Hi, i feel that it is somewhat weird that tomtoms's GPS/navigation runs
Linux, but the only software makes it possible to update the maps and
stuff that exists are for mac or window$. It is called Tomtom Home.
Why is there not an alternative that works on native linux ?


-----------------


A idea is that it is perhaps possible to make software that is
compatible with
the megasquirt ECU family. All of them uses serial communication over
DB9, the newest one has USB in addition.
you can read more about megasquirt here:
http://en.wikipedia.org/wiki/MegaSquirt

The newest megasquirt ECU is MS-III
http://www.ms3efi.com/



It should use the total pulsewith of the injectors and information on
the speed of the vehicle to compute mpg/liters per kilometers of fuel used.




It should then be possible to calculate the usage of fuel, and display
the results.

I think that a tomtom unit can handle the calculations if a suitable
application is written, since they run linux.
It can include gps speed data in the calculations.

A VVS sensor can optionally be used.

Considerations for MS MPG calculating ( the VVS sensor route
)---------------------------------



First, you must get the distance traveled information into the MS.

You will need to configure a digital input pin in the firmware on the
processor.

Then you will need to configure the MS to count bits on that pin,
after a certain number of pulses off of a vehicle speed sensor.
An example is a Ford VSS sensor:
An 8 bit count up variable could register 3/100 of a mile on a Ford VSS.
I dont know how other VVS sensors behave.

It is possible to get VVS sensors that can be connected on the speedo wire.
http://www.magsensors.com/vss.html

Next you need to configure MS3/tomtom to know the GPH/liters per hour
flow of your
injectors ( it should be able to calculate this ) and calculate the
gallons used in that , say 3/100 of a mile window
probably by assuming the pulse-width stayed the same the entire time
of the event. It can now calculate an estimated instantaneous MPG.

If you would like to have a trip cumulative MPG/liters per kilometer
average, it should be
relatively easy to have MS3/tomtom keep a running total of
miles/kilometers traveled
and gallons/liters used.

The more complicated part is historical average MPG. To do this, MS3 or
tomtom would have to regularly
store a total mileage traveled and total gas used figure since it never
knows when its power will be cut.

As far as how many bytes are needed to save all of this information,
the 2nd byte counted up by MS3 can save mileage to 8 miles.
This might be a good time to store that data if one regularly takes trips of
at least this length. If trips are regularly less, you will loose a
significant amount of mileage data.
MS3 or tomtom could store the data on a SDcard.


Ford VSS sensor specs

Ford VSS sensor - DC coupled square wave of 2.22HZ/MPH
(2.22*60*60=7992 pulses/mile). Minimum amplitude is 3.67 V P-P,
max is Vbat. Minimum load is 15K ohms, maximum load is 1K ohm total
for all devices hanging off of it. It does not want or like external
pull ups.
Likely per OBDII requirements, so this may also work for all VSS sensors
manufactured 1996 and after.


GM MPG calculations

For MPG calculation, a running total of fuel supplied and distance
traveled are constantly transmitted. In the '89 Corvette code, there is a
conversion factor at hex address $C620 to convert cumulative pulse
width to gallons of fuel for the dash reading.
The stock '89 bin (ARAP
and APYP anyway) has the factor 3.6875 gallons per hour programmed in
that location. Since stock injectors are 22#/hr GM is evidently
assuming 5.97 #/gallon fuel density. This is assuming stock fuel
pressure, of course.

----------------------------------------

Calculating GPH flow of injectors :
The slightly more complicated part would be this:

If stock fuel pressure is 43.5 PSI at 100% pulse width then:

FLOW = K * SQRT(DIFFERENTIAL PRESSURE)

The differential pressure would be the pressure difference across the
injector. If we ignore variations in manifold pressure on the downstream
side of the injector and assume atmospheric for that then the differential
pressure is 43.5PSI.

So for stock injectors:

22#/HR = K * SQRT(43.5)

Solving for K:

K = 22#/HR/SQRT(43.5)

K = 3.34

 From that we can now calculate the effect of varying fuel pressure at 100%
pulse width on stock injector flow:

At 52 PSI, FLOW = 3.34 * SQRT(52) = 24.08 #/HR
At 55 PSI, FLOW = 3.34 * SQRT(55) = 24.77 #/HR
At 87 PSI, FLOW = 3.34 * SQRT(87) = 31.15 #/HR

Notice that the change in flow versus change in fuel pressure is not linear.
It is a logarithmic function. In general, doubling the pressure gives 1.414
times the flow.
MS-III can be fitted with a fuel pressure sensor.


there is a thread on the MS-III forum here:
http://www.msextra.com/forums/viewtopic.php?f=125&t=35342
the folks at the msextra forum are friendly an can help out.



<<attachment: krbjhvee.vcf>>

Reply via email to