E-Wheelie: New gyroscope == mad behaviour ?

[pbisiac]

Hello, I Just finished to build a modified Elektor Wheelie. My mechanics is slightly different from Elektor kit, as I use a chain, but motors are the same.
The problem I am facing is: when I start wheelie (pushing the presence button) for some fraction of second the algorythm seem to work (motors move slowly, sometimes in the same direction sometimes in opposite), but then both motors start to accelerate until 17V and they never stop. I made these trials with wheelie standing still on a tripod. I tried to ride my wheelie, too, but it runned away and I falled off. I think this is NOT supposed to happen !

In my project I made a little change on Elektor hardware: I wasnt able to find the IDG-300 gyroscope, so I redesigned the gyroboard with an ST LPR550AL. The sensitivity is the same, ±500 °/s, the Zero-rate level is different (1.23 V for LPR550, 1.5 V for IDG), but I don't think ths should be a problem, as firmware makes a "zero point" at startup. I made my tests on this tripod, and the outputs of acceleremeter sticks to 1.5 V, while outputs of gyroscope sticks to 1.23 V (perhaps some little oscillation due to motor vibration, but the problem remains if I disconnect motors).
I programmed the ATMega32 with "official Elektor firmware".
Do you have any idea that could help me ? Thank you very much.

Paolo Bisiach

paolo@bisiach.it'

[pbisiac]

Here I am again...
I am still struggling. My wheelie runs away like a crazy whem I try to ride it, or it starts to turn on itself at high speed, very dangerous too.

I put a 30A fuse between + of first battery and - of second, like I see on Elektor article, but if I try to force my wheelie to stop by holding wheels the fuse blows. This is strange, because the AtTINY25 should inform ATmega when I is > 25A.

To be sure that ATTINY software is working I tried to increment voltage at pin 3 with a 1 K pullup, to simulate an overcurrent; ATTYNY puts thle line CURRFLAG high but duty cycle doesnt change.

I tried also without any fuse (short circuit), but when I mounted on wheelie an tryed to force it not to run away T3 blew up a PCB track. This is also strange....ATTINY should disable MIC2941 and everything should switch off !

I double checked my gyroboard, the ADXL320 is placed with the same orientation of original gyroboard, and my LPR550 in the same orientation of IDG-300 (they have the same output response vs. rotation)

Any suggestion would be appreciated...