'86 3.8l vin 3 performance & fuel economy poor

Okay, this problem is like the plague. As I figure new things out, I will be updating as I find it neccessary. I seriously can't be the only one with this problem. This is USEFUL information for anybody that has A, C, H, or G (turbo 3.8 only I believe) body Buicks. I give some background info as it may be important. If I actually solve this problem, it may be worth linking this thread to the Century, Park Avenue, and Regal forums.

My LeSabre used to get an average of 30-31 mpg consistantly, pretty much reguardless of my driving habits. At one point, we were having some screwy issues with the computer, ICM and timing chain resulting it a no-start situation with spark, fuel, and full compression on all cylinders. Come to find that the nylon timing gear had broke. However, we also learned that no matter what, new ICM's would NOT work, or work poorly. We also had to run the car, shut it off after a few miles driving, change driving conditions, restart. etc. Almost like we had to spoon-feed data to the computer so it could 're-learn' the data from various sensors. All this just to get the ICMs that did start the car to not mis-fire and try to diagnose the problem (ICM was first replacement, then computer, then found the TC was bad, then had to learn how to get car to run again...). The end result was a remanufactured computer, a timing set, and the orignial ICM was the ONLY combination that would allow the car to run properly. Even at that point it was still getting 30 mpg.

Now after the tranny rebuild and sitting for a long time, I get 20. No SES codes. After about 30,000 miles of troubleshooting, tweaking, I finally squeased 21.8 mpg driving all freeway when the engine spun a bearing. Yay! We got a temporary replacement out of a 87 PA, computer with prom for the lesabre version, and the updated sensors. After much troubleshooting, I could get nearly 25mpg all freeway. Then the oil pump crapped out and the dummy light doesn't illuminate until basically 0 psi so never had warning...Yay!

Currently, have completely rebuilt the 86 original engine. I did everything but the machining. Bored .03 over, crank .03 over rods .01 over mains. Original camshaft, not worn down enough to justify replacement. Heads rebuilt too. Anyhow, new sensors everywhere, new Delco rebuilt computer, new coils (BorgWarner) new wires and plugs. The rebuilt 440t4 still in great condition. Still runs like crap, random intermittants, etc. Test every wire on computer harness for conductivity, shorts to ground, shorts to other wires. No problems. Found oxidation on some connectors, cleaned all contacts on all connectors anywhere in or on the car with wire wheel. I eliminate ALL intermittents that had been occuring since the tranny rebuilt. Have been doing testing with different proms too. Still stuck at 20 mpg, also have noticed that the power output doesn't seem to be increasing with rpm (stuck at 100 hp and 150 torque from 3000rpm to 5000rpm for example).

I have a chassis service manual, a DMM, and plenty of free time at the moment. So, I check all the voltages at the computer. Find some anomalies. Can't seem to sort them out. Learned that the wiring colors on my LeSabre are wrong...they match the wiring colors from the A body diagrams. They all go to the right places though. My IAC votage is backwards at key on engine off, but correct when engine idling. My MAS voltage bad when engine off, but normal when idling; second 86 MAS operates identically. VSS signal bad always, even though guage works and computer data through aldl shows correct speed, signal must be a pulse instead of a steady voltage? TCC control voltage bad always, though TCC operation verified as good. 4th gear signal always bad, but TCC works. Crank ref. LO* and HI* from ICM bad. Crank sensor code not setting, and motor starts. EST bypass voltage cycles, should be constant, when running see up to 10 degrees of retard at 5000rpm . EGR diagnostic voltage bad always, showing batt volts when should read in millivolts, though vacumn guage proves that EGR properly functions.

Computer verifies that it is operating in Closed Loop. Coolant temp sens. reads same as aftermarked guage. No SES light or intermittants. Currently using the injectors off of the 87 engine, as they had a better misting spray.

Before we put the 87 engine in, my orignal ICM died. We found a newer ICM in a car at the junkyard, so used that. Replaced it with a type 1 on the 87 engine because thats what the computer we got was set up for. We were using the type 2 that was originally on the 86 when I first rebuilt.

This is the complicated part, and I haven't had an opportunity to re-check voltages noted by the * above:

In the spirit of proprietary design, I noticed something major the other day. My 86 LeSabre wiring harness is meant to be backwards compatable between type1 (magnavox, 1 big coil) and type2 (Delco, 3 separate coils) ICMs for that year, depending on ( I believe) the PROM. Since we've had the engine running on the original type2 coil setup, there has been a shudder at 35mph to 60mph in third gear accelerations only. Wasn't an issue with the type1 coils though the 86 PROM I had was for type2. SO, we go to the local Pep Boys and get a new Borg-Warner ICM. This unit is the same as on a 91 Olds Ciera 3300 we have (this is what got me thinking to begin with). It works PERFECT on that car, but does the same shudder as the other type2 we have on the Lesabre. Then the ICM dies after less than 30 minutes of use. 1 exchange later, we are running again, but same shudder buisness as the other type2s. After scouring over diagrams for the ICM setup, I realized that the type1 wiring is identical to the type2 EXCEPT that the type1 draws a +12v signal from the ecm fuse. My wiring harness has all 14 wires in it. The type2 doesn't use the circuit 439 pnk/blk (pin P on ICM). We have had nothing but problems using aftermarket ICMs... I pulled off the connector, and find that the aftermarket ICM has a pin in that posistion. I know that my original ICM did not. I disconnect the wire from the harness, because the diagrams clearly state that for type2 ICMs, that wire is not used. I'm sure that the replacement models back then didn't have that pin in them, but the new ones do. From a compatability standpoint, I can only guess that the pin is used to switch between settings internally so the ICM functions properly on the newer cars that use that pin (ones that only have 6 wires going into the module) like our Ciera.

As of this point, the shuddering has dissapeared. I haven't had time to do much more testing. Working on a MPG figure. Power seems to be possibly a little better, but still not quite where it was. I believe that was most if not all of the problem the entire time, but I won't know until I get some mpg numbers and some driving time for the computer to re-learn all its adjastable parameters (got to love those transition vehicles...Anybody with a 94-95 LeSabre would know what I'm talking about. OBD1 computer on a OBD2 interface, and no way spare the dealer to read SES codes).
I do have a custom cold air intake and a K&N filter on. I have an aftermarket Merit muffler which seemed a little smaller than the one on it previous but it was the only listing for the LeSabre. I want to disconnect muffler to test if it is restricting my exhuast too much, causing the lower than expected power at WOT. If this is the case, I'll be finding headers, a crossover to mount the O2 sens, and dual mufflers (not just dual exhuast tips like what I've got now). BTW, MAS reads 2-4 g/s at idle, and up to 110 at redline. Using winALDL to test and verify sensor outputs, etc. Anybody is welcome to post their thoughts here as well.

Have you checked the harmonic balancer for separation of the rubber ring and reluctor from the center section (very common issue on early 3.8's)? Having 10 degrees timing retard at 5k RPM shows me either the balancer is completely separated throwing off crank sensor signal, or its partially broken loose causing a rattle, and the knock sensor is retarding the timing thinking the car is spark knocking (pre-ignition). This would explain both the lack of power and bad fuel economy.

Are you burning your own PROMs or using factory programs?

No, don't have the tools.

I've 'retrieved' several from a local salvage yard for the different body styles that use it. This is the list of the ones I've got: CLD, ABYR, ABYS, ABYU. To the best of my knowledge, the ABYS is for the '86 Buick 3.8-vin3 4-door limited type2 ignition 440t4 trans. I failed to write down what the others are from because I had gotten them with the 1227148 computers I was trying back when I had the timing issues and wasn't planning on actually using them.

I did get a number for fuel ecomomy again. I'm averaging just barely over 20mpg...shudder still gone. I believe the harmonic balancer is fine, but I'll have to look at it again. Last I remember it didn't have a rubber ring on it, thought the 87 balancer does. They are different styles from what I could tell, but no surprise here if that's been the problem all along.

Latest update.

The exhuast system is not the cause of the problem. Harmonic balancer is not an issue either.

I've cleaned and re-installed the 86 injectors, and found that one pair of connectors may have been installed (by me) backwards, but changing the injector they are one results in no change in the way the car runs. Also, with the installation of the 86 injectors, the shudder is back. This leads me to believe that the fuel injector harness may have all the wires hooked up backwards or something. Still retaining the symptoms where the torque and horsepower feels like they are dropping off when the RPM increases, as if there is no spark advance. Still no signals on the crank ref. high/low wires on the ECM wire harness.

Okay,

Fuel injector wiring is good, all injectors good, again no change in how the engine runs when the injectors are all hooked up backwards.

All sensors check out as good.

Still no signal on the crankHI crankLO circuits, between the ECM and ICM.

Disconnected the knock sensor and found that the spark is still being retarded by 10 degrees, and absolutely no affect on fuel economy or performance.

found identical car in junkyard, but with the vin B 231 and type 1 ICM, and prom has a 3 digit id.

If I knew what PROMs were for which cars that would help me out significantly. I can't find anything online or at auto parts stores, and dealer won't help. Having problems finding exact match in junkyards for my car to pull ECM and ICM as a pair from.

Click Here (http://www.cruzers.com/%7Eludis/c3xref.html#1227148http://www.cruzers.com/%7Eludis/c3pinouts.html#16032500) & here (http://www.cruzers.com/%7Eludis/c3pinouts.html#16032500) for ECM & prom ID info

Changed to a different PROM...

Still dies immediately after starting, also MAS shows 2g/s when motor has not yet started, and 4g/s when idling around 650 RPM.

The engine power seems to fade in and out under 60-70% throttle.

Also, any higher thottle than that the power seems to completely cut off and just increases RPM to the shift point.

Still have the odd shudder in 3rd, but because the prom has a more aggressive TCC lockup timing (stays locked from 32mph on up under med. acceleration vs. the unlock experienced before when shifting into 4th) have noticed the shudder immediately after shifting into 4th with the TCC locked.

When shifting, see a 14 degree spark retard that goes away. When the shudders occur, get up tp 15 degrees. Also, under WOT have seen 22 degrees of spark retard.

I'm going to get a factory installed Delco ICM as soon as I can, and see if that makes a differance with the factory installed computer I found this weekend at the local junkyard. The car I pulled it from was 99% identical to mine (different color), and it has a different PROM in it as well. Same ICM, Motor, trans, GVW, same options. I wouldn't be surprised by any result but will keep this up-to-date.

Recently noticed that even though I put new contacts in all the connectors, the TPS connector was loose and causing the voltage to creep up to 1.1v-1.6v at 0% throttle. Have patched that up again.

Latest Update:

I believe that the transmission may have been the culprit behind the shuddering as I dropped 3rd gear last week. I got it to shift directly from 2nd to 4th (heavy acceleration), but when I stopped again it locked up, and couldn't go anywhere. Motor wouldn't spin "freely" though still ran. No matter if in D, P, N, or R it wouldn't budge. So, I started testing some of the other gear ranges (3,2,1). Pulled the gear selector down into first, hear a familiar hissing sound from the trans, so I floored it and BAM! unlocks and takes off again. shifts between P,R,N,1&2 fine. Lolz It's going to get parked until the weather breaks. After making the repairs I will have to re-evaluate the performance and fuel economy.

And the saga continues...

Don't take this the wrong way, but you are putting an incredible amount of money and effort into a car worth less than $1000. Why not part ways and just buy another car?

-Tranny Master Rebuild kit - $150

-10 hours labor

-brand new Delco OEM ICM (speculating to the fix to the problem i'm having) $190


If you can find a virtually rust free, immaculately clean and perfect interior, GM car with the 3.8l (30+ mpg) that has less than 10k miles on the motor or tranny in 10 hours for under $340...send me a pm.

Besides, I can't get a new car until I find out what the cause of the problem was. If I did, the car would win...

been busy with school, work, and thesis project but here's the newest updates

- now have 21.045739845 MPG!!!! Thats a 1.045739845mpg increase!
so, after having the heads on my brand-new rebuilt engine fail (valves on 3 cylinders leaking enough to cause misfires) I got the heads off my backup dead engine and $85 in 3 gaskets later it runs again, but now with the 1 mpg increase. This increase may be due to a new knock sensor or the newly replaced used heads, not sure.

- factory computer and factory ICM = shoot and a miss. Did nothing to help or hurt.
- tried using forced-induction during start up with high-flow fuel injectors (from a SC 3800) and achieved the first start in a long time where the motor didn't immediately die. However, injectors too rich to run NA.
-Serial data still shows 2 g/s mass air flow rate with engine off...still dies immediately on start up unless I give it a little throttle. Several MAF sensors show this same condition, including a brand new sensor.
-Runs like crap when temp gets lower than 15 degrees, barely starts.
-spark retard increases with RPM, always present; kind of scary...
-dual exhaust sounds awesome with cold air intake. The stock exhaust manifolds for this car are actual headers; explains why I couldn't find any anywhere.
-feeling more vibrations again that appear to be due to a broken motor mount

I am working on adapting my serial interface through a USB port as I no longer have access to a laptop with a serial port. I am toying with 2 ideas, the first is that the original ECM was proprietary in how it interpreted the PROM even though the PROM is recognized and the new computers think that they are working. The second is that somehow, someway I managed to put the camshaft gear on offset by 180 degrees, and that difference is screwing up the computers timing (unbeknownst to the computer) due to when it receives the cam signal in reference to the crank signal. Both scenarios may explain how flipping the injector pairs do not make any difference whatsoever as the injectors may not be firing sequentially.

Yay happy weirdness that I can't explain.?>.?

I have aquired a computer 1227057 from a 87 Riviera and am now using that, partially because it runs on the 8192 baud ALDL interface as opposed to the 160 baud of the 1227148. I have also been informed that the replacement computer purchased wayyy back when did not match the computer number on the original, but the sales person said it was the only computer in the system for that car/engine (we were sold a 1227148 computer when the original was not a 1227148, but the PROM from the original ECM works in the 1227148 so am trying to find a non-1227148 ECM that works with my PROM). The original ECM number is unknown at this point as it was turned in for a core so very long ago.

The new computer sets a SES light, but I cannot check it? I don't have a pinout to double-check that I don't need to swap some wires on the connectors, but when I short pin a to pin b on the ALDL con., the SES light just blinks very very fast. I believe that I need the body control module off the riviera and/or the CRT to get the codes so i'm kinda screwed to find out what code is being set (I suspect it is for the ESC module).

New computer makes the car sound different, but initial fuel economy test showed the same 21mpg until I tried a 93 shell gas with Lucas Octane Booster. I have estimated 6 gallons in the tank at 330 miles traveled...I seriously thought that the gas guage stopped working. Unfortunatly the performace is still not right.

So my latest theory is that the cam sprocket for the timing chain is manufactured to be compatable across several engines and isn't marked properly for the vin 3 motors. As a result, the side of the sprocket that reads 'front' should read 'back', which puts the guide hole for the cam magnet on the other side of the bolt hole effectively retarding or advancing the timing by at least 10 degrees. What do any of you think about that? Would the motor even run? Like I said, it cannot start cold without keeping my foot on the gas for about 4 seconds after it first fires and the exhuast has a kind of spitting tone to it and is un-even.

also it has been determined with 6 LED's that the fuel injectors ARE firing sequentially at all times (starting cold, running, accelerating, etc.) and that the injector pulse increases with increased throttle, though limitations of the LEDs prevent conclusive readings that the injector pulse gets longer at any throttle above 60%.

Hi did you replace the O2 sensor with one from AC delco or Delphi?

The O2 sensor was replaced with a Bosch replacement from Autozone, but I still have the original sensor for the '86 and an original from an '87.

Hello again,

Please be advised bosch has two factorys W germany and CHINA
and I think the auto maufactures get the genuine parts and guess who get the chinese parts,has your MAF sensor been replaced yet? F/inj. Map sensor
on carb models, on cars vehicles with map sensor the signal can be inverted or have over 4 v (wot) at idle and not set a code.
I noticed you changed ECM's But do you have the correct prom installed verify by calling the dealer and give them the vin# to confirm you have the right chip for your application,does your SES light (key on) off (engine running)does it ever light while driveing,are there any DTC's. is your TPS set correctly to under 1.0v K O E O.Check your Tailpipe is it Black and very sooty or light gray which indicates proper combustion.have you had it checked on a exhaust gas analyzer to check your co% and Hc ppm emissions
is your
converter (cat)original if not you should replace it.those are area's of concern.also 25 + MPG is not that shabby.
Regards,

So 22.2mpg on the new computer...

I seem to have figured out the cold start issue, but need to make a few more cold starts to verify it wasn't a one-time deal.

I disconnected the EGR solenoid from the EGR valve and when I started cold it didn't die and ran a lot smoother. It may also have increased power but like I said before too soon to tell if it makes a real difference.
Also have verified that I didn't put the cam sprocket/magnet on backwards according to how other kits have it lined up, but still trying to get a picture of the original sprocket and how it lines up with the crankshaft interupts just to double check.

Exhuast system does not have any flow restrictions.

If I can figure this out I'm going to find a way to modify my spare motor for this car to fit in a Montana (3400 sucks)

Does anybody with a timing light have a properly running '86 or '87 vin 3 that they can give me a location of the timing mark at idle? I want to make sure my mechanical timing is where it should be...although if it were'nt I think the valve train would have been obliterated by now.