MAP conversion

[Roger Heflin]

On Fri, 25 Sep 1998, Mike Pitts wrote:

> 
> Hi,
> 
> Does anyone have an idea as to what the following 1-bar MAP 
> conversion is actually calculating.  The application is a CPI 
> V6 truck.  The result is used in most of the table lookups which 
> use MAP.
> 
> Value = ((((MAPCounts - 26.0) * 0.59) + 0.25) * 2.0)
> 
> MAPCounts represents the raw A/D counts read from the 
> 1-bar MAP sensor.  The formula limits the useable range of 
> the MAP from 26 to 242 counts (20 kPa to 97 kPa).  Anything 
> outside of that range causes the value to set to 255. 
> 
> Is the formula simply scaling the MAP counts into a more 
> usable range?
> 

First, a comment.  A/D values aren't generally refered to as counts. 
They are generally binary values that represent voltages or some
other signal.  So if you knew what the voltage range was set to
on the input the  (real voltage = (x/256)*range+bottom_rail).

>From the above equation 26 -> 0 (integer) .25 (real)
			242-> 255 (integer) 255.38 (real)

So they seem to be converting their limited operating range to
a full 0-255 scale probably to attempt to give themselves more
steps.  And given that anything outside that range sets it to 
255 (is there some comparisons to do this?), I would assume
that if they equation produced a 255 that would also be a 
problem, so 242 would be an error, that is if the code is handling
255 in a special way.  Since the map is measuring pressure in the
manifold they may want it scaled from 0 to 1 ie. 0 = 0 and 
1=255.  It may also be that something about the sensor
makes it have a minimum value of 26 even if things or lower, or
since it is a map, it would seem to be difficult to get lower
than 20kpa without having a serious restriction even at idle.  The
top value converts to just over 14psi, that may be the highest
anyone expects things to be in the manifold on a NA engine.  I know
my MAP readings on my car have hit 29 (inches of mecury?) on diacom,
I will need to check what things look like in the metric system.
The equation is probably just converting how the sensor reads to
a 0=0 max=255 numbering system.  You really don't want an A/D 
to be going full range, and depending on things you may want to
stay both extreams by enough to be able to "see" bad values while
still within the A/D convertors operating range.

			Roger