550 Intermittent Start Issue

I could use a little help. My 1999 550 has developed an intermittent not start issue

If I start the car from an overnight rest, it starts right up and runs perfectly normally. If I go for a drive, it may or may not restart.

To save time on the forum I have found a way of reliability duplicating the issue. By starting the car from cold and running at idle for about a half hour and then turning it off, wait 5 minutes and then try to restart it will not start. The entire immobilizer system gives the proper indications as per the manual and normal operation. Using the key fob and getting the normal two siren beeps and then turning the key to start it will spin at normal speed but will not start. If I leave it alone for several hours and come back it will start right up.

I have performed the following tests when it is in the no-start condition.

1. I have verified spark by putting an inductive timing light on the spark plug wires and it flashes the light normally.

2, I have verified that the injectors are getting 12 VDC, and by using a 12VDC bulb as a “noid” light I have verified that the injectors are getting a signal to allow them to work, (flashing light at injector firing rate.)

3. With the engine stopped after cranking I have verified that I get a high pressure stream of fuel from the fuel rail test point. I have not actually tested pressure, but it will squirt out a stream 2-3 feet under controlled conditions.

4. I have verified that there is 12VDC leaving the fuse panel going to the fuel pumps by measuring at the panel and also at the pumps. I can hear the relays clicking when going through the start procedure.

5.I have removed the gas cap to verify that a vacuum has not developed in the fuel tank.

6. I have reset the inertia switch just for good measure. No inertia light showing.

7. I have checked all fuses.

8. I have switched relays around to see if the problem would follow a given relay.

9. I have reseated all the connectors on the fuse panel and ECU. The ECU connector is absolutely clean.

10. As I have been working on these tests and time has past, I will occasionally try to start the car. After several hours it will make a feeble attempt to run and might actually run very rough for a few seconds. Again, leaving it alone for 4-6 hours it may start and run normally, leaving it overnight it will reliably start ad run normally.

I’m trying to understand what component that might be sensitive to heat would cause both banks not to run. It never tries to run on just one bank. I have replaced the cam position sensor as that is used by both ECUs. No change.

Any help or ideas would be very much appreciated.

Gary

 

Gentlemen, thank you for your comments.

With regard to the relays, Since most of the relays are of the same type I have moved them around from one function to another to see if anything changes. I realize this is probably futile as both banks are dead at the same time, but I had to try.

I really didn't think the inertia switch would have anything to do with the problem as the car is stationary and the inertia warning light never comes on.

The car has set a P0116 fault code from time to time in this process, but it's my understanding that this is a temp difference code sent to the ECU for comparison of cold start to hot start and is probably more for emissions than anything else. I don't believe it would preclude the engine from starting. If anyone has a better understanding of this please let me know.

After thinking about it overnight the symptoms are similar to old-fashioned vapor lock. I didn't think that could happen in a FI system, but I have read that it might in some cases.

Today I'm going to get a fuel pressure gauge and get an accurate view of what's going on with the pressure.

I appreciate any thoughts.

Gary

 

OK

I replicated the no-start problem starting from cold and running for about a half hour. I then turned the engine off and tried to restart within a minute. All immobilizer functions were normal and the engine spun as fast as normal but absolutely no firing. I measured the fuel pressure while I tried to start it and got about 50 PSI, 3.5Bar.

So I have fuel pressure, injectors that are firing, spark plugs that are firing, but engine that does not run.

I did notice that the relays for the fuel pumps were quite warm so I exchanged them with like relays that were cool. No difference. I have full 12 VDC out of the fuse panel to the pumps so the relays should not be creating heat from an IR loss. I have a 3/4 tank of fresh Chevron gas.

This leaves me with the P0116 code. Can that be stopping things? I am experimenting right now with cooling the interior and fuse panel area with a direct fan while leaving the hood closed to see if the normal start condition comes back any sooner. I have applied freeze coolant spray to the ECU and the fuse panel with no change.

The P0116 code can be reset and when I start the engine from cold it will run for maybe 20 minutes or so and then the code will set again. Don't know what that means. It's possible that the engine would restart up until the P0116 codes is set, but I have not tried monitoring it that closely. I may try that tomorrow.

Any thoughts especially regarding the P0116 code and associated sensor and whether it could be the problem?

I am now out of ideas unless it's the temp sensor associated with the P0116 code, possibility? Unless I find something or someone has a brilliant idea the car will be of to a trusted mechanic.

Thank you.

Gary

 

Qavion

Thank you for the information. I cannot read the banks individually. I read the codes with an OBDII code reader and phone app.

I'm thinking the sensor associated with P0116 is the root cause.

This afternoon I ran the engine for about 45 minutes and the P0116 code came on sometime during that run. At the end of the run, it would not restart as expected. See my comments above for details.

I am trying to isolate the heat problem to the interior ECU's etc or the engine compartment specifically the temp sensor associated with P0116.

To this end when I finished my 45-minute run and had the no-start condition I immediately put a huge fan in the passenger seat blowing at the ECU, fuse panel, and other modules, and ran it for about 30 minutes til everything felt very cold. Still no start. I kept the hood closed during this time so it was very hot in the engine compartment.

I then took the fan and aimed it directly at the intake manifold and ran it for about 45 minutes and the engine started right up albeit with some initial stumbling that cleared rapidly.

I think there is a good chance that the temp sensor is giving a very bad input to the ECU which in turn is giving bad information to the injectors and ignition timing.

The way the engine starts is as if it were flooded, with black smoke and stumbling til it clears out.

My mechanic also thinks this could be the problem. I'm not up for removing the intake manifold to replace the sensor, as I'm not able to bend over long enough to be effective. So I'll have both temp sensors replaced while we're there, as well as update the under-manifold hoses. I'll also have him verify the can bus per the P0116 instructions you sent.

Thanks again.

Gary

I feel fairly comfortable with this diagnosis, though my mechanic will also go through checks to confirm the problem and potential solution.

 

FatBillyBob and Scowman

I have confirmed fuel pressure with a gauge attached to the fuel rail, both individually. Measured 3.5 BAR when cranking under no start condition same as under running condition. I have measured 12VDC at the injector connectors and I have put a Noid light in place of an injector to verify that the injectors are getting a ground to fire them.

I have pulled a plug wire and used a standard inline sparkplug tester as well as a timing light connected to the pug wires to verify a hot spark visually and by getting a normal functioning of the timing light.

The battery is always on a battery keeper so it is fully charged and has almost no voltage drop under starting conditions. IE it spins the engine very rapidly both when the engine starts normally and when it will not start.

As you point out these are basic checks that anyone should make and I thought I had indicated that I had performed them in my prior postings, but perhaps I did not make myself clear.

In any case, I feel like I have narrowed the problem to the ECU temperature sensor. See my prior post. At 81 years old and working with a broken hip I am not physically able to remove the intake manifold to replace the sensor and associated temp gauge sensor so I will be taking the car to a trusted mechanic to go forward.

Just as background I have 50 years of experience racing all over the US and Canada and have done a huge amount of troubleshooting on my cars as well as the cars of my fellow competitors. I am, however, not an expert on ecu-controlled systems like in the 550, a working knowledge, but not expert knowledge.

The bottom line is I'm sorry that you seem to think I'm a bit of a fool.

Thank you for your inputs.

Gary

 

Qavion

Thank you for your input. Yes, I'm aware that there are two sensors and only one feeds the ECUs. I did not have the chart information that you included in your post so I may try to take some resistance measurements and see what I come up with. Because of my hip, it's currently hard for me to get down to the ECU connector but we'll see.

Thanks again for your positive input.

Gary

 

Thank you very much for the schematic. It will be very useful.

Gary

 

Sorry, I was a bit heavy-handed with my reply. I'm dealing with a lot of medical issues right now and I let my frustration overflow into this forum.

Gary