Jump to content
LegacyGT.com

Gas mileage pretty bad


Recommended Posts

Hello, first post here. I just bought an 08 spec b. Thought I was getting 20mpg as my fuel cons says but when I calculated it, I’m barely getting 11 mpg. Below is a list of what has been done to the car. Im wondering if this is normal for a car with this build or if I should start hunting for issues. Car runs pretty good. Thanks for any input.

 

70k chassis miles. 7,500 engine miles

-year 2008

-6 Speed manual

-COBB Top Feed 1050x Fuel Injectors

-Stock Rebuilt EJ255 Spec B w/ race bearings

-11mm oil pump with Stock heads by Outfront Motorsports

-Custom turbo upgrade VF52 20g PNP and Powder coated with upgraded 1bar/14psi wastegate by SoCal porting

-Modified stock TGV Deletes PNP by SoCal Porting

-Process West Vert Mount TMIC

-Mach 2 Recirculating valve

- Invidia Big Tube Header - INV -HS05SW1HDP

- Invidia Downpipe Catted Divorced Wastegate - INV HS08SW1DOC

-Invidia Q300 Catback Exhaust INV HS05SL1GT3

-Mishimoto Performance Aluminum Radiator - MIS MMRAD-STI-08

-Mishimoto Radiator Hose Kit Black Subaru MIS MMHOSE-STI-08BK COBB

-Top Feed 1050x Fuel Injectors (312150) -COBB Silicone Turbo Inlet (724400-BK)

-COBB SF intake system (724100)

-COBB SF Airbox (724150)

-COBB 3-Port Boost Control Solenoid (BCS) (715750)

-AEM High Flow Fuel Pump (315450)

-Grimspeed Phenolic spacers 8mm NGK spark plugs GK-2309

-Stock Sti Cluch and Flywheel,

-Air Pump Delete

-tune by Clark Turner

Link to comment
Share on other sites

A data log is a good idea as I see it in order to see if the system behaves well or if you have a problem. A compression test wouldn't hurt either to make sure it's a healthy engine.

 

 

Fuel consumption is strongly dependent on driving conditions too, so you need a 'normative' measurement where you can do a constant speed drive to get comparable values before and after corrective actions.

 

 

As usual - cleaning maf, ensuring you have a good oil, fresh filters etc. all helps. Add to it - make sure no brake is dragging - abnormal heat at one wheel is an indication.

 

 

O2 sensors can also be culprits, and considering the mods it's possible that some DTCs can be disabled.

453747.png
Link to comment
Share on other sites

You neglected to mention the super important factors:

 

-Driving style?

-Driving conditions?

-Tire pressure checked?

-Brakes checked?

-Roof top accessories?

 

In the right conditions (middle of winter-using lights, heaters, defrosters, warm up time to thaw glass, etc) of when I lived in Chicago, my plain Jane 2003 NA Outback could get below 12-15MPG on a city tank.

Link to comment
Share on other sites

Well it’s pretty cold so there’s a lot of warm up time. It’s a mix of city and highway. There’s a Datalog above but as AWD pointed out it’s missing afr. I have already found a vacuum leak and want to do a boost leak test on it. It has a 14psi (at least that’s what the previous owner states) wastegate but maxes out at like 12.5 so... Edited by boostaru
Link to comment
Share on other sites

I have a similar setup but 18g turbo on a Ej257. My ECU reports 37mpg but it is def much less. I assume it has to do with ECU thinking we are using stock injectors with very low duty cycle but we really have larger ones so we are actually using more fuel. I think I am getting around 15-17ish highway.
Link to comment
Share on other sites

Front O2 sensor. It can go bad without setting a code. Definitely will affect fuel mileage.

yup, came here to say this.

Even with a big turbo, I can get mid 20's driving light on the highway and combined around 18-20. Recently had an uppipe leak that was throwing off the front O2 and dropped to about 11mpg.

 

Winter gas also lowers mileage

Link to comment
Share on other sites

Alright so I’ve started to see some of the issues with this car after driving it more and it could all be pointing to the same issue. So when I hold the throttle steady, it fluctuates from like 1700-2000 rpm. Then when I let off, sometimes it almost stalls then recovers. From the research I’ve done, it points to vacuum leaks. I did find one and it didn’t really fixing the issue. I’ve also reflashed my tune and tried bumping the idle up a bit. I’ve also cleaned the maf. None of that seemed to do anything. I’ve kind of rulled out vacuum leaks but who really knows. I picked up a denso O2 sensor. That’s OEM brand right? I’ll let you guys know if it helps. Any suggestions before I dive into it tomorrow.
Link to comment
Share on other sites

Front O2 sensor. It can go bad without setting a code. Definitely will affect fuel mileage.

 

 

That's why I suggested a data log - it could give some good hint if there are O2 sensor problems.

453747.png
Link to comment
Share on other sites

I believe I reset the ecm. Maybe not. should I disconnect the battery for 10+ mins for that? also, I did a data log in the rpm range that I have been experiencing issues (not the way the COBB video recommends) and did some deeper research. I found my AF corrections are way out of wack in this range. Cobb says...

 

"Negative corrections beyond 8% mean the ECU is pulling lots of fuel. This would most likely be due to a leak in the intake tract. Other potential issues could be a bad MAF sensor, failing o2 sensor, or you may have a map flashed that is not designed for your intake or injectors."

 

So deeper down the rabbit hole I go. Things I did today in addition to the O2 sensor were clean the PCV and inspect the throttle body (clean as a whistle). check out the latest and greatest datalog..

datalog20.csv

Link to comment
Share on other sites

How about the donut gaskets? Are they replaced?

 

 

One way that can be used - get a spray can with brake cleaner, start the engine when cold, spray at suspect points like joints and see if the engine do react. Use small squirts so you don't soak the engine in flammable liquid - it should be enough.

453747.png
Link to comment
Share on other sites

If I'm not wrong there shall be two O2 sensors, one before cat, one after. What I could see from the log the one logged seems to do OK.

 

 

But in addition to the logs I would like to see a "normalized" fuel consumption check by actually filling up the tank at a gas station, then drive highway only at a constant speed like 55mph on cruise control (aside from going in/out of the gas station) for some while (take an hour for this) on a reasonably flat highway, then go in to the gas station and fill up again at the same gas station with the same pump. Then calculate the fuel consumption from the driven distance and the amount of fuel filled up.

 

 

That way you would have a reference consumption to share with us that can tell if you have a real problem or if it's a driving style problem.

 

 

This shall be done on a warm engine/transmission so make sure it's well warmed-up before the test. Notice that it takes longer for the transmission to warm up than the engine.

Edited by ehsnils
453747.png
Link to comment
Share on other sites

Ok that is a task in Colorado where I live but I agree, the norm needs to be done. By donut gasket do you mean intake manifold gaskets? If so, I have soaked them in flammable liquid.. only found one vac leak this way. I know the donut gaskets are a common one but this engine has 7500 miles on it. Not only is it obvious but I have paperwork to back it up. Still not a guarantee but I’m convicted. I’m not ruling it out just putting it on a back burner.

 

The A/F passes but did you look at the corrections? Them puppies out this world. At least from the research I’ve done it throws a red flag that something is in need of attention.

Link to comment
Share on other sites

Air fuel corrections can scream vacuum leak can they not? The ECM is constantly adjusting for "X" air and if inconsistent, such as a leak allowing unmetered air to enter the equation your AFR will be all over the map. O2's only tell the ECM what they see. Sounds to me like a genuine smoke test needs to be performed. Am I thinking right? Trying to understand this myself and assist the OP.
Link to comment
Share on other sites

Seems reasonable - time to do a detailed check for a vacuum leak. If the PCV was just cleaned - try to replace it with a completely new, the current may be erratic in operation.

 

 

Also look for other leaks. And as someone noted - also on the exhaust system.

453747.png
Link to comment
Share on other sites

Vacuum/intake leak or pre-front O2 sensor exhaust leak. If it's leaking in intake, then your ECM will correct for too much air. If it's leaking pre-O2 sensor, then same thing. The O2 sensor reads what the engine is doing by calculating the amount of air versus the amount of hydrocarbons from the burnt fuel, so if there is a leak before the O2, it can make the ECM, again, read too much air and increase fuel intake to compensate.
Link to comment
Share on other sites

Alright so today I pressurized the entire system to about 10psi. I found another vacuum leak. Before I fixed it the vacuum was -7.55 now it’s at -8psi (-16.2 inHg). Is this about what it should be in Denver? No fix yet though.
Link to comment
Share on other sites

hard to say without seeing the learned fuel corrections by air flow range. that can help tell you a lot about where you are losing air. you may also have a boost as well as a vacuum leak. is it still on a factory intercooler? Do you have oil seeping at the crimp seams? TB hose seated well?
Link to comment
Share on other sites

Alright so today I pressurized the entire system to about 10psi. I found another vacuum leak. Before I fixed it the vacuum was -7.55 now it’s at -8psi (-16.2 inHg). Is this about what it should be in Denver? No fix yet though.

 

 

Seems you are on track then for locating the problems you have. The leak you found seems to have been a considerable leak.

 

 

Vacuum values are always a hit&miss though - for the same engine under the same conditions they are useful. Between similar engines they are also reasonably useful, but different model years and different mods impacts the result - and so do the engine temperature at which the vacuum is measured.

453747.png
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...

Important Information

Terms of Use