gm map sensor

[Jeff Webb]

I've had a chance to play around with the GM MAP sensor data.  I've done
a little comparison of the data and equations that people have posted. 
Here's the URL for the stuff I've been working with.

http://mil.ufl.edu/~jwebb/automotive/sensors/pressure/intake-manifold/gm/data/

The gm/contrib directory has the original data files that were posted.

"1bar.pdf" shows a comparison of Walter Sherwin's Delco equation for the
1-Bar sensor with Richard Wakeling's experimental data for his 1-Bar
sensor.  The curve fit for Richard's data is 

1 Bar: Vout = 5.0 * (0.010597 Kpa + -0.116564) Volts

The slope is very close to Walt's equation but the data is shifted by
0.675128 kPa.  This falls within the 1% tolerence of the sensor.  That's
pretty darn good data, Richard.

I also plotted some other data that someone posted long ago (Model 835
and 460).  The slopes match very well, but there is a much bigger
offset.  My guess is that this calibration was done using gauge
pressure, or something of the sort.  If anyone knows where the data in
"gm/contrib/misc/sensor_data.txt" came from, I'd like to know.

The plots shown in the datasheets of two new delphi 1-bar sensors (found
in "gm/datasheets") also match Walter's equation as well.

"2bar.pdf" shows a comparison of Walt's 2-Bar equation with with
Richard's 2-bar data.  Again, the slope is right on, but there's a
little bigger offset of 2.968627 kPa.  This is probably within the
tolerence of the 2-Bar sensor.

2 Bar: Vout = 5.0 * (0.004980 Kpa + -0.054785) Volts

"3bar.pdf" shows a plot of Richard's 3-bar data.  Since Walt doesn't
have an equation for that, I compared it to the plot shown in the
datasheet of a new delco 3-bar sensor (/gm/datasheets/turbo_...). 
There's quite a bit of difference between the two, so I don't really
think that they are supposed to have the same calibration.  Since
Richard was right on for the 1 and 2-bar sensors, I think his data for
the 3-bar would be good as well.

I'd like to make a repository for sensor information of this type.  If
someone has data they would like to submit, please include a file with
your name, email, and as much information as possible about how the data
was collected.  Things like the equipment used, the calibration method,
what is experimental and what is calculated, etc.  Be especially sure to
include the exact part numbers (and identifiying information) of the
devices you tested.  Richard, if you would send me this info, I would
greatly appreciate it.  No hurry, though.

Thanks again to Walt and Richard for sharing your info.  It will be of
great use to me and others as well.

That's all for now,

Jeff Webb
----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org