real time dyno?

[Kalle Pihlajasaari]

Hi,

> All this dyno talk has me wondering....

Fun what.

> The strain gauge on the driveshaft sure seems like a interesting way
> to measure torque (at least relative) in real time, although wiring
> problems would seem to preclude it.  My meager mind has trouble
> comprehending accomplishing the resolution necessary with hall
> effect sensors.  

One would have to fit a self-contained ballanced collar to the shaft
with a radio transmitter that sends the amplified signal to a suitable
received and decoder/demodulator/logger/&c.  Not impossible but
not a quick fix, alternatively you could supply power by having
a permanent magnet attached to the diff and having a set of coils
around the shaft that generate power and then charge a cap/batt
and then send the signal back via radio or another set of coils
that has a modulated signal with a receiving coils on the front
end of the propshaft (yuk).

> Would be interesting to me, although maybe not practical, to have a
> real time horsepower/torque display.  Any other thoughts on how to

Yeah, what about a diy_efi that tries to maximise acceleration or 
torque or speed when you trip the bottom-out switch :-)

> accomplish this with minimal sensors?  Shaft rpm is certainly simple
> enough, but how/where to get the torque reading?  The motor mount
> suggestion sounds feasible, anyone care to elaborate?  Can anyone see 
> any fatal flaws in the concept?

I have been told that rubber shock mounts allow for too much vibration
which would be hard to eliminate from readings but it might be possible
these days to manage something.  You could try and fit 2 standard 
loadcells under the engine mounts and use the difference signal as
an approximation of the torque generated force.

A cheaper sollution might be just to put a linear POT from teh block to 
the engine compartment and just ignore the up down and left right
motion of the engine and have a simple voltmeter on the dash as a 
engine tilt indicator, with the repeatability of a pot it will give
a bit more use than just taking the bonnet off :-)

One method that I have to try out and promises to be workable is to 
use two RPM sensors (uPi, reciprocal time, with just one index) to measure
the shaft speed before and after the torque converter on an automatic
transmission.  Given temperature and RPM and % slip it is possible to
calibrate (or look up from manufacturers tables) the amount of power
and torque directly, a third sensor after the transmission would allow 
for logging of speed as well.

Cheers
-- 
Kalle Pihlajasaari     kalle at data.co.za
Interface Products     Box 15775, Doornfontein, 2028, South Africa
+27 (11) 402-7750      Fax: +27 (11) 402-7751