At the last autoX, I noticed that in 2nd gear, the power would fall completely off at 6,500 rpm. It kept happening. 1st gear was fine. On the way home, I noticed that this same thing happened in 3rd gear as well, but earlier, at 5,500 rpm. If I lifted and stepped back down on the gas, the car would pull well again for 5 or so seconds before the "wall" would come back. I noticed that when this occurrs, the boost holds steady. The car has a boost controller, and I've done the "poor man's parrallel" turbo conversion to rid myself of sequential issues.

What do you all think could be the problem here? The boost is holding at a steady 10 psi, even when the car basically becomes n/a and quits accelerating. It's wierd that 1st gear goes fine all the way through redline, 2nd dies at 6,500 and 3rd at 5,500. Didn't try 4th or 5th, but I bet the trend would continue.

Any ideas on where to start?

 

Mods?

 

Clogged pre-cat or main cat to start with?

Also, a clogged fuel filter might produce those symphtoms.

 

Are you still on the stock ECU?

Does it make any sounds when it gets sluggish? Fuel cut will basically cut ALL power and make a loud 'farting' noise.

 

Mods are minimal. Cat back, and intake. Boost control set to 10 psi. I can't imagine it being fuel cut, as I've seen much cooler temps with higher boost (when setting boost controller) without ever hitting fuel cut.

The clogged cat wouldn't make sense to me. I'm somewhat skeptical of the fuel idea too, although I suppose possible.

I'm thinking a leaky vacuum hose somewhere. Any ideas on which one it might be to cause this symptom?

 

if boost is not falling off, then it's not a vaccum leak.

 

If the wastegate were opening prematurely would the boost drop off in the intake mani?

 

Yes. Well more than likely, but with your mods yes.

You can also boost 10psi with a leak somewhere, and that would cause it to lose power and also overly work the turbos. But you should be able to hear it leaking, it will be noticably louder.

 

I have noticed the same thing after my rebuild and I suspect it to be fuel related. It appears that I have one or more leaking injectors because of a shake at idle, and rich fuel smell, so I would not be surprized if they are failing at WOT.
It could also be an aggressive timing retard, but I am not sure if the stock ecu has enough range to cut power like it feels.
I will be going through the FSM to trouble shoot, and I'll let you know if I find anything.

 

The stock ecu may do exactly what you are experiencing if the coolant temperature is not close to 200 degrees. The ecu thinks the engine is still cold, and acts to reduce power at high rpms. I have had a similar issue. I would be driving on the freeway at night. Coolant temps at 180 for 20 minutes, and the ecu still thinks the car isn't fully warm. If you have the same problem when youir coolant temps are over 200, then you have a different issue.

 

If it help, there are a few CE codes

Code 5 - Knock Sensor open or Short Circuit

Code 13 - Pressure Sensor open or short circuit

Code 42 - Turbo Precontrol Solenoid open or short circuit

Code 43 - Wastegate control Solenoid open or short circuit

I never had trouble with the car doing this before, when warmer or cool. I thought maybe it was because I was running the cooling fans on the course, so I turned them off. And letting the car run warmer didn't help.

 

This is where we are different. I don't have any codes. You need to clear the ECU first (pull negative post, and step on brake for a good few seconds) and then go drive it again.

Those are a lot of codes and if they come back then you may have a messed up wire harness. The only place I can think it where the wires could be broken is at the firewall if the boot isn't seated.
You need to plug the Turbo precontrol, and Wastegate Solenoids back in, or wire in a resister that is 35K ohms.

 

When was the last time you replaced your fuel filter? A clogged filter might cause your problem.

 

SO I took the car out tonight for a drive after clearing the codes. Immediately when I turned on the key all the codes came back on except the pressure sensor code which makes me wonder 2 things.

First, what the hell is the pressure sensor and where is it?

and Second, why the hell is there a knock sensor code if I haven't even started the car. This makes me think that the knock sensor is bad regardless of if it's causing the problem or not.

And here's another monkey wrench. I took it out and drove it and of course the same symptoms occurred. SO I tried keeping my foot in it just to see exactly what was happening. When the power loss occured the boost dropped to about 7-8psi and fluttered there. I kept my foot in it just to make sure I got a good sensation of what was going on. The RPMs climbed, but painfully slow - as in NO BOOST assistance at all. And then, about 3 or 4 seconds later (sputtering started at 6500) by about 6,800 rpm the car came back online, boost up to 10psi and took off like a bat out of hell to redline. Tried 3rd gear, which loses momentum around 5,500 rpm, and held my foot in it. It seems to be somewhat of a pulse thing triggered more by time than anything. Pulls hard for a good 3 seconds then about 3 seconds of dead, then 3 seconds of pull, and then another 3 seconds of dead. THis is why 1st gear pulls fine, because it goes by so fast the cycle doesn't have time to begin.

Any new ideas generated? I'm now thinking it's NOT the fuel filter.

 

Sounds like a fuel issue to me. Could be that you fuel system can generatate only enough fuel for a short time. It runs out (or runs very low), and the engine stumbles until the fuel system recovers. I had very similar symptoms as I noted in an earlier post. Perhaps your ECU is to blame.

 

I just don't think it's fuel. It seems like it would have done this long ago if it were fuel. Temps are warmer than they were during winter and it never did this before. The lengthy delay in power also is bizarre. It seems like more of a mechanical systematic failure than just a fuel issue... If it is a fuel issue, are you saying that the car is just now not producing enough fuel and it's the ECU that's killing hte power?

 

In "MY" case, the ecu thought the car was not warm, so it limited (or changed) the fuel at high rpm. It sounds like you are having identical symptoms. Perhaps the ecu is doing the same thing to you because of the error codes.

 

When I first got my car last year, it would fall on its face hard around 6000-6500rpm when the fuel would get below a 1/2 a tank and then pick back up until redline. I changed my fuel filter and so far no problems.

So I would think fuel problems also if you can't find anything wrong the solenoids/vaccum lines.

 

Well, taking this one step at a time, I decided to see what's up with the knock sensor. I found that the sensor was not plugged in. The problem is that I can't find the plug from the harness side. Can someone help me track this guy down? Where's it exit the main harness wiring from and usually where in the engine bay do they hook up?

 

I think it is a shielded grey wire that plugs in to the emisions harness right behind and at the bottom of the alternator.

 

I'm just not seeing it. There's a wide harness back there. I even removed the alternator. Does anyone know where it physically exits the harness. I can see it on the diagram in the wiring section of the FSM, but there's no substitute for real-life input.

 

You said you pulled this code:
"Code 13 - Pressure Sensor open or short circuit"
Which is the MAP sensor. Definitely check the harness continuity and the sender itself, maybe swap with a spare if you can find someone to loan you one. It seems to me that the TPS and MAP sensor parts of the wiring harness are easiest to damage - I hope that you don't find the problem is in the harness or else you might need a whole new one.

I assume the wg and pc solenoids are showing codes because you are running a boost controller and they are removed. Those codes are ghost codes anyway and harmless.

According to the FSM (page F-178) the knock sensor shouldn't cause the symptoms you describe. The fact that the car sometimes runs fine without being connected reinforce that.

Adam's idea about the temp senders is very good to check also.

Dave

 

I pulled the codes the other night (after the drive where the issues were exhibited) and found that after clearing the code for the pressure sensor prior to the drive, it did not come back during the drive. So I think I've taken that one out of the equation, at least for now.

I've got the intake and all peices off right now trying to find anything that looks broken, and locate the wire for the knock sensor wire.

 

Well, verdict is in! It WAS the knock sensor. Got the car put back together (took a while, waiting on the right size belt for the air-pump delete) tonight. THere is no C/E light on. I checked the code, and the knock sensor code has gone away, and only the old 2 codes remain, which were always there, but didn't illuminate the light.

Result:

FIXED! I could believe it. It's been too long since I've felt the power. (I recently bought a hybrid, so even my SUV feels fast these days.)

ROlled up next to a new GT vert tonight, and really wanted to give it a go, but my passenger wasn't soo keen on the fun. Oh well, I'm thrilled. I'm not sure this would be an isolated incident, so if you're on the stock ECU, make sure you have the knock sensor plugged in!

Thanks for all the great ideas and input!