![]() |
|
|
|
Registered
Join Date: Aug 2020
Location: Belgium
Posts: 324
|
tuning the 3.2,
My 86 is 100% original matching numbers. No 'tuning' or anything else to 'improve'. Has 120k miles when I bought it last year. Overall, I was very happy with the engine behaviour. Easy starting, never quitting, good idle etc.
However there were some absolutely minor things that I attributed to "after all, this is a 35 year old car", like: - the dyno showed me 225hp (near factory), but it felt as if they were not really there - once in a while a very slight bogging when leaving from standstill - the warm up sequence from cold brought rpm up to 1100 rpm (as expected), although with an occasional 'pumping' of 100-200 rpm during a few seconds. Using my Bentley manual (absolutely great reference tool) I went reviewing the stuff that the DME (ECU) is listening to for making up his mind what to do: - first the air flow sensor. Many threads describe how the resistor bridge inside can suffer from wear. I opened my unit, and indeed contact traces were clearly visible. ![]() Various threads describe how to 'shorten' the copper slider so it can run along a non-damaged contact path. Many raise concerns that the contact pressure should be respected. What I did is explained in the pictures below. I first measure and draw the angle of the contact wiper assembly as a reference. Then I bent the contacts from 'flat' to 'Z' shaped while respecting the angle of the wiper (distances A and B are identical). This moves the 2 tip contacts inwards, contact pressure remains the same. The carbon contact path was cleaned and the plastic lid glued back on with silicon. ![]() Then I checked the idle switch, the WOT switch as per instructions. Those were fine. As was the Cylinder Head temperature sensor. Next came adjusting the throttle linkage. I noticed that the butterfly opened only 2/3rd. I checked on the underside of the car, and indeed the throttle linkage 'B' clearance was a couple of mm out of spec. Hence reducing the rotation range of the butterfly. It still does not open to a full 90°, but way better now. ![]() The O2 sensor was checked, seemed ok. However, it clearly had it's age so I replaced it too. Finally the dreadful Idle Control Valve. I have checked it according to the Bentley instructions, passed all tests. But it did not look right (can’t explain…) and the cold idle pumping was still there even after having restored the air flow sensor contact wiper. A new one went in. Now ready for a test ride. The car was started, the warm up cycle went immediately to 1100 rpm and then gradually slowed down to a stead idle. No more pumping or surging. Going for the ride, the bogging from standstill was completely gone, the sleeping horses had woken up, torque in the low range was way up. Grin on my face … The effect should not surprise: when the ecu/dme gets garbage info from the sensors and then relies on crappy output accessories to ‘adjust’ the engine no wonder things are not optimal! Lessons learned: a methodic (Bentley) approach, more labour ( a couple of days) than money (O2 sensor and idle valve) can bring us a long way ![]() |
||
![]() |
|
Registered
|
Nice work.
For me, the retracking of the afm and adjusting the idle with a scope to exactly 880rpm did a lot for idle stability and low speed driveability. At 880rpm (sensor output - read with scope) my tach only reads 650-700 rpm so just by looking at the tach I would never have adjusted so ‘low’ bit these steps are crucial to getting it right. BTW - the o2 sensor is supposed to be replaced every 60k miles. Sent from my iPhone using Tapatalk
__________________
__________________ 1985 Carrera 3.2 |
||
![]() |
|
Registered
Join Date: Aug 2020
Location: Belgium
Posts: 324
|
Quote:
The idle rpm was checked with a stroboscope and a marking on the engine shaft pulley. I am at 800rpm (european version). Still: suppose I want to change it, how did you do it? With the set screw under the yellow cap? (notice that the sealed yellow cap is still untouched after 35 years, that in itself is worth respecting!) ![]() |
||
![]() |
|
Senior Member
Join Date: Sep 2001
Location: Chesapeake, VA
Posts: 1,699
|
I love finding quick gems like this to bookmark for general tuning. I am working my way through an 88 Targa that came back to life in the past year after being stored for 13. Getting close to a solid baseline, but am pigeonholing posts like this for next level fine tuning.
After replacing some door and targa seals this weekend, next up is fuel pump check valve, then oil return tubes. After that is sorted, next phase is spring plates & bushings for all rear (new shocks already), and replace right control arm, and front bushings as well as turbo tie rods. This is my therapy.
__________________
Chris 1988 911 Carrera Targa (driving project started JAN 2022) 1970 911E - Long since gone 1972 911 Targa - gone 1987 911 Carrera - gone Retired FA-18C Driver |
||
![]() |
|
Registered
|
This is a great write-up that those of us with the 32 can really find useful. Thanks for posting. Are all of these plug-and-play? In other words, will the fuel system adjust to the new parts on it's own?
|
||
![]() |
|
Registered
|
Quote:
How cool, still with that yellow cap ! Yes, you turn that screw to the setting that is burned into your eprom in the ECU. Make sure the car is at working temp (oil at 90C) , connect pin B to pin C on the test connector and adjust that screw. If you still have the original Eprom it will be 800 rpm if < MY87 and 880 rpm if MY87 or later. Sent from my iPhone using Tapatalk
__________________
__________________ 1985 Carrera 3.2 |
||
![]() |
|
![]() |
Registered
|
The 'test port' near the coil (under the plastic cover)
![]()
__________________
__________________ 1985 Carrera 3.2 |
||
![]() |
|
Registered
Join Date: Aug 2020
Location: Belgium
Posts: 324
|
Quote:
The caveat with self learning/adjusting systems is that they will correct and adjust for systems where the hardware components (inputs/outputs) are not or no longer within spec. That is what happened with my car, it was doing really great except that I was not aware it could do even greater. The system ran out of headroom, and stabilised on a less than optimal point. I got suspicious only last week, when all at once I got some (minor) surging on the idle, simply because the system was no longer able to correct for the next level of component degradation! |
||
![]() |
|