2017 STI Fueling Issue

hi,

I'm tuning two 2015 STI's with identical setups (ID1050 injectors, AEM 340 pumps hardwired, Tomei type s FPR(43.5psi base pressure), Down pipe, stock turbo) and I'm facing fueling issue at WOT around 5k rpm onwards the fuel pressure starts to drop. checked for leak, voltage drop at pump(13.5V), amp draw(12.8 amps@58psi) everything looks ok. i suspected the FPR might be faulty because i noticed pressure fluctuation while logging fuel pressure so i swapped it with known good Cobb FPR it behaved the same but worse made the engine to run lean 5k rpm onwards. a year ago i tuned same model STI similar setup only difference is bigger FPR(aeromotive) and bigger turbo and it didn't have this issue. the issue starts appearing only when the engine is worm after two to three WOT runs. i attached a video of the pressure gauge while WOT run with Cobb FPR and attached logs of Cobb FPR run, Tomei FPR with pressure sensor, and a static pressure test where i ran the pump and started ramping up the pressure up to 5 BAR to see if i loose pressure. could it be a resonance issue because the fuel system doesn't have a pulse damper or is it an issue somewhere else.

The fuel pulse issue is a big problem while cruising and at light to moderate throttle so I definitely suggest a proper line and rail setup to resolve that, but your fuel pressure drop at WOT is a separate issue.

I've probably helped over 100 shops having fuel pressure drop solve it on 08+ Subarus by fixing the fuel pump install method.

Which o-rings did you use, and how many?

Did you transfer the plastic "cage" from the stock pump's outlet nipple to the new one?

What order do you have the o-ring(s) and plastic "cage" on the pump outlet nipple?

First car I used the same o ring I removed from the oem pump it was in very good condition. The other car I found the bring was damaged so I replaced it with the one supplied with the pump. I used one o ring each and yes I used the plastic cage. Put the cage in first then the o ring.

The correct stackup is cage first, then 2 o-rings on top of it. The AEM supplied ones aren't the ideal size, but when you use both you'll usually be alright.

I put superlube on them for install as well.

Thanks I’ll try that and let you know what happens. Dose this applies even if I use the oem o-ring? I’ll have to stack two o-rings?

I don't suggest using the stock o-ring with the AEM pump. I've not seen that work out reliably. 2 AEM o-rings on top of the OEM cage has been the combo for success with the AEM pump in that vehicle application.

For what it's worth, the same goes for the DW pump in that application.

Just be careful, my old shop and a large dealer of parts I work for have both received incorrect sized o-rings in the correct part # AEM pump kit for that car. It's not common, but has happened, likely just human error sticking parts in the box, so make sure they seem to fit well.

Thanks Mike for the informations. I did stack up two o-ring I have attached a photo but didn’t test the car yet. Unfortunately I only found one o-ring with the pump we lost the other one. so I had to use one supplied with the pump and the other used the oem o-ring. Please see the attached photo.

Hi Mike I just tested the car and it’s definitely better before on the third run it will start to go lean at high rpm now on the 6th run it will go lean but not as bad before it would go up to 12 afr now low 11 and the target is 10.5. As you mentioned maybe it’s the oem o-ring is not reliable and I used it because I had only one aem supplied o-ring. What do you suggest to check/do next.

I certainly understand that's the best you could do with the parts you have. I would get the correct o-rings in there just so that part of the system is as it should be to help narrow things down and have higher confidence it won't be an issue.

Given the duty cycle on 1050s, the pump, you shouldn't be anywhere near the limit of the pump.

Beyond that, if the fuel gets hot enough, turbine pumps reduce flow dramatically as components of the fuel start to boil and cause cavitation, so that's another potential cause of what you're experiencing, based on your description of this happening after multiple runs.

If you have relatively low fuel level so you can service the pump without fuel dumping all over the place, that can allow fuel to heat up more quickly.

Lastly for now, if that particular pump gets run hot or dry with low fuel level a single time, I've seen dozens of instances where the pump can never flow what it used to and needs replaced. I tell customers to keep their tank over 1/4 tank for normal driving, over 3/4 tank during aggressive driving.

Hi Mike thank you for your support I did change the o-ring to aem one and same thing no change. I ran the car between cruising and little aggressive driving for one hour just to get everything as hot as possible and immediately checked fuel temperature at the pump outlet from the tank and it was worm but definitely not hot to a point where it would cause pump cavitation. I did another check where I pinched the feed line at the engine bay and raised fuel pressure to 70 psi and let it sit for more than and hour. The pressure did drop around around 10 psi and stopped there I this this is because the engine bay was hot and after it cooled down the fuel pressure dropped but after that it was holding around 60 psi so the pump check valve and the o-rings are not leaking and able to hold pressure. Pump electrical system from the battery to the pump checked again and no problem there no voltage drop found. Voltage at pump 13.6 volts and all connections are good.

I can say with confidence the OEM FPR that COBB uses is pretty agnostic to temperature and pressure will vary around 0.5 psi from cold to hot, which is a great deal less than the average adjustable FPR at 2-3 psi.

Do you have a singe log you can share which shows you performing consecutive pulls until pressure drops?

Any change the reference line to the FPR isn't 100% secure and it leaks more when hot?

Unfortunately all runs was done with Cobb fpr which I didn’t have fuel pressure sensor installed. And I’m sure there are no leaks in the reference line because I’m using silicon lines and they are tight on the barb plus zip tied. I did test something yesterday and it changed how the fuel pressure behaves. I installed a bypass line between the feed and return pre fpr and installed fuel pressure sensor. Fuel pressure oscillations went down by 50% and fuel pressure is holding better than before. Fuel pressure drop at high rpm was between 5 to 10 psi as heat builds up. Now it only drops 3 psi and holding. I think the issue is related to fuel pressure oscillation due to resonance. That causes cylinders 2 and 4 to be leaner than 1 and 3 because of how the fuel system is routed and once the issue starts to become worse the pressure waves prevent fuel from getting to 2 and 4. The bypass line I installed will fill that gap. Now the car runs a lot better. Even the stumble issue around 2500 to 3000 rpm is not as bad you can barely notice it you can tell only from the aft as it goes lean in that range. Before it would go up to 17 afr now with the bypass only 15. I think the solution here is to upgrade the fuel lines in the engine bay plus fuel rails and pulsation damper. I hope it makes sense and if you have comments on this please let me know.

The 08+ STI has major fuel pressure issues and should never be operated with bigger injectors without a proper rail and line setup, so I thought having those items was assumed, but yes the engine will not run properly without them.

Adding dampers will likely make things worse, since every damper tested when I created the COBB solution did make things worse. I don't suggest just throwing parts at it. I'd get the COBB rail, line, FPR package, and you'll be fine. You'll also be able to zero out the ridiculous load comp tables which are Subaru's attempt to bandaid the problem. Those tables include trims up to 20% to reduce the lean misfiring an up to 30 psi differential pressure drop causes.

Here's an example of a 30 psi variance in differential fuel pressure (which should be stable within a few psi as MAP varies), during a gentle partial throttle acceleration on a stock 08+ STI, courtesy of COBB Tuning.

Differential pressure is stable within about 2 psi with the COBB fuel kit (rails, lines, FPR).