Thermal Sensor and Load Sensor

[xxalexx at ix.netcom.com]

> To:            diy_efi at esl.eng.ohio-state.edu
> Subject:       Re: Thermal Sensor and Load Sensor
> Date:          Sat, 18 Jul 98 16:57:41 -0700
> From:          Andrew Ghali <andrewg at netcom.com>
> Reply-to:      diy_efi at efi332.eng.ohio-state.edu
 
> The problem with strain gauges is that you'll need a) some sort of slip ring
> contacts for the rotating part or b) magnetic power and signal coupling to
> transfer as a variable frequency, right?  The problems with (a) are obvious;
> (b) is a bit complex and probably fairly expensive, and subject to calibration
> problems.  The concept of a phase-based sensor is old, and can utilize the
> same technology used in synchro/resolver equipment.  Yes, filtering would be
> required on any method - your data will be delayed a few turns of the shaft -
> not too bad?  The one problem that none of these methods addresses came up
> in a discussion with a Motorola automotive-group field application engineer:
> work hardening of the shaft.  You are depending on a known modulus of the
> shaft in order to determine the torque from its deflection.  That's what the
> Moto guy told me was the reason the method was not used in passenger cars.
> I suppose race cars with new shafts every race would have no problem, but in
> the long haul...
> 
How about two oppositely polarized magnetic rings on shaft 
See   J. Appl. Phys. 79 (8) 4756-4758, (1996)

alex