diptenkrom

i have been reading about the ability to swap an AFM for a MAF without an ecu upgrade. you put a device inline to translate the signal.

unfortunatly it looks as if you have to make this device yourself, and i do not know where to begin. but i saw one of these on an s4, i have an s5 and it says they are different, but does not specify how.

here is the link ->
http://howto.globalvicinity.com/gv_w...=120&co=1&vi=1


i would like to know if anyone has done this, and what i need (other than the maf , temp sensor, and wiring)? this guy claims that the "translator" can be made for $10.

i want to do this as well as a CAI because this system seems primitive, restrictive and not as precise. just trying to get a bit more response and power on the cheap.

arghx

iTrader: (3)

The whole purpose of a MAF translator is for the new MAF signal to NOT equal the old one from the AFM! It is to reduce the chance of maxing out the AFM on turbo applications! What this person is doing is a nice little experiment, but it has zero practical use.

On the n/a, the AFM is NOT RESTRICTIVE. I REPEAT: IT DOES NOT AFFECT POWER OUTPUT IN A MEANINGFUL WAY. This is according to my own testing. I can show you two 172rwhp dyno sheets, one with an AFM (stock ECU + SAFC) and one on a standalone which in fact had a better tune. Of course there is a margin of error and you may in some circumstances pick up an HP or two, but take it form someone who has more n/a power than most people here.

On turbo applications (especially Mitsu), the purpose of the MAF translator (MAF-T pro and MAF-T Gen 2 are most common) is to trick the ECU into seeing less airflow so the signal does not max out. That leads to unpredictable spark advance curves unless you resort to even more piggyback trickery. On an Rx-7, it is further complicated by staged injection and 2 sparkplugs per rotor.

What the author of that writeup did was the opposite, which was try to get a MAF sensor output signal to mirror a stock flapper AFM signal. This is pointless on an n/a because the AFM is not restrictive, and it is pointless on a turbo car because it will not prevent you from maxing out the AFM. And without ECU datalogging through an Rtek we cannot confirm that the ECU is in fact seeing the same load with both types of airflow sensor. He measured the signals, but he did not confirm by logging gram/sec of calculated air mass from the ECU.

long story short, just don't bother with it. it was an interesting experiment, but that's about it.

vato2000taco

iTrader: (1)

wait an s5 has a maf already stock dont it? s4 has vane type afm and will swap out directly with s5 afm but u loose the use of the stock cold crank fuel map in oem ecu

Evil Aviator

The weak link in the system is the 1980's vintage computer, not the AFM. Hooking a new MAF sensor to your stock EMS and thinking you have something slick would be like hooking a new flat panel monitor to an old IBM 286 computer and thinking that you are God's gift to online gaming.

If you want to build something yourself that is somewhat cheap, more precise than the stock system, completely free of restrictions, and totally immune to reversion waves, then build a Megasquirt EMS and run it in Alpha-N mode.

It would be nice if you could buy a $300 MAF sensor and $20 worth of electronics to replace a failed AFM rather than pay $1,000 for a new AFM. Also, the AFM will be NLA at some point, so it would be nice to have an alternative to spending a lot of time and money dealing with an aftermarket standalone EMS. Other than that, I see no practical use, either.

High-output turbocharged engines are usually better off with a speed-density system, and most serious competition NA engines use an Alpha-N system.

The whole "maxing out the AFM" concept is a scare tactic used by slick marketers to get you to buy their product. It is similar to how they try to tell you that you need their turbo timer to keep your turbo from seizing, or that your engine will blow up if you don't tune it with their wideband gauge, or that your engine will not have a long life unless you use their oil additive or high octane fuel. Complete crap.

No, the S5 also has an AFM, but its a slider type rather than a vane type.

arghx

iTrader: (3)

http://www.stealth316.com/2-mas_liter-per-hz.htm

This depends on the application. For example, in many Mitsu applications, if the Karmann frequency from the airflow meter is greater than about 2200 Hz (which corresponds to about 36 lb/min and mid 300's whp for that car) you will hit fuel cut. There is no MAP sensor or boost-based fuel cut in all the Mitsu EFI systems I have seen. If you flow more air than that you have to do all sorts of crap with bigger injectors and a bunch of piggybacks and dataloggers to get it running right. I've been through that on a Vr-4.

Another point of mine is that there are a number of cars making a decent amount of power on the stock s4 AFM: https://www.rx7club.com/time-slips-dyno-128/dynod-s-afc-stock-ecu-eng-340rwhp-306lb-ft-655194/ . of course he is using an SAFC in that case which affects the AFM signal. In terms of physical restrictions, the AFM is faaaaaaaaaar less important than the exhaust and manifold design as well as the exhaust size of the turbo.

I repeat: exhaust restriction is far more limiting on a rotary than intake restriction. I challenge anyone to prove me otherwise.

the s5 AFM, while somewhat different from the s4, is not a hotwire type nor a Karmann Vortex type. it still has a physical measuring plate that measures the volume of air. From that volume is calculated the mass flow, which is a component in the load reference for the fuel and spark lookup tables in the stock ECU. these parameters can all be logged in the Rtek.