gearwheel encoding

[Walters Chris]

>> I seem to recall some discussion on gearwheel encoding some time
>> ago, and I'm sorry for bringing it up again but...
>>
>> I want to put a 60 tooth wheel (with 2 missing) on the crank shaft of
>> a motorcycle engine that goes up to 12000 rpm.  My problem is this, I
>> don't know what the min. diameter of the wheel should be.  At 3deg. per
>> tooth edge, the tooth length gets quite small as the diameter goes down.
>>
>> I am using a Siemens hall effect IC to pick up the pulses, which I 
estimate
>> will be fast enough at static speeds.  However, at hard acceleration or
>> deceleration, I assume that there will be more opportunity to miss teeth
>> or add more, if the teeth are very small.
>
>The older motronics used to use 129 teeth, or up to 132 tooth
>flywheels, that we're approximately 3mm teeth with 3mm gaps, or for you
>americans out there about .120-.125" (1/8")
>
>This would get you down to a 4.77" diameter flywheel, using the 60
>tooth system.

Sure would be convenient to have the teeth directly on the harmonic balancer 
(for those of us retro- guys)

Further, if this is possible, why couldn't you machine 6 rows of teeth on 
the balancer? If each row was a Grey-code pattern of 64 teeth you could read 
position with 6 sensors for each row.
2^6 = 64 ~ 60 and     360 / (6 * 64)  =  .9375 degrees resolution. A ROM 
could be used to map position to rotational angle. Only difficulty I see 
here is manufacturing the 6 rows each offset by .9375 degrees. And you'd 
need 36 sensors kept in an equally precise housing.

Snake
TWO pretty cool Fords, one very cool Dodge