Zeitronix

iTrader: (2)

You want to do diagnostics before removing internal organs, so check the pressure at the FPR with a boost gauge T-ed right at the FPR, then check fuel pressure before fuel rail and voltage at the fuel pump.

Spunkysandoval

iTrader: (7)

I just thought of another possible cause for this issue.

A pinched fuel line would be a great way to impeade flow. I'm going to take a look under the car tomorrw and inspect all of the fuel lines for any signs of damage and pinching.

Here are some of the electrical checks i'm going to start with

BigT

iTrader: (8)

This is a good indication that the fuel pump is being maxed out. The ecu is trying to compensate by letting the injectors work harder. Same thing happened to me on my 97 GSX when my fuel pump took a ****. I ran out of injector, 850cc, on pumpgas with just a 20g. I couldn't for the life of me figure out why. My fuel pressure gauge even indicated good pressure. But, the car kept leaning out towards the topend. I bought a new pump and everything went back to normal.

GST Motorsports

iTrader: (8)

Nice, only took them six years to fix that one.

Spunkysandoval

iTrader: (7)

Here are my findings so far:

Well, after analyzing all of the conditions and the data that was gathered, it seemed that the fuel pump was the source of the lean condition. The IPW (injector pulse width) and IDC (injector duty cycle) were both maxed out on the sock tune! Unfortunately, evoscan does not currently support fuel pressure logging, which would have been very helpful in pinpointing fuel delivery as the problem.

With this information as a starting point I decided to perform some basic diagnosis of the fuel delivery system. What I discovered was that the fuel delivery system on the Evo X has two voltage operation states. One state is at low load and it supplys the fuel pump with a low voltage that goes through a resistor circuit to drop the voltage prior to going to the fuel pump. The second fuel pump voltage state occurs at high load conditions, not necessarily WOT, but high load. I don't know what the various determinants the ecm uses or what strategy Mitsu used to initiate the high voltage state, but I do know that the ECM powers up fuel pump relay #1 (see picture below) durring higher load running conditions.

After gaining a bit of understanding on the system operation, I began to perform some actual diagnostic steps.

The first step was to verify any simple mechanical issues that would cause a fuel starvation condition such as:
1. disconnected FPR (fuel pressure regulator) hose
2. Pinced Main Fuel Deliver Line (the line that runs from the pump to the engine bay under the car)

Both of these quick checks were ok.

I then refered to the Mitsu Shop Manual for further guidence in diagnosis. I used the following step to verify the operation of the High Load fuel pump voltage state



Fuel Pump Relay is located here



This was the results of my bench test on the fuel pump relay, a bad relay. The resistince was OL, which on this meter means out of limits and that means the circuit is open



Just to verify the test that I was performing, I tested the horn relay, it was good. Resistance measured 0.0



That is all I have for now. I am going to swap the horn relay with the fuel pump relay since it is the same exact part as temporary fix, and I will replace the horn relay at a later time.

I will try and get some more log data with Naji very soon and post the results here with the relay replaced

SiC

iTrader: (1)

Glad you found the problem, Mitsubishi needs to order a recall on this thing, I've heard about this problem too many times. It can lead to engine failure which would be more costly for Mitsubishi to warranty.

Mad_SB

iTrader: (8)

Did you check the vacuum line on the fuel pressure regulator?

11.5, while it may be lean under wot for a stock tune, does not indicate the lean condition you would get if the fuel pump was not kicking up to high speed I don't think, I think if that were the case it would be a lot leaner, but check the other thread that was posted on that subject to see what the op's afr was under wot.

Next step is to pull the plugs and check the piston tops for uniform color. If one does not match the rest, you may have a bad injector on that cylinder.

Spunkysandoval

iTrader: (7)

^ problem has been isolated to a bad fuel pump relay.

nj1266

iTrader: (6)

Thanks for doing all this work to trouble shoot the problem. Please call me so we can get together and log the car to see if the relay is indeed the problem.

nj1266

iTrader: (6)

Today spunky and I logged an tuned his car. He nailed it. It was the bloody relay. The moment we swapped the relay, the car became pig rich during WOT. This is the first time that I was happy to see an Evo X running pig rich

There should be a sticky made with a WARNING in the title about this. This is a very serious problem for the Evo X that owners MUST be made aware of. This might lead, if it has not done so already, to a lot of blown engine.

I am very proud of Spunky. He took the bull by the horns and figured it out. I love people like that

Zeitronix

iTrader: (2)

Proves my theory. Measure the voltage after the relay instead of swapping the fuel pump. Removing internal organs is always much more work with potential destructive consequences. Diagnose first!

nj1266

iTrader: (6)

Thanks for the help As they say measure twice and cut once.

I read about the relay problem before, but this is my first encounter with it on an Evo X. Now I know what to do if I come across this again.

I cannot believe that Mitsu used such a poor quality relay on such an improtant function. The bean counters strike again

Zeitronix

iTrader: (2)

Well, you found out something is not right by diagnostics (logging), right?
I just want to make a strong point here to measure and think. I see too many times, swap a part troubleshooting techniques. Most of the time it creates more damage than good.
Glad to see it's working now. Back to making car fast now.

As far a relay quality goes. Too many times I have seen parts not meeting specifications. Mitsu engineers probably looked at the specs and decided the part is good to go. Even if they actually evaluated the pat sometimes the supplier down the road moves a production to a "lower cost facility" (wink) and the engineering does not find out till failures occur. Long story short, I had major semiconductor supplier move manufacturing to far east and lost the "recipe" how to make the part We used that part for a a well known networking giant. So, it's probably a good idea to bring this failure to Mitsu attention as they may not even know about it yet and trigger some attention.

Spunkysandoval

iTrader: (7)

How do we bring this issue to the attention of mitsubishi?

tephra

iTrader: (6)

dont buy Mitsu again.