No Subject

[Peter Vieth]

Well I'm hoping someone on the list can help me out with this, what i 
think, is an EFI problem... it seems to be too weird for the people on the 
forums for my car... (I've tried to be as thorough as possible so this is 
pretty long)

I went through a lot of crap to get my 86 Acura Legend working well.  The 
previous owner had neglected a lot of routine maintenance lately.  The fuel 
system had been contaminated, he'd ignored the PGM-FI light that was coming 
on for who knows how long, etc.  I think I ironed out all the mechanical 
problems, and I got it working well enough to just barely pass smog in 
September, but I knew it still had some demons running around in it (it 
should have no problem passing smog!)...

	I drive home from college every few weekends and the PGM-FI light almost 
always comes on.  (This light is basically the equivalent of a check engine 
light).  Turning the engine off and then on makes it go  away for 20 
minutes or so usually. The code is always for rear or front o2.  Now I've 
switched these bastards twice with Bosch parts (don't ask about OEM... $450 
each, thats more than I payed for the car), redid the wiring for the rear 
one (the harness in the engine bay was iffy).   I've driven around with a 
multimeter attached to the connector at the ECU and the o2 sensors aren't 
doing anything unusual* when the light comes on.  I'm thinking that there 
is nothing wrong with the o2 sensors.  I'd blame the wiring closer to the 
ECU (maybe even in it?) if it wasn't that I get front and rear codes (the 
wires aren't next to each other on the connector, why would these stop 
working and only these at the same time?).

I'm wondering whether some other sensor is giving incorrect readings and 
the ECU is 'blaming' the o2 sensors but, I didn't think that ECUs this old 
(86) check the rationality of inputs, only check for shorts or open 
connections (and through some testing it doesn't seem the Legend's ECU 
'notices'  when the o2 sensor is open, it just goes into open loop* without 
telling you).

I don't know if it matters but the PGM-FI light comes on at almost the same 
locations one way, and usually only in the morning or at night (in the past 
2 months time of day doesn't seem to matter;  before then it never came on 
in the afternoon).  I'm in California so the temperature isn't severe.

 From Redwood City to Davis:
1 A minute after I merge onto 101N, just as I'm passing this one exit

2 Just as I get in sight of the San Mateo Bridge (92E)

3 Right before uh.. this bridge whose name I can't remember (north of 
Oakland) on 80

 From Davis to Redwood City:
Just about anywhere between Davis and Berkeley

It never comes on when I'm accelerating more than lightly.   With one 
exception, it has *never* come on below freeway speed (65mph).  The one 
exception was when I was pulling away from a curb at night.  That one 
baffled me because the engine wasn't warmed up.

The last time I did the Davis to RWC drive I got the PGM-FI light to go on 
4 times during my trip and 2 more times later in the day (I pulled over the 
first 3 times the light came on and turned the engine off then on).  I set 
my Fluke to log the data I was getting, but nothing interesting was going 
on.  The sensor stayed at .7v the whole time.  When I was barely on the 
gas, I could see the o2 sensor goin up and down around .5v.  Anything more 
than light throttle and it would sit around .7v.  To keep the car going at 
70mph I had to press the gas enough to get it to stay at .7v.  I'd expect 
it to fluctuate around this value but it didn't.  And it stayed that way 
unless I abruptly let off the gas (then there'd be a delay, then it would 
drop down to .1v as the engine slowed down, then it would go back up to 
about .5v until i pressed on the gas again and it went to .7v).  The PGM-FI 
light would come on at random times it seems.  Then the o2 sensor voltage 
would still sit at .7v, but when I pressed on the gas I could watch it drop 
almost to 0v depending on how much I stepped on it.  Awfully lean isn't it...

Stuff recently changed:

Intake air temp sensor:  6 months old.  last one lasted 14 years, so I 
expect this one to be good.

MAP sensor.  3 years old.  I don't think it was broken when it was 
replaced; some dodgy smog tech worked on this car and replaced stuff rather 
than checking to see if it was broken.  Tested it a few days ago, and it 
seems to work. (10Mohm idling, up to 30Mohm when floored)

o2 sensors:  5 months old, replaced 3 months before that.  Had to go around 
the harness for the rear o2 sensor because it was shorting to ground 
intermittently.

2 fuel injectors;  all rubber bits; fuel pump; filter;

Other stuff:
Fuel pressure is not steady.  The needle swings very quickly around 35 psi, 
so fast its just a blur.  But its not more than +/- 3 or 4 psi at any 
time.  After I turn off the engine it does stay at 30 psi (for quite a 
while, I've never had the patience to watch it for more than an hour but 
its always down to like 5 the next morning).

TPS:  last 15% of travel does nothing; 0% throttle is sometimes a little 
different (+/- some small amount I cant remember); otherwise, appears to work.


I had a Honda mechanic test all the sensors in September and he said they 
were all good. Hmm.

Some information on the Sterling 827, which uses the same engine (albeit 
2.7l instead of 2.5)
http://www.keast.com/slugman/auto/827sensors.pdf

Input sensors:
Crank Angle
MAP
Baro
Vehicle Speed Sensor
ECT
IAT 
TP
O2 Sensors
Battery Voltage
A/T Shift Position Signal
A/C Switch


*"normal":  sitting in the driveway I watched the o2 sensor behavior.  As 
the car warmed up, the sensor sat at about .7v.  After the car was warmed 
up, it sat at around .5v, going down to .2v and up to .7v with a period of 
about 20 seconds.  This is what I decided was closed loop.  I disconnected 
the other o2 sensor;  30 seconds later the one I was monitoring went back 
to .7v.

Congratulations if you've made it this far hehe.  Perhaps I should stop 
wrangling with this and get the megasquirt in the car...

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