Announcement

Collapse
No announcement yet.

MAF ---> Speed Density

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • MAF ---> Speed Density

    As my signature says I have an 86 T/A. It runs with a MAF on there right now, however I know that they made models (I think in later years) that were equipped with a speed density system instead. My questions are as follows.

    1. What are the differences between the two?
    2. Are there any advantages to switching to a Speed Density system?
    3. How would I go about the conversion (i.e. what parts are needed, difficulty, proceedure, etc.)?

    Thanks in advance as always.
    -Dan

    1986 350 TPI Trans Am, 700R-4, Flowmaster exhaust, CAI, 180* t/stat, low temp fan switch, SFC's....Sold.

    Project Status: LT1 disassembled, researching costs, searching for project car.


  • #2
    The fuel injection system needs to know how much air is entering the engine, so that it can calculate the required fuel.

    MAF measures the mass air flow directly, by heating a small wire, and measuring the current required to keep it a given number of degrees above the temperature of the incoming air. By knowing the power loss due to heat transfer, and the specific heat of air, the MAF sensor can calculate the mass air flow. The sensor feeds the ECM a voltage signal (3rd Gen Bosch MAF sensor) or frequency signal (4th Gen GM MAF sensor) that it translate into a mass air flow, typically expressed in grams per second.

    Spped-density is just another way of estimating the mass air flow, by measuring the temperature and pressure of the incoming air to calculate its density, and then multiplying the density by the RPM (this is the speed component) X volumetric efficiency (VE) of the engine. The "density" part is easy, but the "speed" part is a problem. The manufacturer has to put the engine on an engine dyno, and measure the air flow, so that they can develop complex tables of VE, as a function of RPM and manifold absolut pressure. This table is programmed into the ECM, and the ECM has to look up the programmed VE for each calculation.

    If you make a change to the engine that improves VE (improved inlet system, or improved exhaust system, a cam, etc.) the table programmed in the ECM are no longer valid. A "tune" is required to adjust the VE tables to reflect the improved breathing of the engine. The MAF system does not have this problem, since it is directly measuring the air flow into the engine.

    In effect, the MAF system should adopt more easilly to moderate changes to the engine, while a speed-density setup might need "reprogramming" a bit sooner.

    But its not a perfect world, and the Bosch MAF sensor used in the 3rd Gen engines was a real piece of crap. It got dirty easilly, required a "burn off" system that heats the "hot wire" to 1,000degF on shutdown to burn off contaminants, had sensitive electronics that failed frequently, a restrictive screen on both the inlet and outlet side, along with some unnecessary cooling fins in the flow area, both of which severely hampered air flow for modified engines. Fortunately, the GM-developed sensor used in the 4th Gens has elimiinted all of those problems.

    The two systems will provide the same level of perfomance, as long as the VE tables are good. I run an aftermarket ECU (MoTeC M48Pro) that is programmed in speed-density, open-loop, with no knock sensors. It provides a very efficient tune (BSFC less than 0.44 #/HR/HP), and has proven to be extremely reliable. But the "tune" is backed up with more than 40 engine dyno and chassis dyno pulls, and when anything changes, its back on the dyno to verify the tune...... not really practical for a daily driver with frequent mods, but OK for a single purpose (track only) application.
    Fred

    381ci all-forged stroker - 10.8:1 - CNC LT4 heads/intake - CC solid roller - MoTeC engine management - 8 LS1 coils - 58mm TB - 78# injectors - 300-shot dry nitrous - TH400 - Gear Vendor O/D - Strange 12-bolt - 4.11's - AS&M headers - duals - Corbeau seat - AutoMeter gauges - roll bar - Spohn suspension - QA1 shocks - a few other odds 'n ends. 800HP/800lb-ft at the flywheel, on a 300-shot. 11.5 @ 117MPH straight motor

    Comment


    • #3
      Thanks for the info I think I understand how that stuff works now, so if possible, would I be able to put a 4th gen MAF on there since it flows better/ is better quality, or would that require changes to the ECM as well? It sounds like it would help at least a little bit in how the car operates with a 4th gen one on there, but I wonder about compatibility.
      -Dan

      1986 350 TPI Trans Am, 700R-4, Flowmaster exhaust, CAI, 180* t/stat, low temp fan switch, SFC's....Sold.

      Project Status: LT1 disassembled, researching costs, searching for project car.

      Comment


      • #4
        Definitely need a new ecm for a 4thgen maf. The ecm's in the maf'ed 3rdgens are sloooooow (ecm speedwise and data transferwise). Although I haven't compared them, the maf hookups are most probably different. The 90+thirdgens used speed density.
        There is always a question over on thirdgen.org about which is better (SD or MAF) and it's really a tossup. Although there's probably a reason that gm went back to MAF in the 4thgens (except '93's) You could probably find more specific info on the conversion and pros/cons on thirdgen.org.
        1989 TA
        Flowmaster exhaust
        Ghetto CAI

        1994 TA GT
        K&N CAI
        Magnaflow catback

        www.nkyfba.com

        Comment


        • #5
          The computer tables are calibrated to the MAF. Making any changes would require reprogramming (probably many attempts to get it right).
          Rob B 95Z A4 Tech Page (Part numbers / locations, how to's, schematics, DTC's...) Home Page - shbox.com

          Comment


          • #6
            And as I noted... the 3rd Gen ECM is expecting a variable voltage signal from the MAF. The 4th Gen PCM is expecting a variable frequency. Can't mix them, without some sort of a signal converter.
            Fred

            381ci all-forged stroker - 10.8:1 - CNC LT4 heads/intake - CC solid roller - MoTeC engine management - 8 LS1 coils - 58mm TB - 78# injectors - 300-shot dry nitrous - TH400 - Gear Vendor O/D - Strange 12-bolt - 4.11's - AS&M headers - duals - Corbeau seat - AutoMeter gauges - roll bar - Spohn suspension - QA1 shocks - a few other odds 'n ends. 800HP/800lb-ft at the flywheel, on a 300-shot. 11.5 @ 117MPH straight motor

            Comment


            • #7
              Ok, thanks for the info, i was mainly just wondering what the difference was between the two. And then I figured if it would help the car run better on one system i might try to change it, but now it looks like I'll just keep it the way it is.
              -Dan

              1986 350 TPI Trans Am, 700R-4, Flowmaster exhaust, CAI, 180* t/stat, low temp fan switch, SFC's....Sold.

              Project Status: LT1 disassembled, researching costs, searching for project car.

              Comment

              Working...
              X