Techno Duck

I have spent over a week now trying to figure out the cause of this problem and have no luck at all.

The problem is, when accelerating normally in any gear at about 3k rpm the car will go lean, going from 14.7 to about 15.5-16 AFR until about 3200 rpm where it goes back to normal at 14.7. Here is a datalog that shows exactly the problem,



Now i believe the problem is heat related. When the car sits and cools off overnight, the first ride i take the car will not have the lean spot issue. I have driven 15-20+ miles without the problem starting. However even after letting the car sit for 2-3hrs to cool off it still has the lean spot issue...so only after cooling off over night does it seem to go away.

I am not sure if this is related, but its possible it is. On more than a few occasions when letting out the clutch to engage first gear while pulling away from the light... the car has bucked very hard as i was letting the clutch out and giving the car some gas (normally pulling away from a stop). As soon as i clutch in, the car returns to normal idle and will pull away again without issue. Tonight on my way home i noticed the car was not running as lean as it usually had.. about 15-15.5 and in some cases it was not doing it all. After being at a stop light and pulling away the car bucked very hard and i noticed afterwards between 3-3200rpm the car was back to running around 15.5-16 AFR. I do not know yet if these two things are related.. tomorrow i am going to let the car idle (and heat soak intentionally) and engage first gear a few times to see i can duplicate the engine bucking.

Here is what i have checked..

-Checked all wiring for the MAF sensor.. i rewired the switched 12v to the cycling valve plug and even had the MAF grounded to the battery at one point
-Pressure tested intake to 17psi, i found 3 leaks (all of which were minor) and repaired them
-Replaced my new TPS with the old original one from my car which is still good
-Cleaned MAF sensor
-Checked bypass valve visually and also bypassed it (essentially ran with no bypass valve connected)
-Checked FPR by running car with vacuum line disconnected.. still get lean spot
-Ruled out the piggyback and associated wiring by bypassing it
-O2 sensor is new (less than 500 miles on it)... when running with sensor disconnected it runs even leaner between 3-3200rpm. I think this rules out the sensor as its not compensating for the leaned out mixture.
-Clean air filter

There might be more i have checked with i may have forgotten.

Fuel filter was replaced in '09 and has 6k miles on it. I am going to have a spare DME and KLR sent to me to rule those out (thanks Brian!). I do not think this is the problem, but i am willing to try anything right now. The only other thing i can think of are injectors.. but dont get why it would only happen at this RPM range..plus the car still boosts relatively normal even though the tune is off (trying to figure out this problem before tuning).

The only original parts still on the car really are the ISV and fuel pump if that matters.

The lean spot can be tuned out by using a piggyback.

Here is the basic car setup..

Vitesse MAF with V-Flex maps and SMT8 based piggyback
60-1 HiFi
Aeromotive adjustable FPR
CEP fuel rail
Deka 72lb injectors

Anyone have any other ideas? I honestly have no idea when this issue started because the car is driven so infrequently now. John @ Vitesse and i have exchanged countless emails over the last week and a half (probably over 200 at this point) and as of now i think we are both stumped. I have even briefly considered just making the 14hr drive and letting John take a look at it in person... hell ive probably spent about the same 28hrs it would take round trip just screwing around with this car at this point.

Techno Duck

Have not been able to verify fuel pressure yet when this occurs. I do have a gauge but its not long enough to route so its visible. One of the things i actually want to do tomorrow is hook up my HeroCam HD under the hood to watch the fuel pressure gauge so i can see what its doing at that RPM range. That way i can rule out the fuel pump for sure which is one of the few remaining original parts on this car.

I can cruise on the highway (just enough throttle to maintain speed) at that RPM range and it will stay lean as long as it stays around 3k rpm. As soon as engine speed increases or decreases in that range it goes back to a normal AFR.

Stephencs601

Jon, What injectors do you have in there now. I know you have done a lot of work to your car, maybe the DME is not letting enough fuel in at the lower RPM? Or there is a flaw in the chip at that lower end RPM. Or as things get hot, they expand? Just shots in the dark.

Chris White

Can you include the intake pressure trace? That RPM range in right in the transition as the turbo is starting to build a small amount of boost.

The reason it doesn’t do that when first driving is the mixture is still being richened due to low coolant temps.

BTW – try it with the O2 sensor unplugged as well.

fast951

Figuring out erratic problems is never easy.. Investigating a problem, identifying possible causes and eliminating possibilities is time consuming.
The problem goes away, then comes back.. Something is triggering it! Once we determine the conditions under which the problem does not occur and compare them to the conditions under which it occurs, I think we can greatly narrow down the possible causes...
The MAF has been on the car for a few years and about 20,000 miles. Problem occurs in original configuration, so why it started now?
Jon just noticed that Idle's AFRs are leaner now when the cooling fan(s) are on.. Are the two issues related? More things to check and to eliminate..

fast951

Disconnecting O2 sensor made it run leaner as closed loop was trying to richen the mixture.

The turbo and the MAF have been on the car for a few years and about 20000 miles. The problem was not there from the beginning, but a new development.

The MAF signal is dropping and in some logs the MAP sensor is dropping as well (or maybe Jon lifted when he saw the lean condition). More variables...

Van

Boost pressure sensor on the KLR acting up? Or maybe a leak in the boost pressure line to the KLR (and after market gauge, if you have one). Maybe the turbo is just starting to make boost, but the computer isn't seeing it so the injectors aren't delivering more fuel yet...

fast951

MAF sensor signal is decreasing instead of increasing as the RPMs increases(in the 2800-3200 RPM range).. Logically, you would think the engine is sucking air after the MAF (vacuum leak). But Jon eliminated the leaks he found..
Hopefully, he can capture a log file when the car runs fine, then a log when the problem occurs..

Techno Duck

Chris, the intake pressure is 'AMP Voltage', the pink line. During that data log i am not accelerating hard enough to be on boost.

Regarding the mixture, the times this is happening i know for sure the car is out of warm up mode. Also note the problem occurs if i am just cruising on the highway in say 4th gear with the RPM's at 3-3200rpm. So it does not seem to be related to the transition from vac to boost.

I have some more things to investigate today.. hopefully i can get some good data.

Another thing i am thinking is possibly a problem with the alternator. It is original with 126k on it, voltage regulator was replaced about 2 years ago. Perhaps there is an issue with heat soak on it..though i have not noticed any voltage drop from it when this lean spot happens. I will check the volt meter when it does happen again. Possibly tomorrow if nothing else turns up i will borrow a spare from a friend and throw it on the car. Its nice an easy to change now with the a/c delete bracket..

This problem has been extremely frustrating and i am trying very hard not to give into what i have been thinking for a while (even before this problem) by throwing in the towel and selling this damn car. Something non-turbo with OBDII seems like a nice choice right now..

Ronin-951

Just a little clairification please ... New, as in stock with harness, or New, as in substitute with soldered wires to harness. [NoNo, reference Air travels inside the wire core]

Van

The pink line stays pretty steady - but the black line - boost, definitely increases. So, I think the big question is: is the MAF doing it's think?

Techno Duck

The O2 sensor is a universal replacement.. wires were not soldered, used the posi-lock connectors provided with the sensor. I have used this type of O2 sensor on the car before with no problems.

Van, think you may be reading the blue line which is Analog #1 input (MAF voltage going into DME). The black line for boost PSI is bottomed out on the graph.. i did not get on boost at all during this run.

Techno Duck

Here is another datalog taken from today.

Run 1 (left of black line) shows the car running where the lean spot does happen. Run 2 (inbetween the two black lines) shows a run with the lean spot not happening. Both pulls were done in 3rd gear, no boost at all.. just accelerating normally. I drew in the orange line so you can see a baseline for 14.7 AFR and 3k rpm. The only difference between these two runs is the car sat is maybe 10-15 minutes of driving around through my development and maybe 1 minute of sitting at idle outside my house.

On Run 2 it looks like it does go just a touch lean, but nowhere near as it does in run 1.

The teal or brown line shows the wideband AFR. Other than the AFR difference.. nothing seems out of the ordinary i think. I am totally stumped by this. Wish i could provide more in terms of datalogging but it has its limitations.




I would say i am 95% sure the problem is heat related.

Josh B

I can certainly understand your frustration. Was feeling the same way about the issue I had. Hopefully you have some luck working it out soon.

fast951

Some findings, based on the logs. When the AFRs hit 15.4 for a second, I noticed that the MAP sensor is reading lower (higher vacuum). In the logs where we do not see 15.4-15.5, vacuum doesn't drop as much.. This can be seen in the graph above, (AMP input volts - Pink color tracer).
With higher vacuum, you get lower fuel pressure...
With a warmer engine, you are getting higher vacuum at the particular RPM range...