Airflow Options

With the release of ECMLink, users have the option to run any number of different mass airflow solutions. They can run a Mitsubishi MAF sensor (1G, 2G, EVO8), any one of a number of GM MAF sensors, or they can run speed density. This page outlines some of the considerations of each.

If you're not sure how airflow plays into the grand scheme of engine management, then please read the Engine and ECU operation page first.

The “standard” options include a 1G DSM, 2G DSM, or an EVO8 (3G Eclipse) mass airflow sensor. Some things to note with these follow.

• Must be installed in front of the turbo.
• Are generally easy to install, but they do add some constraints on piping layout just because of their location.
• Generally provide very smooth, accurate airflow data and work well with our ECUs.
• Are not sensitive to changes in engine VE and atmospheric conditions.
• Use both IAT and Baro inputs to the ECU (preventing you from connecting other sensors to those inputs for logging).
• Can provide some restriction to airflow to the turbo. This can be a hot topic for debate, though. We generally feel this restriction is minimal at best and for most people should not be a major consideration. Just make sure you replace the stock snorkel between the MAS outlet and turbo inlet with a larger pipe and you shouldn't have anything to be concerned with on this topic.
• Have some very real limitations on airflow metering capacity. This is particularly problematic for the 1G DSM sensor. The 2G DSM is better and the EVO8 is really very nice. Some details on the limits can be found on our MAF Data page. Generally speaking, a 2G DSM MAS with ECMLink's MAFClamp function can work well to 30psi on a 2 liter engine. An EVO8 MAS works well beyond that, particularly when combined with ECMLink's MAFClamp function.

A popular option for running a non-Mitsubishi mass airflow sensor with ECMLink is to run a GM sensor instead. Some part numbers of compatible sensors can be found on our GM MAF Part Numbers page. Some things to keep in mind with these follow.

• Can be installed before or after the turbo, which allows for more choices in piping layout. However, installation after the turbo can be problematic. More on that below.
• Do not use the IAT and Baro inputs to the ECU (thus freeing them for logging other aftermarket sensors instead).
• Are not sensitive to changes in air temperature.
• Are very sensitive to moisture. This can (and often does) lead to erratic behavior after a good rain, for example, or anything that produces condensation and/or water droplets before the sensor. This is mainly a problem when installed after the turbo.
• Are very sensitive to piping layout before/after the sensor. Running 2.25“ piping up to a 3.5” GM MAS is a problem… Installing the sensor right after a 90deg bend is a problem… Anything that changes airflow direction or cross sectional density just before the sensor is a problem.

These sensors are designed to work with laminar airflow through the entire cross sectional area of the sensor.

When you run small piping up to a large sensor, you concentrate airflow into a small area of the sensor's cross section. This violates its assumption about airflow through the sensor and throws the airflow reading off.

When you install the sensor after the turbo (typically in the upper IC just before the throttle body), you're asking it to measure very turbulent (not laminar) airflow. This also violates some of its design parameters and throws the airflow reading off.

Honestly, given the typical installation of these sensors on the typical DSM, I'm surprised they work at all. But they do work and they can be made to meter pretty well even though we're pretty much breaking every rule in the book about their installation and operation constraints. I guess that says something positive about the sensor design after all.

We do provide a dial-in procedure that helps get airflow working right and, as I mentioned, they do seem to work pretty darn well once dialed in properly. When I run my car on the GM MAS, I can't tell any real difference between it and a Mitsubishi mass airflow sensor or speed density. It works fine for me and countless others. Some people, however, do have problems. Particularly with moisture issues. Even on my own setup, when my turbo started to develop a problem that allowed some oil to get into the intake side of things, the GM MAF got very upset and I knew pretty quickly something was wrong.

Speed density operation only requires three things: a manifold pressure sensor, an air temperature sensor and some idea of the volumetric efficiency of your engine. Much more details can be found on the following pages.

Basic Speed Density Operation
Setting up Speed Density with ECMLink

Since most people will already want to log manifold pressure anyway, that sensor is typically already available. And installing an air temperature sensor isn't much work at all.

So the only “tricky” part is getting your VE table dialed in. But that isn't really hard either because we provide a couple dial-in procedures that make this process quick and painless.

Dial-in using data from a mass airflow sensor
Dial-in using fuel trim data

Some things to keep in mind follow.

• Requires that you install a manifold pressure sensor and intake air temp sensor. A list of compatible sensors can be found on our SD setup page.
• Requires that you dial in the VE table to match your setup.
• Will require changes to the VE table as you make large changes to your setup (cams, turbo, exhaust, etc.). These can be trivial or pretty extensive, depending on how much the change you're making affects your VE.
• Works very nicely once properly configured.
• Can take time to properly configure.