Alfa Romeo Forums banner

1 - 20 of 25 Posts

·
Registered
Joined
·
91 Posts
Discussion Starter #1
Hello all,

Below is a challenging problem that numerous gurus have yet to solve: This is the third time that it has been posted - new info on this one.

1986 Spider Graduate (L-Jetronic) with 109,000 miles. The following describes the problem:

When cold, the car hesitates, as if governed, beginning around 3,200 rpm. As the car warms up over several miles, the "governer" raises its limit - eventually, the problem goes away. There is a sticking point at around 3,900-4,000 rpm, but once the engine is able to rev above 4,000 rpm, it takes only about ten more minutes of driving to have normal access to the entire rev range. However, the first number of trips above 4,000 reveals a sluggish feel as if the mixture is lean. NEW - NOW even when warm, car bucks while accelerating in first gear.

The following have been changed/checked:
1) new temp sensor
2) new fuel filter
3) new in-tank fuel pump - pressure checks out great at 37-40psi
4) new plugs (what the heck!)
5) Auxilliary Air valve removed, cleaned, tested (works great) and replaced.
6) Fuel lines (metal and rubber).
7) VVT Solenoid - re-mounted - works great.
8) 4 new fuel injectors.
9) new cap and rotor.
10) Cat has been cleared out; thus no catalytic converter anymore.
11) 02 sensor has been disconnected with no change, and has now been reconnected.

The following have also been tried: Swapped out (one at a time) the 1) air flow meter and 2) L-Jet computer - no change in condition.

The car is driven daily. Regardless of the weather, it starts immediately, idles perfectly, and, if you are a short-shifter (below 3k rpm), you'd never know there was a problem with the car. The problem is very much temperature-related because the hesitation is worse the colder it is outside and therefore takes longer before it disappears. Once the problem does clear up, it won't reappear until the car has sat for over 12 hours. NEW - as noted above, this is no longer completely true. Once warm, the car will now continue to buck in 1st gear, but will rev fine once going. Different situation to the hesitation when cold, but probably related. Also, the cold-related hesitation cannot be forced through be using the gears to drive the engine beyond the "hesitation threshold." Downshifting to force the revs beyond, e.g., 3,500 rpm, will simply result in a misfiring engine.

Also - please note: this is not a situation with trying to rev a cold engine too hard - I have had a couple of responses to this effect, and this is simply not the case. This car has always been treated with respect, hence 109,000 miles on the original engine, head, and head gasket. Moreover, a cold engine should be capable of revving immediately to redline - this is stupid to do, but the engine should be capable of this.

Any help is appreciated.

Any ideas are welcome.

Thank you,

Drew.
 

·
Registered
Joined
·
598 Posts
I had a similar problem with my 87 Spider. I changed the O2 sensor and it worked like a charm. The series 3 sensor (L-jet Bosch part # 13005) is very expensive, so I bought a Series 4 (Motronic Bosch part# 13957) sensor and changed the Bosch connector to the heating element. You can also get the Bosch universal sensor (Part # 15735). Also make sure that the ducting from the AFM to the trottle body is whole. An air leak in the ducting will cause some hesitation.

Dave
 

·
Trained (ex)Professional, , 1953-2018 RIP,
Joined
·
16,232 Posts
Since the car runs OK 3200 and below, I'd think that one could pretty much rule out a mixture issue (maybe). I'd look at the ignition advance system although I don't(yet) know how (or even if) temperature could be related.
 

·
Registered
Joined
·
4,450 Posts
What are you reading at the O2 sensor with a Volt meter? Are you sure its even working?

Best Regards,
John M
 

·
Registered
Joined
·
598 Posts
O2 sensors can still be a factor. I agree that it is open loop at 160f. However, the sensor is heated to that temp by it's internal heater very quickly due to emmision concerns.

Dave
 

·
Registered
Joined
·
272 Posts
It's open loop to 160degF CLT (coolant temp). The heater is for hot starts for emissions. Warm up enrichment wouldn't work if GEGO was taking place.

-Peter
 

·
Registered
Joined
·
4,450 Posts
Just another thought. Perhaps you will fit into one of the diagnostic categories. Yes or no to the following questions:

1. Engine fails to start or starts with difficulty when cold?
2. Engine fails to start or starts with difficulty when hot?
3. Engine starts then cuts out?
4. Erratic operation during warming up?....YES
5. Erratic idling?
6. Maximum speed fails to be accomplished?....YES
7. Excessive fuel consumption?
8. Combustion misses in all driving conditions?.....YES
9. CO concentration during idling too high?
10. CO concentration during idling too low?

Looking down the manual I see some possibilities:

Loose connections of system connetors and/or lack of continuity in wiring......check for correct connector connection and electrical contiuity.
Good time to put dielectric grease in all the connectors.

Fuel pressure too low........check pump operation and pressure regulator. Check feed line. Check feed circuit filter.
Could be a clogged filter not allowing enough pressure.

Auxiliary air device fails to close.....check device.
Sounds like you already did this.

Air flow sensor is faulty......check electrical and mechanical efficiency of the sensor.
Sounds like you already did this.

Air supply circuit sealing defect.....check sealing
Ah....my favorite. You would be surprised how little a leak it takes to screw up this system. Really inspect the heck out of all the vacuum hoses and well as the main intake hose that runs from the AFM to the intake...and don't forget the intake runners. Make sure that the clamps are nice and tight on the runners. And don't overtighten the ones on the main big hose running from the intake to the AFM or it will deform and possibly crack the hard hose. And be sure to check the hark hose that is on the top of the motor for cracks. Especially on the joints.

Electronic unit is faulty....replace unit.
Unlikely...I doubt it would run at all.

Ignition system (coil distributor spark plugs) not in order....check system parts and replace inefficient ones.
I suppose you could have a problem with a coil or plug wire...but unlikely. Be sure to check those flywheel timing sensors on the bell housing (just for good measure).

I would also let the car warm up and get an O2 reading and see if its running lean at idle or at the 4000 rpm.

Best Regards,
John M
 

·
Registered
Joined
·
272 Posts
A few things:

O2 voltage is still good even when cold as soon as the heater has warmed the sensor up enough to read. You can still get an approximate EGO reading even when cold. The ECU is open-loop under 160degF CLT but the NBO2 output is still valid.

Keep in mind an NB02 is inaccurate. A UEGO (WBO2) reads proper AFR where an NBO2 will read only stoich, rich and lean. Stoich occurs between .485 and .515v. Anything above that is rich, (.8v) and anything below that is lean (.2v). If it reads 0v it could be bad, or it could be soaked. 0v means it's not getting any reading at all. An overrich mixture can gas soak the O2 sensor and fool you into thinking it's lean. IMHIK.

I'm with Jim on this one. I'd suspect ignition. I didn't see replacement ignition components on the list. I've heard of cases with the 75TS that the cap cracked and ran poorly when cold but the heat expanded the plastic and it ran fine when hot. The rotor could be cracked or worn too. Don't overlook the simple things before diving into the engine management.

The ignition box is behind the side cover on the passenger side under the top hinge. There's a vacuum unit attached to it. Make sure the vacuum hose is in good condition and the wiring is in good shape. Crank position is taken from flying magnets on the flywheel. Perhaps it's reading bad RPM when cold. Even a mass/air system needs accurate RPM. Check the VR sensor on the bellhousing and all the associated wiring.

Just a few suggestions.

-Peter
 

·
Registered
Joined
·
3,506 Posts
Greg Gordon's oldebottles troubleshooting page mentions making a connection to the O2 sensor and watching its values as you drive. This was very helpful to me with our '86 Graduate. Poor ignition would do _something_ to the O2 readings, I suspect. One might not benefit much from the info, but you have a potential diagnostic tool in the O2 sensor that you may not have been using.

Some might recommend (might be me) to do this warm-up thing in the dark and try revving the motor up under no-load conditions with the hood open to look for sparks in indiscreet locations in the engine.

Michael
 

·
Trained (ex)Professional, , 1953-2018 RIP,
Joined
·
16,232 Posts
What I meant by ignition is a lack of advance (which would give this no rev/misfire symptom). Easy enough to check with a timing light. Rev the cold engine and see if the advance curve is to spec.
 

·
Registered
Joined
·
85 Posts
Have you checkeced your timing and valve clearances lately? I also recommend checking: Check cold start injector and Termo-time switch (Greg's page has instructions on this i think.) But those shouldn't affect upper RPM range i think. Also look for vacuum leaks, especially the vacuum timing advance device.
 

·
Registered
Joined
·
281 Posts
Hey Drew... The only times (3 separate occasions) when I had similarly described problems with my '86 (R.I.P.), the problem was one of those resistors that go between the plug and the wire. Do you have those resistors? Did you remove them all? Are they permanently attached to the wires? Just my $.02...

-Dave.
 

·
Registered
Joined
·
91 Posts
Discussion Starter #14
To all,

Thanks for the suggestions thus far! A much better response than last time.

Some more info and questions.

As far as ignition components go, the car has a new cap and rotor. The plug wires are very nice blue aftermarket wires that have been on the car for several years (no in-line resistors as Davy13 queried: good suggestion...those buggers can be quite problematic!). When the car is cold, I'm going to pull the wires and check each one for resistance - perhaps there is a bum wire. I just did the "errant-spark" test by revving the engine in a darkened garage (door open, of course), and there is no evidence of any spark jump. I need to repeat this test with a cold engine.

As far as starting goes: hot or cold, the engine fires right up and idles as smoothly as if it were brand new.

O2 sensor - whether connected or disconnected, the sy,ptoms are the same. Perhaps "P.Webb" could provide a bit more pedestrian detail on how to check the O2 readings. The unit is the three-wire model using a two-wire Bosch plug and a single wire with a spade connector. Is the single wire just the "heater" wire? Do you check voltage across the pins in the Bosch connector?

Given the symptoms, it really seems as if I have an overlean condition.
 

·
Registered
Joined
·
165 Posts
just a thought

I had a similar problem that was temp related an after doing much of what you're doing I found that the ground wire on the thermostat housing was not connected
 

·
Registered
Joined
·
3,506 Posts
My '86 Graduate was running too lean due to AFM signal too small. The car wouldn't run well cold, but ran better when warm. Power was not something I was looking for when it wasn't running well, so I can't comment on that other than it was anemic all around. The O2 sensor showed me _as a diagnostic_ that the fuel mix was chronically lean. I pushed more HC into the engine and the O2 sensor signal climbed to reasonable levels. This confirmed lean running to my way of thinking. You seem to have looked at the running conditions of the engine with the O2 sensor connected and disconnected, but what were the signal levels? Mine were about 0.020 V rather then the ~0.500 V they should be. If your O2 signal seems reasonable, then fuel mix should not be a problem.
 

·
Registered
Joined
·
557 Posts
v. interesting

My 2 cents here: A couple things stand out , Plug wires several yrs old? Throw"em out! Might as well put fresh plugs in too cause poor wires will cause plugs to foul. and it doesn't matter what your resistance test shows because you're not megging the insulation which is where the breakdown occurs. I've found in the past that a mtr that runs like yours (improvrd performance on warmup) is usually do to poor plugwires or a mtr w/ a vacuum/compression concern thats satis fied as the mtr warms up and reseals itself and I don't mean a head gasket. I mean worn rings. I f your sensors were bad O2,Temp and such you'd have a more pronounced and immediate starting and low speed symptom. think vacuum. whats sealing itself as the mtr warms? Keep us posted ok? Joe
 

·
But Mad North-Northwest
Joined
·
10,660 Posts
altmand said:
The unit is the three-wire model using a two-wire Bosch plug and a single wire with a spade connector. Is the single wire just the "heater" wire? Do you check voltage across the pins in the Bosch connector?
The two wire connector is the heater. You want to measure voltage between the single wire and ground. Don't disconnect it...just clip one of your leads to it and one to the engine block.
 

·
Registered
Joined
·
272 Posts
The distributor on an '86 is fixed. It has 1 position and the timing can't be changed like the old fashioned points units. The distributor is empty and is triggered from the flywheel flying magnets. The coil discharge is controlled by the spark box in the rear based on RPM from the magnets passing the VR sensors and load from the vacuum unit. The Milano L-Jet had that stupid cold retard that was eliminated by that factory TSB about 15 years ago. Don't remember if the Spider ever had it. As suggested, a simple timing light will tell where the timing is at cold idle and cold revving. The resistors in the wires are for EMI feedback into a transistorized ignition. It can cause misfires if the flyback voltage from the coil heads back into the transistors in the spark box. If this were the case, I'd expect it to misfire all the time, not just cold.

On checking the O2 sensor: Get yourself a nice DVM because you need to read between 200 and 800 millivolts accurately. Put the (+) wire of the DVM to the green wire on the O2 sensor and (-) to a common engine ground. Start the car and note the voltage @ idle. It should be rich during the warm-up cycle and you shouldn't see any bounce in the voltage. When you rev the engine you'll see a rich condition as you come off idle. That's for 2 reasons. First the throttle position switch determines the car is idling and sets the mixture without reading load (AFM). Once you come off-idle, as signaled by the TPS, the load is read from the AFM (mass/air system) and off-idle enrichments are applied. When the throttle is pushed hard, the AFM flap has an overrun in v/s (0-5) that triggers an acceleration event. IOW, the flap shoots open past it's intended position for the airflow which triggers the ECU of an acceleration event. When you put the throttle on the floor, the TPS triggers the WOT (wide open throttle) switch and instantly goes open-loop and WOT settings. At cruise, when the TPS is open, the ECU reads load from the AFM output voltage.

Once the car warms up past 160degF CLT, it goes closed loop and applies GEGO (gamma EGO). This simply means it applies corrections based on the O2 output. Since the NBO2 is inaccurate, it constantly compensates by enriching and enleaning the mixture to a stoichiometric average. In this case you'll see the O2 sensor voltage bounce from .2 to .8v depending on how fast the engine is turning. It will read the O2 voltage every N ignition events so the faster the engine is turning the more often in elapsed time it will read the sensor and correct. This is what creates a stoich to slightly lean AFR during cruise that gives you reasonable gas mileage. On acceleration, the AFR should go to 13:1 or so and stablize back to stoich.

Spark doesn't read AFR but I believe that box has a CLT input that will slightly retard when cold, but I'm not certain of that. I'm more familiar with the hall-sensor Bosch CDI system on the Milano.

See if any of this helps. If it doesn't I have L-Jet wiring diagrams so you can start measuring voltages and resistances from sensors. TPS, CLT and AFM are the most likely candidates here. We can dive into wiring once the basics are out of the way.

Just for fun, do a cold compression test. Sometimes valve and head gasket problems manifest themself cold but disappear once everything warms up and expands.

HTH

-Peter
 

·
Registered
Joined
·
91 Posts
Discussion Starter #20
What year is your car? My spider doesn't have a ground wire running to the thermostat housing. Upon close inspection, it does not appear that it ever did. I have seen other references to such a wire, but the cars at issue were pre-86.
 
1 - 20 of 25 Posts
Top