Fwd: re: trip computers (a thread from last week)

[albrecht@ix.netcom.com]

If there is enough interest, I'll design a circuit to allow user
calibration of this.  Interested people, please e-mail me directly with 
your current "percent of inaccuracy" on your fuel consumption reading.
Please use the format X:Y, where X represents what your trip computer reports, and Y represents what the gas-pump typically shows.  For example, if your trip computer reports 20gallons, and the gas-pump shows 22 gallons, you would report this as 20:22.  I need the info to get an idea of the range of inaccuracy people are seeing.

Thanks,
John A.
84 XJ6
------Begin forward message-------------------------


Last week there was a thread about the trip computers in SIII XJ6 and
same-year 
XJSs, and how accurate they should be.  I did a little checking, looking at
the system 
schematic.  The trip computer merely counts pulses from the ECU and infers
gas 
consumption based on that and what it knows about the dose of the injection
pulse 
(and how many injectors there are).  That's it.  There is one signal line
going from the ECU 
to the trip computer, after passing through an interface box (buffers the
signal for the 
trip to the front of the car).

The ECU _may_ also care about pulse width but I haven't explored that yet. 
If it doesn't 
this would of course be a source of inaccuracy.  But the whole thing is
only 'sort of' precise, 
anyway, given that's it's really a novelty feature, introduced in a year
when many other 
luxury cars were doing the same thing.  Cheap and available
microcontrollers were 
hitting the auto industry in those years, and auto manufacturers were
anxious 
to show their technical leadership with such a visible device.  Per Chad
Bolles, 
5-10% accuracy is what he sees, at best.  Mine seems pretty accurate -
certainly no better than 2-3% 
though - I'm doing a full-tank(s) test this week and will report next week.

There's no direct facility to improve the accuracy - no calibration pots or
anything.  Per 
Haynes, that oracle of automotive wisdom (!), a faulty interface module
could be a source 
of inaccuracy, but they don't explain how badly inaccurate it would be. 
I'd expect it 
to be totally dead.

David
'84XJ6 VDP
'70 2+2 E-type

------End forward message---------------------------

---

• Prev by Date: 82 xj6 -U-joitn replacement
• Next by Date: New Member Introductory Posting
• Prev by thread: 82 xj6 -U-joitn replacement
• Next by thread: Fwd: re: trip computers (a thread from last week)
• Index(es):
  • Main
  • Thread