Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

2012 WRX - Verifying Factory O2 Vs Wideband AFR Accuracy + MAF Calibration & Base Map Boost Table Starting Point

Practical Reflash Tuning

Forum Posts



Tech Articles

Discussion and questions related to the course Practical Reflash Tuning

= Resolved threads


Afternoon team, have a bit of an interesting setup that I am working on right now and am trying to get myself situated so I can start working through the rest of the Six Step Process for street tuning but feel like I hit a mental block in terms of properly comprehending the MAF calibration / Factory 02 vs Wideband reading accuracy and need to bounce my specific scenario off someone along-with what I am getting hung up on, so hopefully I can firm up what I've learned and mentally correct any misconceptions I have currently.

I have a 2012 WRX that I've been doing some serious work on over the past 3 years, prior to this WRX I had an 05 that I had started to get into modifying - however I was heavily reliant on OTS maps and/or E-tunes with the 05 and wanted to start learning the tuning process myself as I knew the risks I was taking with both solutions. I started with a standard FBO and had it Protuned by one of the most reputable tuners in the region, Matt Miner - the FBO setup consists of the following (all components are still installed):

Invidia Dual N1 Catback Exhaust

Grimmspeed Limited Divorced High-flow Down Pipe

AEM 340 LPH Fuel Pump

Grimmspeed Top-mount Intercooler

Cobb SF Intake w/ After-MAF Silicone Pipe + Airbox

Accessport w/ Accesstuner access

--- car made 300/301

Over the summer I decided to do the next wave of upgrades, there were some additional supporting mods I did such as Cyl 4 Coolant mod, KillerB Pick-up, fresh STI pan etc. but all tuning-specific mods are listed below. Now here's where I'm getting hung up...

During the course work there were a couple sections where it was mentioned that it's strongly recommended that we do not try to tune with changes to the Intake changes that would require MAF recalibration and changes to the Fuel system at the same, as it's almost impossible to recalibrate one correctly if the other has also been changed, is that truly the case or are the certain changes we can work with?

I am curious if I am misunderstanding this as I feel like it's fairly common for both to be altered whenever a car is upgraded from the factory turbo, now the Airbox / Intake housing the MAF sensor itself may not be altered but I've seen a lot of conflicting information online about how any intake changes whatsoever can impact the accuracy of the calibration - Turbo Inlet and Turbo are specifically what has been changed in my case - along-with adding a Radium FPR, Rails, ID1050x Injectors and fuel lines. I just want to make sure I'm not over-complicating this part of the process and really get what exactly I should be doing in this kind of scenario to ensure the MAF calibration table is properly setup.

My second mental block is related in that it has to do with variances between what my X-Series Inline Bosch Wideband 02 Sensor is reading for an AFR and what my fresh OEM Front 02 is reading for an AFR, in specific the Wideband seems to read generally leaner than the Front 02 at times but also richer as well - but still within a range that both sensors should pick up. Curiously, my WRX has been showing some pretty crazy Fuel Economy averages that borderline do not make sense for what I was originally seeing on the Pro Tune - in spite of allegedly running near-stoich at cruise (somewhat lean at idle per-the Wideband) - like 44/36 MPG vs an average of 27/24 MPG with conservative driving.

This has me worried that I may need to go back and check for a mechanical issue IE some kind of fuel leak or maybe I need to re-adjust the Fuel Pressure Regulator (Radium FP Gauge reads just-under the 44 psi mark when disconnected from Manifold Pressure & Capped - 43.5 psi when disconnected is the recommended Fuel Pressure I believe..) or if there are any additional changes I need to make to the tune in order to account for the FPR setup, I've seen a couple posts here and there where people also mention MAF recalibration is required as well as Engine Load Compensation. I guess my question is - outside of visually checking for mechanical issues such as a vacuum leak / boost leak or possible fuel leak, is there anything we can look for when datalogging that may help us identify whether a mechanical issue is present that may make tuning accurately impossible? Also, I know we can take our target AFR and divide by our measured afr to get our correction amount - we use the Wideband 02 log in this case and not the factory 02 correct?

Last question... 3-port EBCS vs Subaru's factory 2-port BCS, I know it operates in interrupt mode but most of the Cobb-specific information related to boost control tuning references the factory 2-port. I'm trying to set a good starting point setup for the necessary boos tables that accounts for the 3-port - especially with the bigger turbo making a safe starting point crucial - but I'm getting myself confused trying to understand what those tables should look like with the change to a 3-port and a fairly larger turbo - at least in terms of a percentage. Does anyone by-chance have a recommendation for a starting point that accounts for both so I can start working on recalibrating, then fuel, then ignition and then circle-back to boost control?

Here's the full list of changes made over the summer, note the necessary changes have been made for the TGV & Airpump deletes, as well as injector characterization and logging for the Wideband 02 - I also uploaded a recent Datalog file from a 10+ minute log session on the highway where I maintained cruise to log AFR readings across the Wideband and the Factory 02 to try and recalibrate.

IAG v2 AOS Street Version

Tomei Turbo Inlet Silicone

Blouch Dom 1.5XTR 8cm2

ID1050x Injectors

IAG TGV Deletes

IAG Fuel Rails

Radium Multiport Fuel Pressure Regulator w/ Fuel Lines

Grimmspeed 3-Port Electronic Boost Control Solenoid

Grimmspeed Flex Up-Pipe

AFE Headers

Secondary Air Pump Delete

AEM X-series Inline w/ Bosch Wideband - (Wired to TGV R for datalogging - Sensor installed in DP from Grimmspeed - has 2nd port)

Hey James, sounds like you've got quite a project on the go. Let's see if I can help out:

1. While you can (and many tuners do) change intake and injectors simultaneously, unless you have previous quality calibration data for the new injectors or intake setup that's been validated then it's not possible to 'properly' tune either. This is because there's simply no way to know if an error in AFR vs target is the result of injector or MAF calibration. Most professional tuning shops that specialise in Subaru will already have well developed calibrations for their injectors and intakes of choice hence they can get away with doing both upgrades together.

2. While there are plenty of people who believe that as long as the MAF sensor/MAF tube remain the same, then the calibration should also remain the same, this is sadly not the case. When the MAF is calibrated this is normally inclusive of the entire inlet system and any changes to a part of that will impact on the accuracy. Admittedly some will be minor but they will affect it none the less.

3. The factory wideband sensor on these cars is useful to a point however they're still predominantly used by the ECU at stoich and the calibration tends to be further off the richer you go. I will generally trust my own wideband sensor, particularly at WOT. 'Usually' both should be very close under closed loop stoich conditions though. Personally I've always used and trusted a MoTeC PLM or LTC but found the AEM X-Series reads almost identical to the MoTeC anyway. Some of your error as far as logged values go could be the result of ground offsets if you've wired it to an analogue voltage input though. That's why I prefer CAN-based widebands as the integrity of the data is guaranteed.

4. Changing to a 3 port solenoid will generally result in more boost for the same duty cycle. I'd start by dropping the duty cycle by 20-30% which should be relatively conservative and then carefully testing from there.

We usually reply within 12hrs (often sooner)

Need Help?

Need help choosing a course?

Experiencing website difficulties?

Or need to contact us for any other reason?