Figured I'd start a thread on here describing everything and quit bothering the SARX guys on Facebook since were not really getting anywhere and could use some fresh ideas.

First, the car is modified, here is a build related to the motor:
AEM intake w/ BHR TB Spacer for nitrous (NOS lines have been taken off and spacer has bolts reinserted so the system is not involved with issue currently)
BHR Coils, MSD wires, Stock Plugs << coils are new, same wires I had before, new plugs during rebuild
Full exhaust, no cat
Air Pump Delete
S2 trans and ACT Prolite Flywheel
Fuel and Oil Pressure Gauges
~1800 or so miles on rebuild

Long story short, I had the motor rebuilt about 1800 miles ago and the car ran perfectly normal. Around that time decided to get the car slowly into the higher RPMs and noticed I was running into the same 7.5k hesitation I was before the rebuild. I had the car tuned by MM before the rebuild and this is when that hesitation really came to life. Got the motor rebuilt, same hesitation, so I decided to just unmarry my Cobb completely to ensure I had no tune whatsoever.

Great idea, car no longer runs right.

Reinstalled the Cobb, and I am consistently getting a high negative STFT once the car cold starts, then after a few when the car drops back down to roughly 1-1.5k the car sits a constant -6 to -7% STFT and you can hear the exhaust popping like its drowning. Then drops into a very rough idle at the same STFT.

If I pull the car out and begin to drive, it runs and moves no problem (roughly 5-25mph) but if I take my foot of the gas and the car returns to idle it just goes straight to 0 rpms. Doesn't even bother idling.

Notes:
I do have a fuel gauge, constant 55-58PSI when running.
If I drive the car at 20MPH, about 2.5k RPM the STFT remains 0% just idle seems to be very off after first starting.
No CEL.
Brake stomped many times, battery reset a few as well.

This we have attempted:
I have replaced the stock coils with BHR coils and wires. Tried stock again. Back to BHR. No change.
I have two MAF sensors, both read the exact same.
Unmarried Cobb again, tried cobb tune, and a different MM tune just to see if it would run.
Smoke test through brake booster line. No leaks found.
Tried stock intake, same issue.
Assures AEM screens were straight 100 times.
Compression test came back well over failing.

Things to yet try?:
Take car to dealership for ECU reflash back to stock?
Pull injectors and send for cleaning?
Get new o2 sensor, but it seems to be reading and not stuck?
Burn car?

Kinda stuck on what is left to try..

 

Possible you have a leaky injector? It's trying to trim the fuel back...

 

It's possible, I'm just not sure of the likelyhood of injectors leaking on these cars

 

Would definitely be looking at the fueling system, even if there is no code for lean or rich you could have a fuel pump that's on the way out, or like others have said leaky injectors.

Also see if you can try a complete ECU reset and have the fuel trims completely re learn (keep in mind you generally have to go for a full drive cycle before it learns what is going on) the new parameters, good luck!

 

It's not car related, but injector related. Half the stuff in our Mazda is Denso / Mitsubishi.

 

The bolts you used to block the N20 ports on the spacer, any chance there is a leak there, did you use thread sealant? Dish soap and water, look for bubbles.

 

Well the chances of a Denso / Mitsubishi injector leaking? If they've proven to be reliable and the chances of them leaking are 1/500 then of course I'll look elsewhere until it comes down to that.

9k, the threads were wrapped with thread locker. I brought the original length bolts home yesterday so I will just remove the entire spacer today. Just trying to get back to the bare minimum until the issue is isolated.

 

Ah ok, definitely check the injectors and seals.

 

Decided to make a video to show the specific problem. Just a quick drive before I tear it apart. The car drives very smooth, doesn't feel sluggish, but as I slow down you can hear the car slightly backfiring as I suspect that's when the negative fuel trims come in as the computer adjusts. When I first fire it up it idles, I slightly rev and the car will idling cold just fine. It is only once the car is warm and I actually drive it. The car starts right back up if I crank it, but I need to keep the rpms up to return home.

 

What were the compression numbers?

 

Fwiw, negative fuel trim means the pcm thinks it's running rich; some of the suggestions listed above are for it running lean, not rich, so just clarifying the situation.

You also need to review fuel trims at various rpms because there are multiple trim settings; idle, cruise, wot, etc. The results there can possibly help provide some direction on where the issue may lie, assuming it has anything at all to do with the engine shutting off, which it might not ...

 

IIRC he was getting them sweeping + and - but his post here only says negative so if he could clarify that would help for sure.

 

The compression test came back with every face showing over 8 at 260-270rpms if I remember correctly. I could double check exact numbers.

To clarify, I was getting numbers swinging back and forth originally. I had found that the MAF sensor has leaks where a mechanic used much longer and different screws to tighten down the sensor. Since then I have fixed that and the numbers do not swing nearly as much. They do still go in each direction, but negative seems to be 90% of the time. these numbers appear when driving slowly and idle when the car first starts. I have not tried WOT since I was not sure if that would cause damage to the recently rebuilt motor and current issues.

I did however take the plug wires back off and tested the MSD wires from BHR. 3 of 4 wires test about 40ohms and 1 tests 310ohms. I'm not a pro on plug wires, but I'm assuming that wire has seen better days?

 

Compression sounds solid based on those numbers.

I would say you have a bad wire if your test was done right.

 

Are these wires possible to get locally, or are they all custom to the kit? I need 1 I suppose

 

Nah any decent speed shop should be able to make you one.

 

God this car is a pita. $30 for a damn wire

 

Okay. I checked all plugs, replaced the wire, nothing usual going on there. I have to keep brake stomping in order to get the car to start. Seems to be getting -5 to -6% stft however again I'm just let the car idle for a while, which it does fine as long as I don't move it, I began to increase rpms and let go and while idling it will catch itself for the first couple times then won't a few tries later. I know when I picked up the car after the rebuild he said a slight rpm dip was normal due to the flywheel change.

I did however keep the car running this time by holding the gas and look through the cobb. Things I noticed were: stft is -5 -6 idling, if I increase rpms its 0. If I let go, they go crazy. Around 600 rpms the trims jump all the way to roughly -18% if I hit the gas they even back out. AFR readings seem perfect, in the 14s while idling, and move normally through the rpms. What stands out to me is the MAF sits 1.3-1.6v the entire time regardless of rpms. I'm not sure if that's normal for idle but I thought it moves 0-5v??

 

-5 or so isn't any issue. The MAF output should be somewhere around 1.13V - 1.17V at normal idle speed and increase with airflow/rpm from there. Maybe try fitting a known good MAF sensor and see if that resolves the issue; do you know someone else nearby with an RX8 that you can swap theirs in with? Also a can of carb spray can help figure if you have intake leaks; rpm will surge if it sucks in the carb spray through a leak spot.

 

I have two sensors, one which was the original, the other a denso purchased last year. Same results from both. It was leaking and giving erratic readings originally so I actually rtv sealed one to the aem intake for the hell of it. That's when it was much more steady on negative fuel trims. Figuring it may be a bad MAF I then sealed the other. Same results. I'll try some carb cleaner and propane just to be safe, but I'm not counting on anything

 

What intake is on it? I've seen the screens get out if whack on the AEM/Mazdaspeed intake and cause problems in the past.

 

Also, are you doing your own Cobb tunes? The first MAF calibration cell can't be changed or it causes problems too. Otherwise it seems odd that your MAF reading is so high if the sensor is good unless the MAF map was altered.

 

It is the AEM intake and I have verified the screens are not bent and in the correct position. Yesterday I swapped out the throttle body and I am able to drive the car without it dying. However, it is very rough to start but once running it seems fine. I am however, still getting trouble in the upper RPMs as I was before and after the rebuild. I am going back to the dealer today to see if they can reflash the car. I had MM tune the car, but noticed the hesitation really shine once he has began tuning it. I'll see if anything changes if they can reflash it.

 

Uninstalling the Cobb should have reverted it back to stock.

You need to get some fresh (qualified) eyes on it.

 

Man I have seen these kind of thing sooooo many times, people swear they went over everything and then there is vacuum cap on the LIM or something stupid like that.

That antonio guy on the SARX FB page went thru something similar, had two shops look at it and then he brought his car by the office and his jet air hose was disconnected.