worf928

LH
Poor fuel flow
All eight injectors misbehaving in the same exact way.
The Evil Eye....

What else?

When the MAF was out I stuck my inspection mirror under the intake's skirt to check. It's pretty clean underneath and the breather hoses look healthy and properly connected. There is a leak in the secondary vacuum system (when isn't there?) but that's not going to add enough false air to do sqaut IMO. The car idles rock solid and at the correct rpm. Doesn't die during overrun. No indication of a gross leak as far as I can tell.

Yeah. The gas pressure in the rail is still the ONLY thing that resembles a clue. I've, just a few minutes ago, finished wiring up a gizmo I've been meaning to put together for a while: An ARM1 a/f meter with a plug for the lighter socket, aligator clip for ground, and one 02 sensor connector. I need the harness side 02 connector to 'finish' it. For the momenent the three wires from the plug I do have just have to be fitted over the pins on the 02 connector. There's T for the signal that plugs into the ARM1. It's almost a plug-and-play a/f meter. But, it's good enough for now assuming tests on one of mine tomorrow shows that it works. If it does it will enable a reading of the closed-loop 02 output while the car's being driven etc.

Yes. And after the second failure I suggested that the shop may be the problem. After the fourth failure...

The cats are on the suspect list. Does anyone know if cats need a 'burn-in' period?

ZEUS+

Cats usually need a short break in period to become familiar with the gases. Usually after the smell is gone they become efficient.

WICruiser

To reduce the NOx the cats need to see a rich mixture, that is why the A/F is suppose to swing back and forth between rich and lean. The emission results seem to indicate that the cats are working, just they are only working on the oxidation side (getting rid of the HC and CO) and not on the reducing side (to get rid of the NOx). Is the air pump putting to much air into the exhaust resulting in a "lean" exhaust mixture even though the combustion chamber is not really leaner than intended? This may have all been covered but I just thought a new voice in the mix might open a new thought process.

Bill Ball

Dave:

You can borrow the LH out of one of you cars. Any of them should work fine in the 87. Heck, you'd even gain the ability to use a Spanner.

I'd attach your A/F meter first and see if the A/F mixture is cycling as it should. You probably know it won't do this until the sensor warms up - just a couple of minutes. But then it should cycle from rich to lean extremes about once a second at idle or cruise. If it's not cycling, then I'd swap the brain.

It would be worth running the car for a short while and see if you still get the air/vapor in the fuel rail without heating the rails much. Since the smog test is done with the hood up, I can't see the rails getting heat soaked, so you have to see this vapor phenomenon at lower temps for it to be a factor in the test.

worf928

Good to know. The car's been driven about 500 miles prior to the last test so, I have to assume it's past the burn-in period. The car didn't smell funny when it was here. I found out today that the cats are Magnaflow cats and each one (there are two) is supposed to be able to clean up 500hp worth of motor. So, two should be overkill.

Actually, I have a spare. So, no borrowing required.

What I don't have is a known good MAF except in one of my cars. My spare is core worthy only. Apparently rebuilt MAFs are as scarce as honest politicians right now.

Theoretically yes. Since the 87s have the 7-pin (IIRC) connecter, my spare '91 LH would give a little help.

That's good info. Thanks for posting that. It was useful for my test. I just finished a test run in my '89 with my new ARM1-based Gizmo. It's pretty cool. Feels like I'm driving with K.I.T. staring at me though. I'll bet there's something I could buy that would do the same thing but it was fun to put it together. And it's installable in about a minute and completely reversible too.

Anyway, in my test I did see the voltage (multi-color LEDs if you've not seen the ARM1) cycling from almost-full-lean to almost full-rich. I noted that the cycle time seems to be proportional to RPM; Slower cycle time at low RPM and dancing along at 3k or so. Display shows full lean during overrun (throttle off engine braking) and full rich at WOT. So, I think it works. I can see how the a/f ratio is related to what I'm doing with the throttle.

This would certainly seem to be true if the 'gas' is fuel vapor. If the 'gas' is in fact air getting sucked into system then it might be a time-of-running issue too.

Question: If the a/f mixture is in fact really lean I would expect no cycling? Or will the cycling just tend towards lean? Or do you mean if the o2 output doesn't vary at all it must be the brain?

In essence: how can you distinguish between the a/f ratio being constantly too lean (except at WOT) and a brain with a fried o2 control loop? With a bad LH will the output be nearly-constant? And with too lean dancing around degrees of lean?

ZEUS+

If the O2 is cycling above and below .5 volts it is usually doing its job. Computers are always shooting for stoichiometric efficiency. In a perfect world we would always see .5 volts. Sensors can be lazy and not respond quick enough. A false signal from the 02 could force lean or rich conditions. You can try focing the system lean by creating a large vacuum leak. 02 should have low volts, .1 to .2. Then try forcing the system rich by spraying carb clean in a vacuum hose. 02 should have high volts .8 to .9. 02 should respond quickly to these tests. Just an easy 02 test.

worf928

This would be a test for positive function of the 02 sensor (and my test gear.) I have tested this effectively this evening on my '89. So, I know my test gear works.

So, I'm thinking that if, with this '87, I can get some positive indication, as you describe, that the 02 sensor (should be new) is doing anything then:

1) Any periodic cycling of the a/f reading would indicate that the LH is trying to do something. If the a/f is cycling in the lean range (0.0 to 0.4 volts) then the motor's running lean.

2) If I see no cycling then the LH is suspect.

In this case, it will be a good idea to test, on the failing '87, continuity and resistance from the 02 connector black wire to pin 24 of the LH itself.

3) If I see normal cycling of the volts and indication of 'centered' cycling around 0.5 volts then I shoot the motor in the head...

Observations will of course be made with the car driving down the road under light to moderate load that recreates something like the dyno environment.

John Speake

QUOTE :What I don't have is a known good MAF except in one of my cars. My spare is core worthy only. Apparently rebuilt MAFs are as scarce as honest politicians right now.
>>>>>>>>>>>>

Hi Dave,
928 Intl carry my rebuilt MAFs . They have stock.
>>>>>>>>>>>>>>>

Quote:
Heck, you'd even gain the ability to use a Spanner.



Theoretically yes. Since the 87s have the 7-pin (IIRC) connecter, my spare '91 LH would give a little help.
>>>>>>>>>>>>>>>
12 way recatngular.....


Your idea to check out the O2 loop is good. O2 loop is only active after engine gets to normal water temp.

worf928

And apparently hot off the press? Just ordered.

New information today: The 02 harness on the LH-side of the plug is frayed and showing copper. If the shielding around the black wire is compromised I would think that we might get some cross-coupling from the 12v wire for the heater. Yes? Add a few 10ths of a volt to a 0.5 volt signal and the LH will be constantly trying to lean the mixture out. Yes? No?

ZEUS+

Yes, if 02 is getting extra voltage there will be the false signal forcing it lean as I suggested. This just would not coincide with the emission test results HC will be higher.

Bill Ball

I'd like John to "Speake" to that.

John Speake

It's unlikely there would be a small "leakage". Either the wires touch so O2 output shorted to ground or 12v. To have a high impedance path to have a reduced voltage needs something to give the high impedance path between the two (bare) conductors.

You would need something burnt (i.e. carbon deposit) or some conductive gloop or other to achieve that...

If no LH cycling, knowing O2 sensor is new, and wiring to LH proven as OK, then yes, LH is suspect.

worf928

You lost me here. The emissions results above indicate lean running right? HC 15 PPM, C0 0.01%.

The way it makes sense to me is: Let's say the 02 sensor is reporting 0.5 volts (stoich) but, because of a partial short (say 0.2 volts for discussion's sake) through the damaged harness the LH is seeing 0.7 (rich). Then the LH will drive the mixture leaner until the 02 sensor reports 0.3 volts (lean) and the LH sees 0.5 volts (stoich.) So because of the postive voltage offset the LH is controlling the mixture to a lean condition rather than stoichiometric.

worf928

I have seen such small 'leakage' on my '89. "Something burnt" can be pieces of wiring insulation from a previous "electrical event." The wiring diagrams indicate that the black wire (02 sensor signal) is shielded on the harness side. I would think that parasitic effects were considered non-trivial when they decided to shield it?

If the pin 24 of the LH is shorted to 12 volts rather than the 02 sensor's signal then the LH will most definitely believe the motor is running rich.

John Speake

Yes, if burning is evident, then leakage is possible. But you only mentioned bare copper - abrasive damage ?

I would thing the screen on the signal wire is more to protect against interference signal pickup. But yes, it is a sensitive signal wire.

Let us know what you find.