enduro

Hi Folks.

Is there anyone that could give me ideas on what to check next, I'm running out of ideas on what could be wrong. The engine starts immediately, warm or cold. It runs great, idles great and is in general doing anything I ask it to do. However the fuel economy is very bad and the exhaust smells of gas.

The setup when I got the car:

1995 Z28 5.7L. Original engine, long tube exhaust pipes, EGR valve/exhaust intake blinded. 4L60e automatic transmission. Original ECU - no mods on the maps.

Here are the logs: https://www.dropbox.com/sh/322m7ydqf5djc6f/9nC40GwMIS

Logging the car with Datamaster on a short trip I found that all BLM cells I hit (all below 2000 RPM) were very rich. I then added the following stuff:

High flow air filter (no mesh), throttle body air foil. (This was not done in attempt to fix the rich values, just stuff that I had already had) - if anything this should have leaned the car because of possible higher air flow.

Using Datamaster I logged the car under a several hour long trip. What I see from the LOG is that the BLM on the cells that I'm hitting indicates very rich AF mixture, which matches the bad fuel economy and gas-smell from the exhaust. Most of my usage was around and below 2000 RMP with few occasions of revving the engine 3000-4500 RPM. At higher RPM the BLM values look right, but I hit the cells so rarely, I don't know if they would stay that way if I hit the cells under longer period of time.

Due to error code 37 the TCC doesn't lockup, meaning it's slipping (and thus partly responsible for bad MPG), but the 108 BLM is what I'm trying to figure out.

What I've suspected and checked so far:

1. Fuel pressure - 43,5psi and the rail holds the pressure for long time. It drops approx 5 psi when the engine is running.
2. No vacuum leak that I can find, no fuel in vacum hoses.
3. Fuel injectors, I pulled the fuel rail and there is no sign of any leak (tested for 30 min with 43,5psi pressure on the rail. The injectors are stock (part no: 7095004 - 24lbs).
4. Oxygen sensors. The Datamaster log shows that they are reporting through the whole voltage-band (0volts to 1volt). When the engine is running they swing from 5mv to 900mv, just as they are supposed to do, and most often they are well above 500mv indicating rich mixture. As the left and the right side of the engine are behaving very similar (reporting the same BLM values), I find it hard to believe both the sensors failed in exactly the same way at the same time.
5. Checked the spark plugs. They are black/sooted except at the tip where the spark is - there the color is brown. All plugs look very similar.
6. I've used infrared temp gauge to measure the exhaust pipes - all indicate that the engine is not misfiring, the pipes in the front are a bit colder - think it's because of how close they are to the fan on the radiator.
7. MAF sensor - it seems to be working, and I can't see that there is any dirt in the sensor, however I don't have anything to compare it to in order to analyze if it's reporting incorrect values.

I would be grateful if anyone had any tips or ideas on what to do next, I'm out of ideas.

Thanks,

Gunnar

Lonnies Performance

iTrader: (16)

Your diagnostics are great.... you gathered a lot of info.

Your o2 sensors if bad would read low & the ECM would add fuel.
Same with a dirty MAF.

Since your BLM is 108, your ECM is subtracting all the fuel it can.

I cannot view your datalogs, but would suggest checking the MAP sensor to see if is it reading correctly. Should read 98-102 key on, engine off (assuming the weather is normal & you are not at high elevation). You can usually get your local weather readings off the internet.

Get a vacuum gauge & compare to the MAP readings. 0" = 100KPa, 30" = 0KPa, & naturally, 15" = 50KPa. See if you are close to rule out this sensor.

Also look to see if the coolant temp sensor & intake air temp sensors are reading correctly. Either of these reading cold will richen up the mixture.

The hardest things to check are the ones that have not completely broken yet....

enduro

Thanks Lonnie.

I forgot to mention that I have already changed the coolant temp sensor. Both the intake air and coolant temp sensors seem to be correct. The engine is running at 184f (I bought "cold" thermostat) and the intake air reads 105f to 120f. And, I'm at sea level.

I find it interesting what you say about the MAP sensor. I found a log I did on the 11th of august last year. The MAP key on - engine off read 101.5, however the kPa at that day according to historic weather data was just below 100 (and it rained . I will check this tomorrow, but I wonder, if the MAP sensor is failing (as it seems reporting approx 2% higher kPa), shouldn't it be telling the PCM that the air contains a little less oxygen and therefore the PCM should use less fuel? Or am I misunderstanding this completely?

I'll try to get my hands on vacum gauge to investigate this further. Also, I've already ordered new O2 sensors as they can play a major role if they are as you say "on their way to break" :-)

Thanks again.

Lonnies Performance

iTrader: (16)

If the MAP is only 2% off, it will have minimal effect on your tune.

The 108BLM tells me the ECM is pulling all the fuel it can & the car still is rich.

You could try out another MAF sensor to see if it is reading incorrectly.
Do you know anyone you can borrow one from? LT1's are getting more scarce everyday.

I'm not sure what the approximate idle flow should be for reference... but something seems to be telling the car it needs more fuel.

enduro

Hi Folks.

Problem solved (but a new introduced instead I hate to leave a thread "unsolved", so I decided to share with you how this went. The last few months I've been doing experiments on my engine trying to find out what is causing the rich mixture. When I got the car it didn't have the original air-filter or air filter box, but KN high flow filter. As the filter was sitting right next to the MAF the air it was taking in was up to 170f in some cases. I had however not been able to find any connection to intake air temp and how rich the mixture was. When I learned this from logging the engine I decided to build a cold air intake. To my big surprise the first logging with Datamaster I did (yesterday) after the CAI was installed revealed that the engine is not running rich any more. The mixture is at all "banks"/cells most often 128, and only rarely 2-3% rich or lean and I can see that the PCM can easily adjust the fuel pulses to correct these deviations.

The problem therefore must have had to do with the MAF and how the high-airflow filter was affecting it's readings. I had changed the MAF without it having any effect, but my theory is that due to how close the filter was to the MAF air must have been creating vortexes/turbulence that caused the MAF to make wrong readings. Now the air flows after a straight tube for 20 inches before it meets the MAF.

A slight drawback is that the engine now sometimes stutters/hesitates after idle. This has never been a problem before so it must have something to do with the CAI, but this is not causing any problems for me so I don't think I will try to solve it.

Regards from Iceland

http://www.jeppaspjall.is/download/file.php?id=17758

Lonnies Performance

iTrader: (16)

With airflow measurement devices, the recommended minimum straight runs before the device is 5x the diameter & 2x downstream. This is typical in the instrumentation world, but rarely possible in automotive applications. Often the weirdly shaped tubes from the factory have a purpose... to even the air flow out across the device.

Poor inlet conditions will cause all sorts of problems as you found. This is even worse if someone removed the honeycomb air straighteners in the MAF.

Glad you solved the problem.

enduro

Thanks Lonnie.

The first MAF I got had the mesh/honeycomb removed. I got my hands on Z28 and swapped the MAF's. The second one was original with the mesh but it didn't change much. From what I've read it's a common mod to remove it to gain more airflow and some extra HP. However, the way I'm using the engine, HP is rarely needed (but very important when it does

Again, thanks for your input, you obviously know the lt1's quite well

Thanks,

Gunnar