When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Guys, I have an '86 Corvette pace car that I have been troubleshooting off and on, for about ten years.
I have tested, cleaned, and/or replaced, just about everything pertaining to the TPI during this time.
The car has a fresh engine and runs pretty good, but will always foul the plugs if left to idle for very long.
It slowly varies the RPM between around 800 - 1100... Sort of like you would expect if the A/C compressor clutch was clicking in and out of engagement.
Besides my own efforts, I have also taken the car to a couple of otherwise competent shops, but they were also unsuccessful at finding the source of the problem.
If anyone here feels that they might have some information and/or expertise that could help me fix this issue, I would certainly be interested in hearing your opinion.
First thing I would recommend if the engine seems to run fine, is a wideband sensor
Fouling plugs is a sign of running rich condition, pollutes the oil, degrades oil, fouls plugs, etc... not good
The wideband will help you diagnose the rich condition. You can install the sensor then perform maintenance and changes until it reads a nice lean a/f ratio at idle and cruise.
I target in OPEN LOOP 14.8 to 15.2:1 air fuel ratio for idle and cruise, like this
See how anything above 55KPA is lean for low rpms? That is ideal to preserve the plug condition, you get nice clean plugs out of the engine, clean oil, long lasting oil and plugs this way.
I'm not suggesting you run YOUR car in open loop and tune it this way, it would be very difficult. I am only showing you how these things can be accomplished so you understand the theory and application of using a wideband sensor. None of the above tuning is possible without the wideband sensor.
The AEM unit is ideal, $150~ from authorized dealer, you can install it, tune your engine up until you fix all the issues, then remove it and re-sell it when finished for almost what you paid. Or leave it as a classic permanent installation!
[QUOTE=Kingtal0n;
The AEM unit is ideal, $150~ from authorized dealer, you can install it, tune your engine up until you fix all the issues, then remove it and re-sell it when finished for almost what you paid. Or leave it as a classic permanent installation![/QUOTE]
Thank you for the prompt reply.
First, let me say that I'm really just trying to get this car running good enough to sell it, so I'm not seeking perfection here.
On the other hand, what you have suggested does sound very interesting, although I am almost totally unfamiliar with the whole process.
I read through the thread links you provided last night, but a lot of what they discussed was somewhat over my head at this point.
I thought those old ECUs had to have the eproms pulled and flashed outside the unit itself...
Let me ask you this; are you saying that the computer can be programmed on the fly... or am I misunderstanding?
Parts shotgun strikes again. Some things that might help with a diagnosis:
What exactly have you changed? What parts did you use?
What is your fuel pressure? Does it drop after you shut the engine off?
Have you gotten a datalog to see what the sensors are seeing?
What have you done to verify it's not a vacuum leak?
Is it throwing any codes?
Thanks Komet.
" I honestly don't remember everything that I have changed... After all, as I said I've been piddling with this thing for over ten years, and there has been plenty of other drama in my life in the meantime, so that the details just aren't that sharp anymore.
I can remember changing the map sensor and the tps sensor after checking them. I did a lot of diagnostic investigation, checking sensor readings, and checking for vacuum leaks. Even built my own pressurized smoker for that purpose.
Totally disassembled, cleaned, and adjusted the entire TPI assembly. And as I said earlier, I've taken it to a couple of reputable shops, where the car was subjected to their own diagnostic procedures.
The fuel pressure initially spikes to 60 PSI, then holds at 42. After shutdown it will hold there for a while, but after an hour or so it will leak down to about 28 pounds.
However... I have just finished cleaning out the tank, and installing a new fuel pump assembly. The car had been sitting in my warehouse for around six years, and the gasoline had basically evaporated down to a syrup. I had a couple of stuck injectors, so after running for a little while longer, I think the small amount of leak down I'm currently experiencing will disappear.
I do not have a datalog, not do I have any code information, as my reader will not interface with those older computers.
First it sounds like your injectors need to be cleaned! They dont just free themselves up completely. This could easily explain why some cylinders are very rich and others are lean, causing some plugs to foul. The ECU is adding fuel in closed loop because oxygen sensors read an average across all cylinders- so 1 cylinder can be 20:1 lean, causing a near-misfire, the others will work their way to 13:1 or 12:1 causing rich fouling. As an example.
Originally Posted by Rich...
Let me ask you this; are you saying that the computer can be programmed on the fly... or am I misunderstanding?
Well, not exactly. The ECU is 'dumb' all it can do is measure sensors (perhaps poorly sometimes) and go from there. It has no clue how well injectors flow or how good the maf or map is working or calibrated. Or whether or not the spark is strong, or weak. Or the fuel quality,... etc...
So this is how diagnostics would go with the wideband sensor.
First, everything needs to be tip top mechanically, e.g. Clean the injectors. Put in new plugs. Compression test the engine make sure they all cylinders read around the same number.
Use new wires, cap rotor etc... clean up the oil from the pcv system and pressure test the intake.
You can't rely on smoke. Here is a quick intake pressure test vid
A pressure test ensures the intake isn't leaking at any individual runner which will cause that 1 single cylinder to run leaner than the rest, which will skew oxygen sensor readings (for the ECU and for the wideband BOTH).
Once the engine mechanical is up to date and fresh and new and ready.... you can diagnose using a wideband properly.
Next you would install the wideband so it can read all cylinders, like after a Y-pipe. The exhaust cannot have ANY Exhaust leaks or it won't work. So this another critical aspect to not overlook, no leaking upstairs and no leaking downstairs.
Now with the wideband installed you would run the engine in open loop at first with a fresh ECU memory (delete the stored closed loop learned values by reset the ECU)
And watch the gauge behavior. It takes a while to get used to how the wideband moves and reacts but after a few sessions it will become more intuitive.
Lets say initially the gauge is reading 13:1 or something rich, and always seems to stay there.
This would point your diagnostics to things that make the ECU cause the engine to run rich, for example, improper MAF Placement, or wrong MAP calibration. You might try changing the map sensor to see if it helps for example. Investigate fuel pressure, try adjusting the fuel pressure down a bit to see if that helps. You are basically making diagnostic adjustments and then review the wideband sensor to see if something helps significantly and if so how to dial it in over time to achieve ideal mixtures. There really isn't a whole lot to "DO" though, as the ECU Is very simple, it basically goes MAF/MAP -> ECU -> Injectors. The injectors flow based on fuel pressure. And thats it. You can play with MAP/MAF, fuel pressure, and engine health until it runs well and shows a good air fuel ratio, say 14.7 to 15.2 at idle or cruise is fine.
Then go WOT and make sure it drops to mid 12's or around that. 11's is too rich. 13's is too lean for WOT. 12.5 is good, 12.8 is good, 12.2 is fine a little rich but fine.
Your "tuning tools" besides directly programming the ECU are based around the fuel pressure, MAP configuration, MAF configuration, and the basic mechanical diagnostics you perform such as pressure leak tests and maintenance related.
There is a good chance as you go through all of this the issue will be corrected by itself thanks to maintenance and careful diagnostics without fooling with the ECU programming, as long as the engine is still mostly original, there is no reason to reprogram the ECU.
At some point you would re-enabled the closed loop to watch the wideband cycle around the set point of closed loop properly, which is 14.6 to 14.8:1 air fuel ratio all the time. It will jump back and forth if closed loop is working properly. That is another powerful useful diagnostic from the wideband sensor.
Then you can remove the wideband if you want- engine is tuned up healthy and done.
First it sounds like your injectors need to be cleaned! They dont just free themselves up completely. This could easily explain why some cylinders are very rich and others are lean, causing some plugs to foul. The ECU is adding fuel in closed loop because oxygen sensors read an average across all cylinders- so 1 cylinder can be 20:1 lean, causing a near-misfire, the others will work their way to 13:1 or 12:1 causing rich fouling. As an example.
Thank you for taking the time to write such a detailed and helpful reply.
I understand that the injectors could be flowing unevenly, however I should mention that the plug color and/or fouling, always looks pretty even and consistent between all of them. I have also checked the exhaust temperatures with my laser thermometer, and at idle all eight exhaust ports check virtually identical.
Well, not exactly. The ECU is 'dumb' all it can do is measure sensors (perhaps poorly sometimes) and go from there. It has no clue how well injectors flow or how good the maf or map is working or calibrated. Or whether or not the spark is strong, or weak. Or the fuel quality,... etc...
So this is how diagnostics would go with the wideband sensor.
First, everything needs to be tip top mechanically, e.g. Clean the injectors. Put in new plugs. Compression test the engine make sure they all cylinders read around the same number.
Use new wires, cap rotor etc... clean up the oil from the pcv system and pressure test the intake.
You can't rely on smoke. Here is a quick intake pressure test vid https://www.youtube.com/watch?v=rYZmZqn3-x0
A pressure test ensures the intake isn't leaking at any individual runner which will cause that 1 single cylinder to run leaner than the rest, which will skew oxygen sensor readings (for the ECU and for the wideband BOTH).
Okay, hold on now, I'm scratching my head over this... There will always be at least one valve open, leaking air into at least one cylinder, right? Plus the idle air port? So how are you able to get the manifold to hold pressure like a sealed vessel? My smoke machine was pushing about 25-30# consistently, and I couldn't find any leaks. The only smoke that leaked out, was going down into the cylinders and coming out of the PVC hole, dipstick tube, etc. I never tried to get it to hold the pressurized smoke inside, but I'm reasonably sure it would not have. The engine is pretty fresh in this car ( an .060 overbored 350 with a torque cam), it has less than 1,500 miles on it. I have checked the compression, and it is between 185-190# on all eight.
Once the engine mechanical is up to date and fresh and new and ready.... you can diagnose using a wideband properly.
Next you would install the wideband so it can read all cylinders, like after a Y-pipe. The exhaust cannot have ANY Exhaust leaks or it won't work. So this another critical aspect to not overlook, no leaking upstairs and no leaking downstairs.
Now with the wideband installed you would run the engine in open loop at first with a fresh ECU memory (delete the stored closed loop learned values by reset the ECU)
And watch the gauge behavior. It takes a while to get used to how the wideband moves and reacts but after a few sessions it will become more intuitive.
Lets say initially the gauge is reading 13:1 or something rich, and always seems to stay there.
This would point your diagnostics to things that make the ECU cause the engine to run rich, for example, improper MAF Placement, or wrong MAP calibration. You might try changing the map sensor to see if it helps for example. Investigate fuel pressure, try adjusting the fuel pressure down a bit to see if that helps. You are basically making diagnostic adjustments and then review the wideband sensor to see if something helps significantly and if so how to dial it in over time to achieve ideal mixtures. There really isn't a whole lot to "DO" though, as the ECU Is very simple, it basically goes MAF/MAP -> ECU -> Injectors. The injectors flow based on fuel pressure. And thats it. You can play with MAP/MAF, fuel pressure, and engine health until it runs well and shows a good air fuel ratio, say 14.7 to 15.2 at idle or cruise is fine.
Then go WOT and make sure it drops to mid 12's or around that. 11's is too rich. 13's is too lean for WOT. 12.5 is good, 12.8 is good, 12.2 is fine a little rich but fine.
Your "tuning tools" besides directly programming the ECU are based around the fuel pressure, MAP configuration, MAF configuration, and the basic mechanical diagnostics you perform such as pressure leak tests and maintenance related.
There is a good chance as you go through all of this the issue will be corrected by itself thanks to maintenance and careful diagnostics without fooling with the ECU programming, as long as the engine is still mostly original, there is no reason to reprogram the ECU.
At some point you would re-enabled the closed loop to watch the wideband cycle around the set point of closed loop properly, which is 14.6 to 14.8:1 air fuel ratio all the time. It will jump back and forth if closed loop is working properly. That is another powerful useful diagnostic from the wideband sensor.
Then you can remove the wideband if you want- engine is tuned up healthy and done.
Okay, I get what you are saying regarding the HP Tuners/02 sensor setup now. At least if I go that route I will be able to read the fuel mixture ratio in more or less real time. Not exactly the quick fix I was looking for, but it would be empowering to have direct access to that data, I admit. If I can't find a way to shortcut the troubleshooting part of this repair, I may very well go that route, as it does sound interesting.
An open intake valve doesn't leak air out of that cylinder much. It will just fill the cylinder and dead end there. Some leaks out past the rings though which is why you check the dipstick for flow, or other issues. A lot of flow into the crankcase and out the dipstick is indicative of a problem with ring sealing on that cylinder or two with open intake valves or that somehow it is getting into the crankcase some other way. This is why we listen carefully, make sure the engine is setup right in terms of intake and crankcase sealing. Since a compression test is very easy to do and review numbers for all cylinders you can confident that when one or two cylinders leaks into the crankcase will more or less represent what they all will do if they all show the same compression under the circumstances.
Intake pressure holding even with a cam and overlap situations,
Alot of people will first say its impossible to perform a pressure test with intake valves open and overlap in the cam, but I've tested a hundred engines this way of all shapes and sizes and continue to do so. The leaking is very small.
The pressure test is a critical test, People approach me to tune their cars all the time and I won't even consider tuning an engine until its had this critical test. There are three things we always do before tuning: pressure test, timing test, and compression test.
Here is the entire pressure test from compressor cover to intake manifold.
Notice there is a differential of pressure from cover to intake manifold caused by the open valve, ring sealing, and overlap conditions. All engines will have this situation but the leaking is so small that we can still adequately perform the test. In turbo forced induction apps you must test the entire plumbing because all leaks no matter how tiny will affect the compressor negatively, raise exhaust temp and exhaust pressure and facilitate piston over heating and excess compressor wheel speed causing engine/turbo demise.
An open intake valve doesn't leak air out of that cylinder much. It will just fill the cylinder and dead end there. Some leaks out past the rings though which is why you check the dipstick for flow, or other issues. A lot of flow into the crankcase and out the dipstick is indicative of a problem with ring sealing on that cylinder or two with open intake valves or that somehow it is getting into the crankcase some other way. This is why we listen carefully, make sure the engine is setup right in terms of intake and crankcase sealing. Since a compression test is very easy to do and review numbers for all cylinders you can confident that when one or two cylinders leaks into the crankcase will more or less represent what they all will do if they all show the same compression under the circumstances.
Intake pressure holding even with a cam and overlap situations,
Alot of people will first say its impossible to perform a pressure test with intake valves open and overlap in the cam, but I've tested a hundred engines this way of all shapes and sizes and continue to do so. The leaking is very small.
The pressure test is a critical test, People approach me to tune their cars all the time and I won't even consider tuning an engine until its had this critical test. There are three things we always do before tuning: pressure test, timing test, and compression test.
Here is the entire pressure test from compressor cover to intake manifold. https://www.youtube.com/watch?v=w1sb5Y1J068
Notice there is a differential of pressure from cover to intake manifold caused by the open valve, ring sealing, and overlap conditions. All engines will have this situation but the leaking is so small that we can still adequately perform the test. In turbo forced induction apps you must test the entire plumbing because all leaks no matter how tiny will affect the compressor negatively, raise exhaust temp and exhaust pressure and facilitate piston over heating and excess compressor wheel speed causing engine/turbo demise.
Okay. Well again I appriciate your carefully detailed replys. I will certainly take all this under advisement. I'll give the manifold pressure test a try just out of curiosity, when I get time. Always interested in learning something new. Just hope I don't blow out any TPI gaskets in the process.
Well two things to consider
1. If you think the engine can handle boost pressure, then it will handle the pressure test
2. If you are NOT forced induction and no plans to make boost, just use very low pressure like 2 or 3psi max in the intake manifold
No need to go 22psi like I do, thats only because I run 22psi of boost pressure into it every day. In other words you test to the boost pressure you intent to run, a little over that, just to be sure.
I like komet's response, I agree with it. And another thing to consider,
3. Positive pressure is just as rough as negative pressure, so for example if your engine pulls 25" Of vacuum during Decelleration, well 25" of vacuum is like 12.5psi of negative pressure, which means the engine needs to be able to withstand -13psi of boost and +13psi of boost, either way it will blow the gasket out with the same force. Makes sense? -5psi is the same as +5psi in terms of PSI which stands for pounds per square inch....
When the engineers design these manifolds they always (Well, they are SUPPOSED to) always make sure it can withstand forces slightly above what it will actually experience. So for example the LS1 intake was designed for a naturally aspirated engine like the TPI intake, but the engineers said "well it could go as deep as 28" Of vacuum so lets make it withstand 30" or 32" Or something slightly more so it never fails" This is a common engineering practice, its called safety factor. For example in a building when they design the force the building can withstand, it has a LOT of extra safety factor, so it can withstand far more force that is normally applied. The safety factor adds materials, adds weight, so it isn't often easy to get a bunch of safety factor into everything, like helicopters for example. Helicopter needs to be very light weight so they can't just add 5x safety factor everywhere it would be too heavy.
Finally just a reminder, pull the pcv valve and test it with the pressure test for leaking, and also make sure you don't pressurize the crankcase. Pop out the dipstick tube and make sure there is no heavy flow from the dipstick indicating you have a line somewhere pushing air into the crankcase. Crankcase seals cannot withstand much pressure for long. 2 or 3psi would be okay for a brief time but 10+ Psi NOO non o nonono <3
I have a 1992 camaro rs 5 speed that I swapped a 5.7 tbi in from a 1990 gmc v1500. According to my research they made a 1990 v3500 5.7tbi in a 5 speed manual. I orderd that eprom. Will it be compatible with my camaro ecu or do I have to buy and wire in a v3500 ecu?
05-24-2022 | 09:08 PM
