Running issues with '88 Testarossa.

Robert - perhaps this may be of help. I suspect that the flexible rubber (black) tube that you found going into a cannister that had vacuum pulled through it is the vacuum assist for the baffles in the air conditioning / ducting system under the dash. If you remove this black tube from the cannister (reservoir) located over the left rear wheel, the air venting in your passenger compartment will no longer work. That is, the ducting will not change when you press the A/C buttons from defrost to dash flow.

These ducts are vacuum assisted.

The important issue is that if the vacuum tube is removed or violated, then air will leak into the manifold of the left bank (that is where the vacuum is generated). This can lead to a mixture problem. I know this because at one time my ducting failed to work (defroster would not work). Upon finding the source of the problem, I noted that the end of the black tube had become dry and cracked, and was no longer creating an air-tight seal with the cannister. It was sucking air thereby eliminating the negative pressure. I clipped off an inch of the tube to come to fresh rubber. Problem solved.

Hope this helps.

Jim S.

 

You must have the injection ECUs plugged in when you measure for the 8V at the airflow potentiometers -- so your getting nothing when they were unplugged was the correct result, but it's odd that you didn't get the 8V when you plugged them back in. The injection ECUs make the +8V out of the same +12V on the red wire at the water thermoswitch so you might go back and confirm that you've still got the +12V on the red wire (I say +12V, but it's really the output of the alternator so the ~14.5V is OK for what I'm calling "+12V").

 

Rob – I emailed you Fig 3 (the injection system) from the 459/86 TR wiring diagram for your model. It is so wonderfully simple compared to the US version , but it has the exact same architecture for the wires going to the airflow potentiometers so I can’t see how it should work any differently. It is true that I only have Diagnosis Sheet N. 6 (for the US TR) and your model is actually covered by Diagnosis Sheet N. 5; however, I have reviewed the actual schematics of both, and, in both cases, there is no reason why having the ignition key “on” or “off” would make any difference (as long as you’ve got the red wire at +12V). In Diag. Sh. 6, it does say “injection ECUs connected and ignition key ‘on’” for the measurement conditions at the airflow potentiometer wires so it wouldn’t hurt to retry with the key “on” (but I’ll be shocked if it makes a difference -- my guess is that “with +12V power applied to the injection system” -- i.e. red wire at +12V got confused into “with the key ‘on’” somewhere along the way).

I’d be very surprised if you had the exact same fault inside both injection ECUs, so you might want to try confirming that +12V power AND ground is actually reaching the injection ECUs. Having the red wire at +12V is a very good sign that the +12V is reaching them because the food chain is:

+12V from battery -to- relay C terminal 30 -to- relay C terminal 87 -to- 1/6 injection ECU pin 1 -to- 7/12 injection ECU pin 1 -to- 2-way splitter -to- water thermoswitch (and the other branch goes to the throttle microswitch)

Since you are making the +12V measurement at the end of this chain, there’s a good chance it’s OK at the ECUs IMO. Checking the ground is a more difficult situation because the food chain is:

Ground connection at engine -to- 3-way splitter where:
One branch goes to relay C terminal 31
One branch goes to 1/6 injection ECU pin 2
One branch goes to 7/12 injection ECU pin 2

Since your relay C is working, that confirms that its ground connection isn’t horrible, but it really doesn’t guarantee that the other ground connections are OK (if there is a problem at the 3-way splitter). Bottom line is that what you’d like to measure is the voltage between pin 1 and pin 2 right at each injection ECU with everything plugged in and while the red wire is at +12V (or the engine is running) -- should be +12V -- but this is difficult without the special interconnection box. I can’t recall if you can remove the back connector shell enough on the 25-pin connectors to make a measurement while they are plugged in.

If you do have +12V power between pin 1 and pin 2 at the injection ECUs, but no 7~8V between pin 18 and pin 14, that would be a bad sign IMO.

PS For the Throttle Microswitch Resistance Test, the “140 ohm when cold” part only applies to US version TR. On your model (KE-Jet without Lambda), it should be 0 Ohms at idle and infinite ohms off-idle regardless of cold or warm conditions – sorry for the confusion (and yours seems OK).

 

I forgot to add that you can try making an easier-to-do resistance measurement to see if the wires going to pin 2 on the injection ECUs are a good ground (but I think that the functional voltage measurement with everything plugged in and actually working is a more reliable test for determining good vs bad). To make the resistance measurement:

1. key "off"

2. unplug both injection ECUs

3. unplug the large round C12 connector from the triangular black box

4. measure the resistance from the female metal pin 2 terminal in the (unplugged) 25-pin injection ECU connector on the harness side to the engine block or cylinder head -- should be 0 Ohms (or a few tenths of an Ohm maximum).

5. repeat step 4 for the other 25-pin injection ECU harness connector -- should also be 0 Ohms (or a few tenths of an Ohm maximum).

 

Certainly, and it's probably the same for every version KE-Jet TR (I've made this same measurement on my TR twice before, but obviously forgot the details so am relearning ). There is no direct connection from a key powered +12V source to the injection ECU (on any KE-Jet TR) so it must have enough smarts inside to look for the presence of another signal that is based on the key being "on" -- like the tach signal from the ignition ECU being present (even if "0") -- just a guess.

Those translations are a little misleading. Both relays are only involved in firing, or inhibiting from firing, the cold start injectors (not the general A/F ratio during warm-running).

Yes, it's sort of bad news when everything measures OK.

 

Vacuum leak...intake manifold...control tube...brake assist?

Jim S.

 

This is a very confusing thread to say the least. I agree with jselevan in that it could be a vacuum leak problem. But why only at such a specific rpm range? What specific conditions are occuring at the rpm range with the car warmed up and running down the road? What sort of vacuum problem would affect BOTH cylinder banks? Could the vacuum problem be intermittent, not always there, but happening only under certain conditions? Did the mechanic know of this "surging & bucking" condition when he set the fuel mixtures? Sorry to be so vague, without any direction....just tossing out some random questions.

 

Rob -- You'll need to make your own decisions about where to go next or what to do (some things are not a trivial amount of work so I'm uncomfortable saying do this or do that). You've completed the initial electrical tests that check the various input signals to the injection ECUs. The more difficult electrical tests are checking what the outputs of the injection ECUs are doing under various conditions -- i.e., measuring the voltage from pin 10 to pin 12 (or the current flowing from pin 10 to pin 12), but this can be dangerous -- if pin 10 gets shorted to pin 12 (even momentarily), it will blow up the injection ECU (this is where using the special interconnection box is much safer than a manual probing method). I don't know if the pin 10 to pin 12 tests on Diag. Sh. N. 6 are exactly the same as those on Diag. Sh. N. 5 for your model, but I think it would be better to do some of the simple tests on the other systems before digging into the difficult electrical tests (and you may need to get a copy of Diag. Sh. N. 5 if you want to do the more advanced electrical output tests):

Ignition -- why not put a timing light on it and make sure the idle timing is OKish and that the advance seems to work OK? It's relatively easy and quick -- unlike a 400i .

Intake -- per Jim's suggestion, study the places where (unwanted) air could enter the intake tract (without deflecting the airflow metering plate) and try to make sure things are OK -- no cracked hoses, the brake booster will hold a vacuum, etc..

Fuel -- the advantage of the KE-Jet system is that it doesn't have a warm-up regulator so there's really only one critical input pressure to measure -- the regulated supply pressure. I have to agree with Jeff's sentiment that it's hard to see how a problem here could affect both banks simultaneously, but it wouldn't hurt IMO to confirm the regulated fuel pressures are in spec. The more advanced fuel measurements involve measuring the actual discharge rate from each injectors (at low and high flow rates) either using the factory rotameter device, or I've had reasonable results with 100mL graduated cylinders (getting about 20mL in 3 minutes holding the airflow metering plate in the approximate idle location, and about 40 mL in 1 minute holding the airflow metering plate down 5~6 mm -- and you could do more intermediate points):
Image Unavailable, Please Login
Image Unavailable, Please Login
Welcome to Gremlin hunting -- it's not easy

 

What is the air-fuel ratio at idle? I see where at cruise it was very lean, and on acceleration that the mixture level never got down into the 14 range. If the a/f ratio at idle is also lean, then perhaps the actual base line mixtures are off? If the car had O2 sensors it would be great to see the signal values being sent to the ECUs, but unfortunately that is not the case here.

I'm thinking the only way to get to the bottom of this is to start from square one. First verify that the air flow metering plates (engine off) are in the proper position in the barrel portion of the air diffuser housings. Next would be to verify throttle plate actuation, they must be in sync with each other, not just close but perfect. Check the routings of all vacuum hoses and their condition. Is there a way to put vacuum to the Ingition brain and measure the output changes with the car sitting still? Unplug and inspect every dog-gone electrical connection, including the fuse panel, and re-plug everything. Verify that the auxiliary air valves are working properly. If I can think of any other nonsensical task to perform...I'll post.

Accept my apologies for rambling on here. But I would like to see this problem solved. The reason is that about a year ago I read through Robert's postings about the electrical system fixes that he has done...and using that information I was able to fix my own Testarossa's mother board. So I would really like to see a successful conclusion to this thread....Robert certainly deserves it.....

 

This is really puzzling, and as such quite interesting. Perhaps more so for us than for you trying to fix it. Silver lining is when this is done you and Steve will be the top world experts for the TR ke fuel injection. But it seems to me that you may have two problems:
1 Car runs way too lean
2 car bucks at a well defined rpm (and throttle opening)
The two may or may not be related. The strong dependency on rpm may indicate that a connection or wire enters into some sort of resonance and either opens or short the circuit. If that is so you would have to make all the measurements with the engine reving at 2500 rpm!!! Could be an old solder, a kinked wire, corroded connector. Could even be in the ignition. If you go through the critical rev range quickly the resonance has no time to get going. For example if the wires in your flywheel rev sensor vibrate and short (or disconnect) themselves at 2500 rpm that would affect both the ignition and the fuel injection through the tach relay which would kill the fuel pumps I think.
Could you trick the ECU into thinking the engine is cold which would tend to enrich the mixture and see if the bucking problem remains?

 

Robert...regarding the AFR at idle of 15.8? Are you saying that no matter how much you adjust the mixture screws the AFR stays the same? Or is the AFR staying at this level, while your doing everything BUT adjusting the mixture?

This next bit of advice comes with a disclaimer...."use at your own discretion".....

IF EVERYTHING checks, all electrical and mechanical components, with nothing apparently wrong. Then I would --slightly-- tweak the mixture a bit more rich. First off, stick a wrench into the mixture screws and use some method of marking the position of the wrench/screws in their current position. This way you can go back to a baseline if you get things messed up. Now turn both adjustment screws about 10 degrees clockwise from their current position. Then drive the car and see what happens. If things improve try another 5 degrees turn and see what happens. The mixture adjustments should be done in VERY small incremental changes. An AFR of 15.8 is too lean in my opinion, you want to be in the 15.0 range at idle. At wide open throttle an AFR around 12.8 to 13.2 is okay - again just my opinion.

I'm getting the feeling that the problem is AFR related.