on my bike and now it works like a charm. At home.
http://img.photobucket.com/albums/v368/ ... P_0153.jpg
http://img.photobucket.com/albums/v368/ ... P_0159.jpg
Unfortunatley, when riding my bike on a real street, the spokepov starts drawing images and keep on working until the speed is lower than -approx - 20km/h. When I go faster than this speed, it stops flashing and only one led per side stays on.
Weird things happen when speed goes back under 20km/h: the spokepov starts back flashing but out of sync regarding the wheel revolution period.
Eventually, it goes completely off, with no chances to revive it even removing the batteries.
The only component that changes from the list specified by Limor is the Hall effect sensor, since I only could find a SS443A ( http://it.rs-online.com/web/search/sear ... &R=1811441 ). It's an unipolar effect which I thought it could be used.
I used a latching hall sensor before with the same result.
When I'll go back home I'll try to reflash the uC firmware, eventually changing the uC too.
Any suggestion?
THANKS in advance!
SpokePOV woes
Moderators: adafruit_support_bill, adafruit
Please be positive and constructive with your questions and comments.
-
- Posts: 150
- Joined: Tue Oct 09, 2007 1:06 am
My italian... it is, how do you say, not so good eh? But I ah try to help you, yes??
Seriously, the behaviour you're describing I've experienced myself. Everything is cool, don't mess with the firmware, its cool. What you're seeing is the microcontroller not seeing the sychonizing pulses from the hall effect sensor. So provided you have the correct orientation for the sensor itself, and you're using a very powerful magnet like the one Lady Ada sells, you need to get the magnet closer to the sensor.
The rated supply voltage for that sensor starts at 3.8 (vs the recommened part's 3.5V), so a 3 cell stick would put you in safer territory there. What color stick?
I'd venture to guess, that the signal is marginal, and when you go fast, there isn't sufficent time for the sensor to trip. The microcontrol thinks that the wheel has stopped, and goes into the standby mode with one led lit per side. When your speed come back down, you're on the edge of the sensor operating, and so the microcontroller is ready some pulses and missing others so it sequences the LEDs with the wrong timing showing a stretched (in the direction of spin) image at random intervals.
Seriously, the behaviour you're describing I've experienced myself. Everything is cool, don't mess with the firmware, its cool. What you're seeing is the microcontroller not seeing the sychonizing pulses from the hall effect sensor. So provided you have the correct orientation for the sensor itself, and you're using a very powerful magnet like the one Lady Ada sells, you need to get the magnet closer to the sensor.
The rated supply voltage for that sensor starts at 3.8 (vs the recommened part's 3.5V), so a 3 cell stick would put you in safer territory there. What color stick?
I'd venture to guess, that the signal is marginal, and when you go fast, there isn't sufficent time for the sensor to trip. The microcontrol thinks that the wheel has stopped, and goes into the standby mode with one led lit per side. When your speed come back down, you're on the edge of the sensor operating, and so the microcontroller is ready some pulses and missing others so it sequences the LEDs with the wrong timing showing a stretched (in the direction of spin) image at random intervals.
- zero the hero
- Posts: 2
- Joined: Wed Jan 09, 2008 7:38 am
Thanks!
I'll not touch the firmware, then!
I'm already using 3 AA cells as power supply, they are rechargeable and of 1.2V each so I'm on the lowest edge of normal power supply needs of the sensor I'm using (thanks for pointing that!). I suppose that using fresh non rechargeable batteries would improve the situation, but I don't have it at home actually...
I used these sensors because they were readly available in Italy.
Ladyada, is it possible to buy just a couple reccomended sensor from you? I've already bought the PCB for this project, as well as a complete set of board for making a x0xb0x (which I completed succesfully! Thanks Limor for this great project!!!)
I'll not touch the firmware, then!
I'm already using 3 AA cells as power supply, they are rechargeable and of 1.2V each so I'm on the lowest edge of normal power supply needs of the sensor I'm using (thanks for pointing that!). I suppose that using fresh non rechargeable batteries would improve the situation, but I don't have it at home actually...
I used these sensors because they were readly available in Italy.
Ladyada, is it possible to buy just a couple reccomended sensor from you? I've already bought the PCB for this project, as well as a complete set of board for making a x0xb0x (which I completed succesfully! Thanks Limor for this great project!!!)
-
- Posts: 150
- Joined: Tue Oct 09, 2007 1:06 am
Zero, take the board off your bike, take your magnet in your hand and pass it over the sensor to get an idea up close how close it needs to be, and what orientaion is the best. Then, when you have characterized the interaction, go back to your bike, and see how you can recreate it.
Recognize that the sensor requires a specific orientation, and that the magnet polarity is also important.
Recognize that the sensor requires a specific orientation, and that the magnet polarity is also important.
Please be positive and constructive with your questions and comments.