SSTAT

Hi all - I have a 97LT1 stock original fuel system. Its a road race car so I had to take it in for routine dyno certification. last dyno 6 years ago the AF ratio was perfect at 12.5 throughout. This time power is exactly the same but AF went up to 14-14.25. I checked for vacuum leaks, fuel is a little old but not enough for this I think, so Im checking my pump, here are the numbers:

With the fuel pressure regulator still hooked up to vacuum:
Idle 40psi, at throttle tip in it goes to 43 for a split second, then falls with more throttle to a low of 32 above 3000 RPM.

With the fuel pressure regulator unhooked and vacuum plugged:
Idle 46 PSI and no change at all with throttle.

It still has the original fuel pump and regulator. So I think this means the pump is good and it needs a new regulator? Im going to town to get a new fuel filter just to see if it makes a difference. Thoughts?

SSTAT

New fuel filter made no difference.

SSTAT

Is there something I'm not clear about? I'm just trying to find out if the problem is with the pump or the regulator based on the results of my testing. Thanks for any help!

Injuneer

Unclear.... we’re the first test and second tests performed with the engine under load, or with the engine idling in neutral?

With the engine under max load, above 5,000 RPM, the fuel pressure should not drop more than a few PSI below the “no vacuum” value.

How much the pressure drops with vacuum at idle depends in part on how big the cam is. Stock cam vacuum will typically drop the fuel pressure about 8 PSI at idle.

SSTAT

Thanks! I gave results at idle and revving the motor (in neutral) both with and without vacuum above. It fell to 32 while revving to anything over 3K with vacuum, without vacuum it was 46 and didn't budge no matter how high I revved it. I have read it should never be below 40 psi. What would you do next to sort it out? Yes it has a big cam.

Injuneer

Without the vacuum line attached, the pressure should stay at the “no vacuum/idle” value, in this case 46 PSI. Should never change. GM spec is 43.5 PSI, but anything in the range of 41-47 PSI is acceptable.

When you reattach the vacuum line at idle, the pressure should drop proportional to intake manifold vacuum. In your case, 36-38 PSI would be expected with a stock cam. Maybe only 4-6 PSI drop with a more aggressive cam.

Revving the engine with no load on the engine is of little value. You would expect the pressure to rise as the throttle blades are opening, level off as the engine reaches steady state RPM, and even drop below the 36-38 PSI value as the throttle closes and the revs drop off. All of that happens very quickly. 32 PSI is suspect, but not conclusive.

The real test is taping the gauge to the windshield, taking the car out on the road (or on a dyno), and go WOT, taking it above 5,000 RPM. At peak RPM, under full throttle, the pressure should approach the “no vacuum/idle” reading, in your case 42-44 PSI. If pressure falls below that at any time during the WOT/under load process, the fuel pump may be on its way out.

SSTAT

I think Im on to it. I ffound that the custom pipes/check valves (from the air pump) that came with the my old AS&M headers were rocking back and forth in their fittings, and one header collector was missing 2 nuts and I could move the gasket back and forth with my hand. I think exhaust air leak was telling the computer it was getting too much air so the engine was dumping gas it to correct it (my actron scanner showed trims at +25% at idle) Thanks for the help all-

SSTAT

OK, so again the exhaust has reeked of gas and puked black stuff out the exhaust on start up for for the last 7 years, but its a pure race car and lasted 7 years so no biggie until I rebuilt the motor and the dyno showed it lean at idle?? and getting lean at 6K. Scanner showed fuel trims up everywhere, 5.5 to 25. So I fixed two exhaust leaks in the passenger side and replaced some vacuum hoses, isolated the engine from the air pump (both air valve locations are sealed, the vacuum port for the pump is sealed). I got a vacuum tester and the engine is pulling only 11" of mercury at idle (checked in 2 places), seems low despite my big cam (overlap is 17). Makes you think of a vacuum leak right? But I sealed and pressurized the intake and found no leaks and hosed the motor down with started fluid on 2 occasions with NO bump in idle.

Repeat testing showed the fuel trims on the drivers side all got better (where the collector and air leaks were), from a range of 5.5 to 17 down to -0.8 to 6.3. The passenger side got slightly better 12.5 down to range of 7.5-12.5. Vacuum went up from 11 to 12. Interestingly the O2 sensor voltages on both banks are mostly less than 0.45 (which I read equates to perfect af ratio). Of the 19 numbers I wrote down all but 2 are less than that, from 0.05 to 0.33 which means they are telling the computer its running lean thus explaining the high fuel trims I think. HOW IS THIS POSSIBLE WHEN THE THING REEKS OF GAS ALL THE TIME?? (yes its from the exhaust - not a gas leak). I just thought of the EGR pipe that goes into the passenger header. It was fractured last time I built the motor and I had it welded but had to kind of jury rig it into place. Definitely not well sealed so maybe thats causing the high trims on the pass side. I ordered a EGR blockoff plate but I don't imagine that would cause a vacuum leak though since its already open to the exhaust would it?

I think all that would be left would be to take it to a tuner and see what they can do or find. Maybe its just been a bad tune all these years. Thoughts anyone?? Thanks-

Injuneer

My response is going to get sort of long, so I’ll post in several parts.

When you refer to a “scanner” do you mean a handheld device that shows real time data for a few sensors and PCM parameters? If so, does is have the ability to store a short time interval of the data? Or is this a software scanning and data logging setup?

Sounds like you are copying/writing down single frame data, hence the seeming issue with the O2 sensors reading lean most of the time. And that makes the O2 data somewhat unusable. The PCM reads the O2 sensors up to 9 times per second. They change rapidly, because in closed loop the PCM is toggling the A/F ratio back and forth rapidly between slightly lean and slightly rich, using the short term fuel trims (STFT). That is required for the catalytic converters to work, and the PCM operates that way whether you have a cat or not. Completely normal for the O2 sensors to operate between 0.0XX – 0.9XX volts in closed loop. Sounds like you may have just happened to write down more on the low side, than on the high side.

To tell whether the O2 sensors are reading rich or lean, you have to data log for a reasonable period of time, then take the average value over that time period. If the long term fuel trims (LTFT) are stable, the O2 readings will average out to 0.450 volts +/- say 0.025 volts over a long period of time. In that case, it indicates the LTFT’s are stable, and the STFT’s are averaging 0% correction (or a 128 INT). If the engine was running lean (“true” or “false” lean) the STFT’s would start increasing above 0% (average over a period of time). If the PCM saw that the STFT’s are averaging significantly above 0%, it would start increasing the LTFT’s until it returned the average O2 readings to 0.450 and the STFT’s to 0%.

Yes, 0.450 volts indicates the stoichiometric A/F ratio of 14.7:1. That means there is exactly enough oxygen available to fully combust the hydrogen and carbon present in the fuel. 14.7:1 is "ideal" only in that sense. The engine would get better gas mileage at a leaner A/F ratio. But at 14.7:1 the TOTAL emissions are minimized. If it ran any richer, carbon monoxide (CO) and unburned hydrocarbons (HC) would increase dramatically. If it ran any leaner, oxides of nitrogen (NOx) would increase dramatically.

So… you need to start data logging. You will see the value in that as I respond to the other questions you raise.

Injuneer

Let’s address the EGR situation first. If there is a leak at the corrugated metal riser tube, you can have two problems:
1 – when the EGR system is not activated, the fast moving exhaust gas in the #8 primary will educt (“suck in”) air through the opening. That air hits the O2 sensors, O2 sensor indicate (false) lean, PCM jacks up the LTFT to provide more fuel, passenger side of the exhaust is running rich… and maybe smells of fuel.
2 – when EGR system is activated, the intake manifold vacuum pulls in exhaust, plus the air that enters through the crack or opening in the corrugated metal tube = a sort of vacuum leak, and air entering the combustion system that hasn’t been through the MAF sensor, so no fuel is being added to use up that available air. PCM jacks up the LTFT’s to add fuel.
Definitely have to block the EGR pipe connection on the #8 primary. You can leave the EGR valve in place, and just remove the line to the EGR vacuum solenoid. Then plug the vacuum nipple on the intake manifold with a vacuum cap. Or remove the EGR valve and put a blanking plate on the EGR port. In theory no need to block the connection of the back of the intake for the metal corrugated riser tube, but I would suggest blocking it off anyway.

SSTAT

THANKS SO MUCH! I can build this motor in my sleep but Im a complete idiot on the computer stuff. Off the bat its an actron scanner. You can watch the numbers live and I know it has a recording function but I dont know for how long, I'll look into it. Also the LTFT before I made the fixes was 17 drivers side and 25 passenger side (maxed out as I understand). After the fix drivers dropped a lot to 3.9-6.3 but passenger side stayed pegged at 25. I dont know how long it takes LTFT to adjust but thats why I tended to believe the 'lean' reading on the O2 sensors, or is that still not a valid assumption? I see what I can record with the scanner and THANKS AGAIN!!!!

Also - when the EGR is activated, and if there is a leak at the riser tube, would that effect engine vacuum? doesnt seem possible since exhaust air is coming in anyway, just trying to explain my low vacuum some how, thanks

Injuneer

You indicated above in post #8 that the passenger side LTFT came down to a range of 7.5 - 12.5. Now you indicate it’s still at 25. Clarify.

I have more info to add, but I ran out of time yesterday.

Injuneer

What are your cam specs? Overlap causes low vacuum. Based on my experience 11 to 12”Hg is not unheard of for a “big” cam. You can track vacuum in a data log by adding a new column in Excel and subtracting barometric pressure (BAR) from manifold absolute pressure (MAP). MAP - BAR = vacuum. You may have to convert the units from kPa to “Hg depending on the scanner units.

A lot of overlap can also result in a stinky fuel smell at idle and low RPM. Intake and exhaust valves are both open. The fast moving exhaust creates a scavenging effect, which helps pull the remaining exhaust out of the cylinder and helps pull the incoming A/F charge into the cylinder. With a lot of overlap, some of the A/F mixture is pulled into the exhaust stream. The O2 sensor “sees” the oxygen, indicates lean, and the PCM uses the STFT (and eventually the LTFT) for the idle cell (cell 16) to add even more fuel. Now you are pouring some raw, unburned fuel out the exhaust. PCM is seeing the lean O2 reading. But the exhaust reeks of fuel. Remember, the O2 sensor can only measure the oxygen content of the exhaust. It can’t “see” gasoline/hydrocarbons.

A good tuner can solve the A/F issue vs. overlap. I don’t know the details, but I know it can be done. I am not a tuner.

There is a huge benefit in doing a data log that shows the engine sensors and calculated parameters over a long period of time. See next post... I'm on an iPad and I don’t want it to time out and lose my post.

SSTAT

Sorry, in post#8 I was talking about STFT, I didnt think the LTFT would have changed quickly enough to be relevant. Its a Brett Bauer cam, the overlap is 17. I dont know which specs are usually talked about on the internet but this is what it lists: Gross lift = .568/.562, Lobe lift = .379/.375, Duration @ .020 = 275/264, Duration @.050 = 224/236, LSA 106.5. And Im using 1.6 rockers (1.5 stock).

SSTAT

Holy crap! I just got done reading all over the freaking internet about high overlap and smelly gas exhaust! Maybe thats it! From what Im reading you can get the smell with as little as a 4-6 overlap. I still need to verify on dyno that the lean condition is fixed or false so I scheduled some time 2 weeks from now. My scanner says it will record 'until the memory runs out' but I can set what PID to record and it will let me print them out. I take it I just need the O2 readings or do I need the fuel trims too? It will let me graph individual O2 readings in real time on screen but I cant imagine you would be able to see the patterns you are talking about on a graph on that tiny screen.