1999 Chevy Cavalier long crank.

We recently bought this car.  Car has 117,600 miles on it with the 2.2L and the 3 speed automatic.  Previous owner recently installed new AC-Delco plugs, O'reilly spark plug wires, and a new battery.  Car also has both new GM O2 sensors.  Car has a long crank when cold and sometimes when hot.  There are no codes stored in the PCM and all monitors have been run so no codes have been cleared.    Here is what I have tested so far:

Checked fuel pressure at KOEO:  results were 42 psi and spec is 41-47.
Checked fuel pressure running:  Results were 37 psi and spec is 31-44.

Fuel leak down:  Spec is no more than 5 psi in 10 minutes:  My result was 3 psi in 10 minutes.  Next leak down spec is drop psi down to 10 psi and the pressure should not leak down more than 2 psi in 10 minutes.  My result was psi went up to 14 psi in 10 minutes.

I checked Scan data for the MAP, IAC, ECT, IAT, Baro, and TPS.  All data seemed normal except the TPS was not really in spec.  Alldata states that with the throttle plate closed, voltage should be at around 0.6 volts.  My results were 0.47.  WOT should be close to 5 volts and my results were 3.76 volts.  Both of these tests were done KOEO.  I also watched the data in graph mode and the voltage was smooth with no dropouts.  My IAC counts at cold startup were 95 and then dropped to about 25-26 at a hot idle.  Car idles in the 650 rpm range when warmed up.

I checked power and grounds to the TPS and had 5.04 volts on 5 volt reference wire and a good ground voltage.  I also used a jumper wire from the 5 volt reference wire to the signal wire and watched scan data and the signal wire showed 5 volts so I know that my wiring is good to the computer and the computer can read the signal voltage.

I watched engine rpms while cranking and it shows in scan data so it appears that my crank sensor is working.

O2 sensors are working correctly.

I cleaned both the throttle body and the IAC valve and bore.

Scan data shows no misfires or history of misfires and I don't feel any misfires.  Car runs good and has no issues running down the road.  Car does not overheat or get hot.

So my question is, are my TPS values skewed enough to cause this issue and if not, what else should I be testing or looking at?  Thanks for any input.   

Today I verified that I had spark coming out of both coils so the coils and ICM seems to be working correctly. I did discover that the previous owner had put what looks like dielectric grease on the coil terminals and I cleaned it off. I will let it sit again and start it again and see if that grease was causing some issues.

Sorry for the late reply. To answer your question about the TPS voltage, I do not think the TPS values are far enough out of range to be causing the issue you are having. I would expect that for the TPS to cause any issue while cranking, it would have to report near WOT and would cut injector pulse all together.
It's been a long time since I saw one of those cars, but if you were to watch injector ms and TPS on the scan tool, you may even be able to see injector on time go to 0ms above 80% throttle.
Does this engine have the remote mounted MAP? I have seen the hose connecting the MAP to the intake crack, fall off or collapse.

Thanks for the reply. No, the MAP sensor is mounted in the intake below the throttle body and right under the TPS. My MAP readings were also normal.

I'm about at a loss to which direction I go now.

I know you mentioned checking the ECT sensor in your first post, but it may be worth another look. An intermittently lying ECT will skew the cranking injector pulse width to the point of a no start/hard start/long crank.

Check the ECT and IAT after sitting overnight, before starting the engine. They should both agree to within a few degrees of each other, and with the actual ambient temperature.

I did check the ECT and IAT against each other before the car was started for the day and they were within a degree of each other and the temps were consistent with the ambient outside temps. ECT was at 95 degrees and IAT was at 96 degrees. Car still had a long crank when I did go to start it.

It sounds like you've been pretty thorough. Can you scope the cam and crank signals while cranking?
A missing cam signal will cause a long crank time as it will take longer to synchronize injector firing. All though you would think would also set a code.
A weak VR type crank signal could also cause a longer than normal crank time as the low cranking speed coupled with a weak sensor may take longer for the signal to reach the amplitude required for the computer to recognize the signal.
Also, don't overlook the battery. It wouldn't be the first time I have ever seen a battery that could sustain cranking but not supply enough voltage while doing so to support critical electronics.

I plan on trying to scope them tomorrow. Today I tested injector pulse with a noid light but unfortunately the car was starting fine because I had already started it previously. I was getting injector pulse the entire time. I'm going to let the car sit overnight and try it again because it will usually long crank after sitting for a while.
The battery is new and has good voltage. I voltage drop tested the starter today and it only lost about a volt when starting but the car was immediately starting with no long crank when I did the test.

I also checked the 5 volt reference at the MAP sensor and it was good. I also jumped the reference wire to the signal wire and I got 5 volts on the scanner so that circuit is good.

The biggest issue is that you only get one chance to test anything because once you start it, it will usually not long crank on additional starts unless you let it cool down for a long while.

This morning I also primed the fuel pump several times before trying to start it for the first time of the day and it still had a long crank which again verified to me that it's not a fuel related issue. Thanks again for the reply.

So I did take some cam and crank captures with my scope but I don't know how to post pictures here. The captures are not great. The cam signal looked normal to me but I never could get the crank signal to look like any of the crank signals that I have seen of other vehicles. I messed around with the voltage and time base but never got it looking good.

Long story short, I became a parts changer and replaced the crank sensor and did the relearn for it. That did not fix the issue and the scope capture I took for just the crank sensor looked the same as the sensor that was previously in the car.

I gave up and called a local mobile mechanic to come look at it but he flaked out on me and did not show so I guess we are just going to let my fiancé's son drive it like it is until it breaks for good. I have no other ideas or tests to do so I'm basically done messing with it.

That's frustrating. I get it, some times you get to a point where it just makes more sense to try a part on the car than to keep poking around. I think we've all been there.
It sounds like the car will still be reliable in it's current condition.
If you do end up finding a resolution to the long crank issue, please be sure to update us here.

This car has been very frustrating. Today I did the MAP sensor output test and it passed. I originally did the test wrong and thought it failed so I went and got a new MAP sensor and it wasn't until I tested the new one that I realized that I had done the test wrong. I guess I will have to take the new sensor back if they will take it. It wasn't installed just hooked to the connector so I could do the vacuum test on it.
My fiancé's dad and his brother-in-law think it has something to do with the oil pressure sensor and the oil but I cannot find any information in service information that ties the oil pressure on this car to how it starts.
Tomorrow I'm going to test the ohms on the ECT sensor when it's cold to see if the resistance is correct. I tested it hot and the resistance was in spec at around 2.64 ohms. The last time I checked it after the car had sit for a while, the coolant temp was at 104 degrees Fahrenheit and the ohms reading was 1.4 k ohms which if my conversion is correct is about 1,400 ohms which seems to be on the low side but I'm not sure. Ambient outside temps are in the 90's here so the cold ohm readings may be low.

I really don't ohm test temperature sensors. I'm not saying you shouldn't, it's a viable test. I prefer to check the data against another temp sensor before the engine has been run, then I like to disconnect the sensor to make sure the voltage PID goes to full 5v or the temp PID goes to -40°f. This rules out resistance in the circuit.
As far as the oil pressure theory, i can see why a lot of people want to go down that route on older GM vehicles.There's some SD videos that cover the operation of the oil pressure switch and how it goes into the fuel pump circuit. Essentially, it's a back up control circuit for the fuel pump relay. Assuming there was an issue with the fuel pump relay control circuit, eventually after cranking the engine enough oil pressure will be made to close the oil pressure switch which will in turn active the fuel pump relay. You've tested fuel pressure all ready, I think you would have noticed if it i wasn't building fuel pressure until after some excessive cranking.

The captures are not great. The cam signal looked normal to me but I never could get the crank signal to look like any of the crank signals that I have seen of other vehicles. I messed around with the voltage and time base but never got it looking good.

I would keep looking into the Crank Signal. This is a 2-wire VRS (Variable Reluctance Sensor). That means that the sensor should produce its own AC voltage. The waveform should be an AC sine wave. If the air gap is too wide, the amplitude of the signal could be too low and it could take the ICM longer to get a good sync. 

Connect the negative lead of you scope to the yellow Crank Sensor wire and connect the positive lead to the purple wire. For cranking, set your scope screen time to about 500 ms - 1 second. Set the voltage scale to about 500 mv - 1 volt. As the engine speed increases, the voltage output should increase, too. 
  

I only ohm tested the sensor because I wanted to rule out the sensor even though it tests good compared to the IAT sensor.
On this car, I don't see the oil pressure switch being tied to the fuel pump circuit according to the wiring diagrams unless it's internal to the PCM.

I did replace the crank sensor and it made no difference with this issue. All the wiring that I can see looks fine and the connector looked fine.

The mounting surface for the crank sensor was clean and not rust jacked but I cleaned it anyway with the same results so I don't think it's an air cap issue and there is no adjusting when installing the sensor.

I did hook up the scope as you suggested when I was doing my tests as well as hooking the negative lead to the battery with no difference in the scope patterns. This is my fiancé's son's car and he has it now, but I will scope it again when I get a chance.

This problem seems to me to be temperature related since it mainly only does it when the car is cold and the first start of the day or if you let the car cool down too long. I'm not sure if that would be a crank circuit issue or not. Thanks for your help.