When tuning a MAF based engine management system, the first priority is to calibrate the MAF sensor. Since all fuel calculations and load targeting are based on MAF sensor readings, it is very important that the MAF sensor is properly calibrated. Here is a detailed how-to:
Mass Air Flow Sensor (MAF) Calibration Procedure Using VersaTuner for Mazda 2.3l DISI Turbo Engines
Step 1 - Health Check
Before attempting to calibrate the MAF sensor, it is very important to check for and eliminate any boost or vacuum leaks. Any air that enters the engine via a leak after the MAF sensor is not measured by the MAF sensor. As a result, these leaks will cause the MAF readings to be skewed in parts of the calibration curve. If you do not have a boost leak tester to pressurize the intake system, a good visual inspection, a wiggle test to find loose clamps, and a thorough listen for whistling leaks is a good start.
Boost leaks can exist anywhere from the turbo compressor outlet to the cylinder head. To check for boost leaks, first open the oil filler cap to prevent excessive crankcase pressure from damaging the oil seals, then connect a boost leak tester to the turbo inlet. Spray soapy water on all of the boost hoses and connections as well as the intercooler. Pressurize the intake system and look for any bubbles or foam appearing around any of the connections or hoses. Correct any leaking fittings or hoses.
Vacuum leaks can exist between the throttle plate and the cylinder head. Test for vacuum leaks the same way you do for boost leaks. If you don’t have a boost leak tester, listen carefully for a whistling sound while the engine is idling. An old-school technique to test a suspect connection for a vacuum leak is to idle the engine and spray carburetor cleaner around the area in question. If there is a leak, there will be a discernable change in the idle.
Air intake leaks can exist between the MAF housing and the turbo inlet. This section of the intake system is not under boost or significant vacuum. As a result, it is harder to detect leaks. Leaks in this area usually come from loose clamps or the PCV hose connection. Inspect the clamps and hoses for loose or poor connections.
Once you are certain that there are no boost leaks, proceed to the next step – Cleaning the MAF sensor.
Step 2 - Cleaning the MAF Sensor
The MAF sensor sits in the intake air stream between the air filter and the turbo inlet. Unfortunately, air filters do not remove 100% of all contaminants. Over time, the MAF sensor can become dirty, oily or contaminated with road salt. This is more prevalent when driving on dirt roads or in climates where road salt is used to melt snow and ice on roads. Low mounted cold air intakes tend to pick up a lot more road dirt and salt.
A pictorial how-to on MAF sensor cleaning is located on the VersaTune user forum.
To clean the sensor, unplug the electrical connector from the sensor. Inspect the connector pins for signs of corrosion or any bent or damaged pins. Next, remove the sensor retaining screws. Remove the sensor from the MAF housing. Spray the MAF sensor element with an appropriate cleaning solvent. CRC makes a MAF cleaner that works well to remove contaminants from the sensor without leaving a residue.
Once the sensor is clean and all of the solvent has evaporated, re-install the sensor and reconnect the electrical connector.
Step 3 - Clear the Fuel Trims
If you have made any changes that might affect the MAF calibration, such as fixing leaks or cleaning a dirty MAF sensor, the fuel trims should be reset so they can be re-learned. Disconnect the battery or reflash the ECU to clear the existing long term fuel trims.
Step 4 - Drive the Car
In order to re-learn the fuel trims, drive the car through varying loads from idle to moderate acceleration. Monitor the long term fuel trims. When the long term fuel trims no longer change after driving, you have successfully learned the fuel trims for the current operating environment.
With the car at normal operating temperature and in normal weather conditions, find a safe location to operate the car to log the fuel trims. Log using the MAF Calibration preset in VersaTuner’s virtual dashboard. It is important to log Desired and Actual Equivalence ratio in lambda. This makes open loop calibration much easier. Operate the car at progressively higher loads starting from idle to moderate acceleration. Hold the accelerator pedal steady for two to three seconds at a particular MAF voltage. Then accelerate slightly and repeat at the next MAF voltage. The resulting MAF voltage log should look like a stair step pattern. By holding the accelerator pedal steady, you allow the logs to capture data without acceleration enrichment or deceleration fuel cut. When LTFT and STFT are both steady zero, you have finished logging the closed loop part of the MAF table, and are in open loop operating mode. Find a safe location and collect a couple WOT logs to obtain open loop data at high MAF voltage. Repeat this logging process a few times to ensure that you have enough data points to perform a thorough calibration.
Step 5 – Analyze the Logs
Open the MAF calibration logs in the VersaTuner log viewer. Use the VersaTune MAF Calibration Worksheet attached here to help organize the data and calculate the multipliers to apply to the MAF tables.
Note: At this time, the MAF Calibration worksheet only works in Excel 2007 or newer. I am working on compatibility with previous versions.
Since it is a rather tedious task to record data every .039 MAF volts corresponding to each cell in the MAF Calibration tables, try to record data at least every 3 or 4 cells if possible. Try to get data points from multiple logs to get an average. This will prevent a one-time event from creating a high or low point in your new calibration.
Closed Loop Calibration
At each point in the log where MAF volts and throttle position are steady for a second or more and Actual Equivalence Ratio is the same or very close to the Desired Equivalence Ratio (+/- .010 lambda), add the STFT and LTFT to get the combined fuel trim. Record the combined fuel trim on the calibration spreadsheet next to the corresponding MAF voltage. Do this at as many data points as possible throughout the MAF voltage range where LTFT and STFT are non-zero. The worksheet will calculate the multiplier to apply the cells of the MAF Calibration tables in Tune Editor.
Open Loop Calibration
By definition, fuel trims are not applied in open loop operation. Enter into the open loop section of the worksheet the Desired and Actual Equivalence Ratios from your logs where fuel trims are steady zero. The worksheet will calculate the multiplier to apply to the cells of the MAF Calibration tables in Tune Editor. Manually interpolate and fill in the multipliers for any MAF voltage data point that you do not have data for.
If all LTFT’s are +/- 5% or less and open loop Actual Equivalence Ratios match Desired Equivalence Ratios, the MAF calibration is acceptable. With a little work and patience, you can get LTFT’s to zero, but due to seasonal temperature changes and changes in fuel formulation it is difficult to achieve LTFT’s lower than +/- 3% for a sustained period.
If your LTFT’s are greater than +/- 5% or you are a perfectionist, proceed to the next step.
Step 6 – Modify the MAF table in VersaTuner
In VersaTuner’s Tune Editor, open the tune that was installed when the logs were recorded. Open MAF Calibration 1 table. For each MAF voltage, multiply the g/s value in the cell by corresponding multiplier from the worksheet.
Once you have applied the multipliers to all cells in the MAF Calibration 1 table, select all cells in the MAF Calibration 1 table and copy them to the clipboard. Highlight the first cell of the MAF Calibration 2 table and paste the modified cells.
Update the Tune Properties information such as version and description and save the modified tune.
Step 7 – Flash the Modified MAF Table
Using the Tunes tab in VersaTuner, flash the modified tune to the vehicle.
Repeat from Step 4 on until trims are within acceptable limits and open loop Desired and Actual Equivalence Ratios match.
Spreadsheet updated February 10, 2012.
MAF Calibration 101
- Steve @ VersaTune
- Lead tuner
- Posts: 1159
- Joined: March 29th, 2010, 12:58 pm
MAF Calibration 101
- Attachments
-
- VersaTune MAF Calibration Worksheet v1.01.zip
- (26.75 KiB) Downloaded 4899 times
Re: MAF Calibration 101
Thanks this was useful information.
After entering a series of MAF calibration adjustments the visualisation has a slight zig-zag appearance. Page 23 of the tuning guide advises to "apply smoothing as needed". Is there an easy way to smooth the curve?
After entering a series of MAF calibration adjustments the visualisation has a slight zig-zag appearance. Page 23 of the tuning guide advises to "apply smoothing as needed". Is there an easy way to smooth the curve?
- Steve @ VersaTune
- Lead tuner
- Posts: 1159
- Joined: March 29th, 2010, 12:58 pm
Re: MAF Calibration 101
You can either do it in Excel with a 3rd party smoothing tool, or do it visually in Tune Editor. We are working on adding a smoothing function to Tune Editor.
-
- Posts: 383
- Joined: August 3rd, 2011, 11:29 am
Re: MAF Calibration 101
Found 2 boost leaks on my car, was coming from a hose and from where the cold pipe meets the tmic. but even after fixing them both, I couldn't get more than 10psi in the system with my shitty compressor. I can hear air escaping from somewhere near the cold pipe area, I might try getting another gasket or could I ghetto it with some silicone?
- Steve @ VersaTune
- Lead tuner
- Posts: 1159
- Joined: March 29th, 2010, 12:58 pm
Re: MAF Calibration 101
It's not ideal, but it can get you by until you get the proper gasket in place.
The silver/grey Permatex ultra rtv is the strongest. Let it cure fully and vent the intake before you start it.
I doubt there's a risk, but RTV fumes can kill an O2 sensor. Never use RTV anywhere near an O2 sensor.
The silver/grey Permatex ultra rtv is the strongest. Let it cure fully and vent the intake before you start it.
I doubt there's a risk, but RTV fumes can kill an O2 sensor. Never use RTV anywhere near an O2 sensor.
Re: MAF Calibration 101
I have been reading this all over and over trying to understand it on how to calibrate it. Its a really tedious task.
-
- Posts: 358
- Joined: March 29th, 2010, 11:16 pm
Re: MAF Calibration 101
http://www.mazdaspeedforums.org/forum/f ... ple-60321/
read this if anyone has some confusion on how to do a maf cal. Sometimes its easier with a different person explaining it.
read this if anyone has some confusion on how to do a maf cal. Sometimes its easier with a different person explaining it.
Re: MAF Calibration 101
Hello All...
Any idea on how long (or far) it takes for the ECU to settle and establish LTFT..???
Thanks..!!
Any idea on how long (or far) it takes for the ECU to settle and establish LTFT..???
Thanks..!!
Tuned with by
Now not a Beta Tester. Proud Owner of a Personal License.
I use ECU Tools for iPhone to help with Boost & AFR conversions..!!
Mods include: ETS TMIC, K&N Filter and Turbosmart BOV.
Now not a Beta Tester. Proud Owner of a Personal License.
I use ECU Tools for iPhone to help with Boost & AFR conversions..!!
Mods include: ETS TMIC, K&N Filter and Turbosmart BOV.
-
- Posts: 383
- Joined: August 3rd, 2011, 11:29 am
Re: MAF Calibration 101
Depends how good the calibration is, I try to give it at least 20 miles.
- Steve @ VersaTune
- Lead tuner
- Posts: 1159
- Joined: March 29th, 2010, 12:58 pm
Re: MAF Calibration 101
it depends on how you drive. If you cover all of the possible operating conditions for some period of time, it will populate the LTFTs. I recommend several drive cycles rather than one long drive. Include some idle time too.