Need imediate help, Bad miss after hot starts: Paging JJ

[GC89]

I dedicated this whole week to figuring this issue out and if I cant the car might have to stay parked this whole year so I appreciate any help you can give me. I am running stock fuel system and turbo, maft pro in speed density, I have checked all wiring and settings.

Symptoms-

The car runs great until it is shut off. I have taken it on up to hour long drives with no issues, great in boost as well as cruising and at idle. As soon as the car is shut off after running long enough to reach normal op temp, (10-15 min drive or idling for 20+) it misses bad when restarted. Sounds like it is dropping cylinders all through the rpm range. Closest sound discription is a subie flat 4. Also the first few times I attempted to make it do it on purpose to diagnose it seemed to need to sit at least a few min in between restarting. I am also getting a code 24 and 35. My best guess is something ignition related due to acting this way all the way through the rpm range but I havent ruled out fuel

What I have swapped out-

Ignitor
Coil pack & plug wires
CPS
O2 sensor

Currently supplying 2.46v to hac, 3.59v to tha for temp sensor. If I have read correctly the temp sensor will assume 20C ambient temp if the voltage is incorrect so that should be no problem. Jetjock may be of help here, what sensors are there that affect ignition and or timing? Also what sensors would only affect the car upon start up and not if they threw a bad signal later while it was running. My best guess is a sensor or electronic component that is becoming heat soaked and causing the timing to jump or cutting ignition.

I am at a loss here and only have through this weekend to really focus on this. After that I wont have much time between work and class and it may have to sit. I can provide a log tomorrow of before and after this happens and or a regular video if it helps

Thanks Guys

 

[GrimJack]

Check the VSV that's connected to the FPR... or just disconnect the FPR next time you hot start it, see if the problem goes away.

 

[jetjock]

^ This. Also check rest pressure...

 

[GC89]

by disconnect the fpr, you mean the vac line on it right dave? Ill see what I have to check fuel pressure JJ.

Also I started trying to get rid of the code 24 and 35 just as a start and cant. I have tried sending every voltage imaginable to tha via v-out1 as well as 0v and put a 2.2k resistor between tha and E2 on the afm plug nothing satifies it, any ideas? is it even possible that an incorrect hac signal could cause this? isnt that what the code 35 relates to?

 

[jetjock]

It sounds like vapor to me but until you check rest pressure it's all a guess.

The HAC shouldn't cause this symptom but it still needs to be fixed. The 24 may be due to the Maftpro. Something about how long it takes to initialize. From what I understand it's not an uncommon issue. Might shoot a PM to member CRE, iirc he's familiar with the problem.

 

[GC89]

I have it set so it powers on before the ecu with acc and ignition so Im not sure why its still throwing it, it is also recent. Ill focus on that later if it cant cause the symptoms though. By rest pressure you mean pressure at the rail after the car has been run and shut off?

 

[jetjock]

Yes. Because the rail is above the engine the fuel in it can boil after shutdown. Happens after about 15-45 minutes and makes for hard starting and funky idling until the vapor gets cleared. Rest pressure prevents vapor by increasing the boiling temp of the fuel. Same theory as the cooling system. A quick way to check is to jumper the fuel pump on in the diag block for 30 seconds or so before attempting a hot start. If that resolves the symptom rest pressure is likely to be at fault.

 

[IJ.]

I have the MoTeC in mine set to run both pumps at 100% to clear the lines for hot starts it made a HUGE difference, maybe a cap/diode/button to act as a primer on hot starts?

 

[GC89]

Just rigged up a pressure gauge, from a cold start its showing 33-34 w vac attached 40-42 without. Sounds like that could be the problem the only issue is it doesnt clear up. I thought I had it fixed yesterday and took it for a drive to the golf course. drove 15-20min parked for 2 1/2 hours and it did it upon start up. messed with a few settings in the maft and let idle for 10 min before deciding to limp home. Even after the 15-20 min drive home 10 of that on the highway it was still stumbling the same

back with rest pressure soon

 

[jetjock]

Well, if it's doing it after 2.5 hours it's unlikely to be a loss of rest pressure.

Ian: That's the way to do it...

 

[GC89]

It seems to be temp dependent. Usually after 2.5 hours if started again it will behave normal, but that was when it was cooler. It would usually require at least an hour so cool down on days in the 30s-40s before it would start and run normally again. On warmer days like yesterday (high 60s) it seems to have to sit longer. This is why I was chasing sensors looking for one that was getting heat soaked and failing. I dont understand what sensor could be affected by heat but not cause it to start running this way durring a drive, only after a startup. What components cannot change the way it is running once started but cause issues if giving a faulty value or no value upon the ecu booting up? I have also pulled plug wires one at a time while running this way trying to isolate a dead cylinder with no luck, each one caused a drop in idle. I also considered an injector failing when hot but that would also be only one cylinder and would not have changed when the plug wire was pulled

 

[jetjock]

How long does it take for this to clear up?

 

[GC89]

Usually an hour or more, it seams like the whole bay just needs to cool off. Like I said it seems to be temp dependent. The majority of the time a 2.5 hour cool off would be enough but it was warmer than it has been yesterday.

 

[jetjock]

I mean once it happens does it ever clear up during driving or must you wait until it cools?

 

[GC89]

As far as I know It has to cool off. The longest drive I have ever taken it on in that state is about half an hour just to get it home

 

[jetjock]

Need to figure out if it's fuel or spark. If it were here the first thing I'd do is put it on the ignition scope. You could check each sensor when the engine is doing it to see if they measure what their supposed to. Not going to be the O2 sensor. If EFI related it's probably the load or rpm signal and since you changed the CPS that leaves the Maftpro sensor.

Or you put a bad CPS on. I've seen more than one randomly act up when heated. I'm leaning towards the load signal but I dunno. I usually don't "do" modified cars, at least not over the internet. Not to mention problems like this need good test gear. That said these kind of things usually aren't related to the EFI system. In fact most problems aren't.

 

[GC89]

Wouldnt you know today it wouldnt do it for me. Took it on a 5 min drive and let sit then a 10 min drive, and then 15+ and still nothing. Seems like the fuel is ok, holds 40psi at the rail after shutoff. Tomorrow Ill take it on a long 40 min cruise and test the map sensor output while its doing it. If that isnt it Ill take a multimeter to the cps to be sure. If both those check out what would you look at next JJ?

 

[jetjock]

Multimeter isn't going to cut it. Need a scope, preferably a DSO.

Forget the other sensors. Why? Once hot the engine doesn't need but the load, rpm, and knock signals to run way better than what you're describing. With a stock ECU the engine will even run well without a load signal, at least up to 3000 rpm. The others are trim sensors and do little more than handle cold starting, emissions, idle control, and some very minor performance issues, most of which aren't even noticeable.

Put another way the ECU is easily able to get the mixture within 1% of optimum without them. Seriously. Any stock GTE that won't run near indistinguishable from normal when hot with the coolant, O2, and TPS sensors disconnected has something else wrong with it. You may even want to try this. Unplug them after the engine is hot and when behaving normally. It'll let the ECU run on default values. If one of the sensors is shifted that'll tell you. Or you could measure them.

The other thing you may want to do is cycle the important EFI connectors. Better would be to spray each one with something like ACF-50, Corrosion X, or Deoxit while they we're apart. Carefully inspect the wiring at each one and also verify good ECU grounding. Electrical maintenance is one area usually neglected on these cars.

 

[GC89]

Once again couldnt get it to do it for me today. I changed the voltage to the hac, still have the code, but that couldnt have fixed it. I also disconnected the cold start injector to hook up a gauge. I started thinking maybe the coldstart injector was malfunctioning and flooding out cylinders 3 and 4. Hooked it back up and still ran fine. Tomorrow I think I will take it on a long hour or so drive and if that doesnt do it move on until it happens again. Am I correct that placing a 2.2k resistor between THA and E2 in th afm plug should get rid of the code 24? that is the next issue I will be addressing.

 

[GrimJack]

Pretty sure that's how I ditched my code 24... although to be honest I'd have to go outside and check the connector to be certain.