Lucy's wierd startup issue - New info

[Facime]

Ok so some of you know that since the very first day I got Lucy she has been giving me fits in the starting department. She always cranks, but sometimes will not fire up. It used to be that the only thing that cured the issue was time. If I let her sit, say over night, she would fire right up in the morning. Well after replacing and/or swapping nearly every part I could think of, the occurance of the issue has decreased but the mystery continues.

Now for the new info. For the last month or so the problem has taken a wierd turn. Now sometimes when I get in the car and turn the key, nothing happens. Usually once, maybe twice...turn the key, nothing. Then when I try again it will turn over but fail to start. When the failure occurs it turns normal but it doesnt act like it is even trying to start. (say, like you unplugged the coil packs or something like that)

Through a fluke I tried starting it after I had pulled off the 5th and 6th injector connections. The car cranked then stumbled then started. I shut the car off, plugged back in the injectors and it again started and idled smoothly. The last 3 times the car has failed to start I pulled off the last two injectors and she will start. wierd huh?

So, Im now thinking that Lucy suffers from a rare occurance in an EFI engine...flooding. Flooding alone does not explain the "no crank" issue that occures immediately preceeding the "crank but no start" condition. I have to believe that the two conditions are linked in some way, but I have yet to come up with a logical explanation why.


Background info: I am getting spark when in no start condition. I have fuel in the rail and jumping the FP does cause fuel to surge through the return line. The problem screams electrical but electrical problems dont just fix themselves over and over under certain conditions. Some days it will do it at every stop and some days it wont do it all.

Im still baffled...anyone have any suggestions?

 

[CRE]

Have you tested the clips for 5 and 6 for continuity between the two terminals? There may be a short in the harness for one or both of them.

 

[Facime]

but how would having a short in them make it not want to start when they are plugged in, start when unpluged, then fire on all 6 when I plug them back in? I dont think thats an issue. If it was intermittent I would notice stumbling and lack of power when driving it. That is not the case.

I had been thinking that the cylinders are getting flooded with fuel and when I unplug the injectors it allows the system to clean out. But a thought just occured to me...Does anyone see logic in this idea, maybe I have low fuel pressure and when I unplug two injectors it allows a slight raise in the pressure allowing the other injectors to introduce enough fuel to fire.

That still doesnt explain why this occures right after a no crank start attempt.

At least now I know how to get my car started and it wont strand me anymore. Im thinking about wiring the two injectors to a switch in the cabin to shut them off when I need to start it...lol

 

[CRE]

You're right about the intermittant thing.

I don't think it's flooding or the low fuel pressure though... low fuel pressure would be even more of a problem past idle. As for the flooding idea, pull the injectors and test them for leaking. They'd have to be pouring out fuel to flood the cylendar in the one second it takes to turn the key all the way over, if that's the case you'll know real quick once you pull them.

Do you have high or low impedence injectors? I just can't keep which years have which injectors straight.

 

[Facime]

Low impedence, soon to be Denso 550's though. Also have a Walbro going in at that time. I will be installing an AFPR then as well, so I will be able to monitor my fuel pressure.

Also you may have misunderstood one thing. When this occurs its after the car is up to operating temp, and is shut off and left for 5 to 10 minutes. Returning to the car after say going in a store, or stopping for fuel.

In the last thread I made about this car all of the "heat soak" questions were asked, answered and possible problems were investigated. It doesnt seem to specifically be heat related, as its done this in a variety of temperatures and conditions. Oh yeah and still none of this explains the lack of turning over on the first of second try.

Does anyone think I should bypass the otherwise properly working factory alarm?

 

[CRE]

Keep in mind that while this may not be specifically related to the outside temps, some things such as the injector and fuel pump resistor packs create heat.

Unless there is something horribly wrong with your car's wiring pulling the clips from the 5 and 6 injectors to get it started shoudn't have anything to do with the stock alarm.

That said, I'm lost and dumbfounded. I look forward to the answer though.

 

[Facime]

Heh, so do I. I suspect as I continue to put money into the car and slowly upgrade things, something will make it go away and I will never know exactly what it was. As far as Im concerned its a gremlin in the system and/or the car is possessed.

 

[Facime]

still looking for some expert advice here to help me make sence of this problem.

 

[chriso]

Have you checked your ignition switch?

 

[Facime]

no, I have not. Im not even sure I know how to. However, if memory serves correctly I do have a spare in my shell (just the switch part that goes behind the key cylinder) I suppose its worth a try, though I dont know how it would affect the starting so long as its turning over, but Im willing to give it a try, thanks for the suggestion.

 

[jetjock]

Weez my man, I normally don't deal with those who exhibit anthropomorphism because I feel it's weird behavior but since you've always come across as an otherwise normal and decent guy....

Flooding is unlikely unless your injectors (including the CSI) are leaking so you may want to check for that. You can do it by monitoring fuel rest pressure after pressurizing the system. If it drops off too quickly you have a leak somewhere. Does it have also trouble with hot starts?

Just FYI the TCCS has the same flooded start mode common to all EFI cars in that if you hold the pedal to the floor it'll cut injection off until the engine starts. Just like in the old days with carbs but done electronically through the TPS. Course if the injectors are leaking that won't help.

You need to fix the cranking problem first because the ECU depends on the starter signal to do several other things. A missing or intermittent STA will completely prevent your cold start injection system from working. The STA signal can be checked at the TTS or CSI but if it were me I'd check it at the ECU first and then check for it at the TTS and CSI. If you have 12 volts with the key turned to start the no crank situation is likely bad starter solenoid contacts.

The ECU also needs to see CPS output during cranking. It uses this in combination with STA to fire all the injectors at once during a start so if the engine doesn't crank it won't do that. Add in a non-functioning CSI system caused by the missing STA and you get hard starting. Lack of STA should be generating a code so check for it too.

Bottom line is you have multiple problems but you need to fix the no crank one first. That'll get you back to where you were and you can go from there. None of this is magic, it just takes a systematic approach to find the problem.

 

[pimptrizkit]

hows the power wire to your ecu look?

it's on the power side of the batt, and is in a gray heat sheild. check to make sure it's in good status,

that wire was my no crank problem, it had wiggled lose, and had a bad connection.

 

[Facime]

Thanks JJ I'm glad to see you back here, I always value your analytical approach to things.

Here is the thing I think I need to reitterate though.

1. 95% of the time the car starts normal when hot.
2. 100% of the time it has ZERO symptoms cold.
3. 100% of the time the car runs normal and makes excellent power after it is started.
4. During the 5% time when the car cranks but fails to start, it occures 100% of the time after the temporary "no crank" condition.
5. The "crank but no start" condition is ALWAYS preceeded by the "no crank" attempts.
6. My instinct tells me the two problems are related.
7. I can not locate a problem when the condition is not present (in other words I have to wait for it to act up before I can do any further diagnostics, and it never seems to happen when Im near my tools and a reasonable place to work on it)
8. 100% of the time, unplugging 2 injectors gets the car to start after roughly 5 seconds of cranking.
9. The "no crank" condition has never lasted more than 3 attempts, and usually will only do it once. (adding to the difficulty in diagnosis)

I dont have a fuel pressure gauge so I cant check for leak down, but if that is a problem it will be cured when I swap in my 550's (soon). Holding the pedal to the floor used to actually help in the past, but since the "no crank" condition has popped up it no longer helps.

My main questions at this point are:

1. is there some electrical connection, sensor, control unit that would cause the system to have spark and fuel but no fire, that could also work one minute then not the next, then work again later?
2. how would unplugging two injectors cause the engine to start relatively easily when just minutes before it acted like it would never fire.
3. How would any of this be temperature related? (this never occures cold)

Lastly, I loaned out my digital meter to someone. When I get that back I will carry it in the car and try to do some diagnostics the next time it occures.

 

[jetjock]

1) is there some electrical connection, sensor, control unit that would cause the system to have spark and fuel but no fire, that could also work one minute then not the next, then work again later?

Well, connection and sensor-wise it goes like this: As long as you have CPS signals to the ECU you'll have spark and as long as you have spark you'll have injection if the feedback from the ignitor to the ECU is ok. When you say you have fuel I'm assuming you mean injection. Fuel in the rail ie; working pump, regulator, return line, etc, is useless by itself so I'm trusting you when you claim spark and injection. If you don't have spark the ECU will cut injection within a few seconds to protect from hydrolock and catalyst overheat. It's why a 7M crank/no start diagnosis should always begin with spark. None means to work backwards towards the CPS. Fuel should come last.

2) I can not locate a problem when the condition is not present (in other words I have to wait for it to act up before I can do any further diagnostics, and it never seems to happen when I'm near my tools and a reasonable place to work on it)

You can instrument it beforehand. They're are many way to do this, some not all that difficult. For example I have leds on things like my VSVs and the CSI. best to carry a meter and familiarize yourself where the signals can be accessed. Also have a plan on what you're going to check prior to it acting up.

3) 100% of the time, unplugging 2 injectors gets the car to start after roughly 5 seconds of cranking. How would unplugging two injectors cause the engine to start relatively easily when just minutes before it acted like it would never fire.

That's strange. If they were leaking unplugging them wouldn't matter. Either it's unrelated or could be a wiring problem. If it was a wiring problem it'd also show up after start though. The ECU may be doing something weird but I doubt it. Do you smell fuel at the tailpipe when this happens? That's where a gas analyzer comes in handy because the tailpipe will be off the scale with HC is such a case.

4) How would any of this be temperature related? (this never occurs cold)

Well, from a TCCS standpoint the coolant temp sensor has the greatest authority over starting enrichment. Also ignition timing and idle. For example if you unplug the coolant sensor the ECU defaults to a temp of 180 F. If you then try and start a cold engine it'll be a chore because the ECU will think it's already hot and won't provide cold enrichment. The CSI should still work though as it's an independent system. As I said both the ECU and CSI need a valid STA signal to work. That signal is taken from the same wire as cranks the starter.

I don't quite follow this: "100% of the time it has ZERO symptoms cold."

Prior to the no crank I though your complaint was cold starts? Forget about rest pressure then as it's almost always cause bad hot starts but an easy way to check pressure is to drill and tap the CSI bolt for 1/8 npt. That way you have a place to install a gage when needed. I have a quick disconnect on mine so I can pop in a gauge in seconds. If you get an AFPR it'll also solve that problem.

The only relation I see between the problems is the lack of STA and CPS signals when cranking so fix the no crank issue first. One miracle at a time. It's the simplest and all other problems are moot if it won't crank. The problem could be in the ignition switch, clutch safety or neutral start switch (if you have an auto), or more likely the starter. Common failure modes there are bad solenoid contacts or worn brushes. Try rapping the stater with a long socket extension or pipe next time it happens.

I assume the MIL comes on every time you turn the key on? Codes? No codes would send us in another direction entirely but focus on the no crank thing first. You could also check all the connectors associated with these systems. I've seen some pretty shoddy connectors in my time. That's bad because any EFI system depends on good electrics more than anything else.

It's tough to troubleshoot from afar so that's the best I can do until you check some stuff. In the end it'll likely turn out to be something simple and completely explainable. It's just a machine. It has no crazy mind of it's own you know, even if you do refer to it in the female vernacular

Even if you get it fixed, and assuming you're still there, next time I drop into CVO I'll bring the test gear needed and one of my gas analyzers to go through the system but that probably won't be until spring. I can't go in there with the jet so I have to use something smaller and winter weather over the Siskiyou is nothing to fool with down low. It puckered me up on more than one occasion when I lived up there. For now get your meter back and study the wiring. Ask if you have more questions.
__________________

 

[Facime]

I don't quite follow this: "100% of the time it has ZERO symptoms cold.

By this I mean that none of these problems have ever occured when the car is dead cold. It starts promptly when cold, and actually will start just fine with the CSI unplugged. I have also never experienced the "no crank" issue when cold.

Prior to the no crank I though your complaint was cold starts?

no as I just mentioned above, cold starts have never ever been a problem. These symptoms ALWAYS occure when the car is at full operating temp.

also to cover some of the other things you mentioned, Its nearly impossible to tell if the starter is at fault, because like I said it only fails to crank once or twice when I try it and ive never had it do it more than 3 tries. Usually it is one try: nothing -> second try: crank, but no start. Im not willing to buy a starter just as a diagnostic tool. Also this problem has never thown a code. Ive never seen a code 11 on this car either, the only codes its ever thrown is fuel cut.
One of the possibilities I thought was that the temp sensor for the ECU was bad and telling the engine it was cold 100% of the time and causing difficult hot starting. I thought I made a deal with someone that was gonna net me a sensor, but that seems to be on hold at best. I did find some bad connections in that area but have already replaced all the bad wires and connectors there. Wireing under the hood has been suspect in a number of locations that I have found and since fixed or ruled out. The wiring under the dash however looks mint and untouched. The car shows no signs of corrosion or rust anywhere and all connectors in the cabin "appear" ok.

I guess the thing that is so baffling is the conditions in which the failure occures, and the remedies that have presented themselves over time. Near the end of next summer I intend to rebuild the engine and I was going to use a new or at least mint used engine harness. Until then, or when one of my attempts to fix the problem for good works, at least Lucy is no longer leaving me stranded. And yes I refer to this machine by a female name, when you'vehad as many cars as I have (many of the same make and model) its easier to give the notable ones a name for reference purposes. I dont really believe it is alive or for that matter even has a "personality" BUT a car CAN have a persona...and Lucy fits this one to a T!

 

[jetjock]

You could yank the starter, disassemble, and inspect. You can get parts for it at the dealer. If it were me the next time it does a no crank I'd hook a meter to the starter control line and see if it's active. Or you can hook it to the STA signal (same thing) at the ECU and leave the meter connected in the car, ready to tell the tale. Since it's such a simple circuit there are lots of ways to skin a no crank cat.

I realize naming inanimate objects is human nature but it's always struck me as odd. Machines have no soul. Nothing but metal, glass, plastic, etc. Guess I've been working on them too long. I look at my all stuff first and foremost as assets (or liabilities). I also view some of them as things that'll kill me in a heartbeat if I'm not careful. Therefore it's tough to become overly attached to them.

 

[Facime]

I know a surgeon and he looks at the human body that way. Replace the metal and plastic with bone and tissue. Obviously, (well to me anyway) people have a soul, but on the operating table he becomes detached from the fact that what he is working on is alive. Its a coping mechanism.


I dont think there is anything wrong with the starter. Im much more inclined to follow your STA line of thinking than anything at this point. I also think that with my fast approaching upgrade things will get better, or I will inadvertantly stumble across the problem. Like I said, at least the car isnt stranding me anymore.

Thanks again for the mental exercise JJ, its nice to have a rational mind like yours to bounce stuff off. I admit my mechanical nature is more of a "solve it with logic rather than procedure" than the other way around. I think thats why Im so stumped because this problem seems to defy logic.

 

[jetjock]

Rational? You must have me confused with someone else

It'll turn up. Could be something as simple as a funky clutch switch. Whatever it is once it starts leaving you stuck somewhere you'll become motivated to hunt it down. Remember, machine problems never get better, they only get worse. I wouldn't blow it off too long unless you're willing to put up with the hassle. As for the hot crank/no start that could be a loss of rest pressure. The system keeps fuel under pressure for a few hours after shutdown in order to raise it's boiling point and prevent vapor formation. It's no different than what your cooling system does when the engine is running.

Anyway, good luck with it. Feel free to PM if you have any questions.

 

[Facime]

but you see thats just the thing. It used to leave me stranded, and now it doesnt, and the problem has to some extent gottem better on its own. Well ok maybe not better...but different, and now its manageable.

In the next month Im upgrading AFM, injectors, headgasket, head bolts, exhaust and turbo. I imagine during that process I will discover more out about the car.

I guess what I hoped to accomplish with this thread was someone to give me some new directions to look for possible suspects. I will follow your suggestions though as time allows, and see if anything stands out as being odd.