I have tuned several high performance engines with MAP sensor based EFI. It ends up working very well and can make huge power quite well. The biggest problems I ran into was at light loads, especially on a Honda with big cams. Idle was near impossible to tune nice with just the MAP sensor for a signal. The motor only pulled 12 inches of vac at 900 rpm, but it shot to 18 inches of vac by 1500 rpm. This suggests it needs less fuel, but it really needs more. For a test, I added a small ford hot wire MAF sensor and looked at its output. I used it to help me tune the V.E. table values for the MAP sensor. Had I just tuned it for the MAF, it would have been simple, but the sensor was way too small as it was from a 3.0L Taurus that topped out at just 160 hp and this little 1.6L made close to 300 hp at just 12 psi of boost. Once I got in the ball park using the flow numbers from the MAF sensor, I completed the tune with a wide band O2 sensor system. After seeing how nice the data from a MAF worked, I actually picked up a KV AFM from a 7MGTE to try and use on my turboed 22RE 2.4L Never finished that idea, but it looked very promissing.
Most MAF's and AFM's only measure part of the air flowing through them. The mesh (or screens) on the sensors ensure that the ratio of flow being measured to the bypassed flow holds constant as the total flow changes. Most Ford units have the same electronics, but the housings have bigger bypassing passages on the higher power motors. My 3.0 Taurus meter was identical to the 5.0L Mustang meter except for the size of the un measured hole. I believe Toyota did the same thing with the 7MGTE and the 1UZFE units. The passages next to the meter bar are bigger on the 1UZ. The ECU is programmed with the ratio of signal to CFM of air. Change the amount of bypass air and the mixture will be wrong. The same signal off of a 1UZFE meter means more air than that from the 7MGTE meter. Using the bigger injectors fakes out the ECU to provide the proper mixture again. I used a strong shop vac to compare the signals of several AFM's. If you do anything to get more air through the AFM, you need to make sure it provides the same signal frequency for a given amount of air flow, or you adjust the injector size for the proper air flow. Adjusting the injectors (and/or fuel pressure) can get the mixture back over a fair range, but also keep in mind the spark timing will also change with this type of mod. Using a bigger air meter that gives a smaller reading (and using bigger injectors to compensate) may result in perfect mixture, but the ECU "thinks" the engine is under lighter load so it runs more ignition advance. This could result in detonation.
Personally, I am leaving the AFM alone until I am certain I exceed it's limit. I will monitor the vacuum drawn between the throttle and the AFM (and between the filter and AFM). If it is a small amount of vac, the AFM is not a problem. A guy who put a 90 mm throttle on a Honda was annoyed when he didn't get any improvement in acceleration. It turned out he was not pulling any vacuum behind the stock throttle, so going with a bigger hole made no change. All it did do was make 1.2 throttle pull as hard as wide open before, and 1/2 to WOT made no change at all. It just made it near impossible to drive smoothly as it would rev hard from a tiny touch on the pedal.
Once you do go to boost though, the MAP sensor is king for several reasons. A HUGE AFM that cam meter that much flow will not have near as good of resolution as the MAP sensor. Blow off valves on MAF (AFM) systems need to be connected back into the system after the AFM or the mixture goes out the window when it opens. Any leak after a flow meter will mess with mixture, MAP sensors don't car how the air gets into the manifold, it fuels the same whether it is a leak, the brake booster, or even the A/C control head idle up VSV. The bad part about MAP is that any change in the flow rate of the engine requires a re tune. By chnaging the cam in my 2.4 turbo I had to decrease the V.E. table values from idle to 2500 rpm by almost 10%, then increase them nearly 20% from 4000 rpm on up 5500. At 6500 or so it was about back even as the head could no longer flow up there. The ported head required the 5000 to 7000 rpm ranmge to be increased another 10%. All this for the same MAP sensor readings. I actually had to drop max boost from 15 to 13 psi because the injectors could no longer keep up. A MAF system should just see the new found flow and inject the extra fuel (within it's flow range)
Gary M.
