OK, I was gooffing off on the web and I found something that caught my attn. Someone(?!) Apparentley upgraded their AFM to a hot wire MAF, using all the stock computers in his car. I kind of understood what he did, but I'm pretty stupid when it comes to anything computer, so I don't know what some stuff is, and what stuff means(well alot of stuff). Still, I think it's an interesting write up. How about if some of our forum "techs" take a look at it. It's at www.carlton24v.co.uk/afmmaf.htm

Let us know what you all think. Is this doable for our cars? can the "average joe" do this? Can someone explain this in laymans term?

 

i saw that about 2 days ago, was thinking the same thing... the average joe could do that.... if he could program the pic microcontroler, and had dataloggers and all sorts of other good stuff like that

anyway, i know some of the people on this board have to be electrical engineers... what do you say guys... how about trying to make the module to let us replace the afm with a maf sensor.... something common like the one from mustangs or volvo's or something, im sure we would all pay well for a lil box that would remove the greatest restriction in the intake

 

a team of engineers got together to bring us the datalogit, I'd really like to see this happen. AFM is a big bottleneck and huge nuisance when i comes to providing more air for our cars!

Goals:
eliminate AFM
work with stock ecu
provide more air/better response
low cost


/me prays

 

^^^^^^^^^^^^^^^^^^^^^^^YES!!!

 

I remeber a while back mazdaspeed7 was working on a mustang maf conversion. Last I remember the car was running, but like crap. Don't know what ever happend with that. Someone out there's got to have the stuff and knowhow to do this project with. Let us who weren't blessed with the digital intellegence to pull off such a project know And hell yeah, I'de pay a little somth'n-somth'n for a module to do this. I'm going to bed(work 3rd shift), I'll check back later. Come on guys(and gals)!!

 

Adam never got his conversion to work. If it was running like crap, it was unrelated.

 

LOL!

 

that conversion would never work on an s4 computer even with all their extra cards.

 

When I get this beast running I am going to experiment with running two airflow meters (stock S4) in parallel, leading to an S-AFC and finally to the ECU. Then I will get the pair tuned on the dyno with a wideband and see if having two restrictive orifices is better than one. Each airflow meter will have a 2.25"-ish outlet pipe collecting into a 3" throttlebody feeder pipe . . . with two cone filters in the cold air box of course. Every electrical engineer knows that two resistors flow more current than one, hence my reasoning that two restrictive airflow meters flows would act the same way. Bue I am still months away from doing this. One problem I need to solve is how to put two cone filters in the cramped engine bay corner . . . maybe one on each corner but with the airflow meters in the same general location as stock, with two ram air boxes fed from my bumper ducts . . .

 

the R&D costs and time would cost more then a haltech... therefor I doubt anyone would ever build one.

 

the e6k isn't all that expensive...hmmm

before any of that, people who would know how to actually make it happen would need to be found

 

How much of the sensor range is there to spare? And would it be conceiveable to use the internals of the stock AFM and put them in a less restrictive housing, perhaps something custom out of fiberglass?

 

I was thinking about trying to make a custom AFM for my car. The internals would be moved into a "more round" housing. A problem would be that flap thing(trap door doohickey). It would still need one of thoes door things to measure the airflow. Probably just cut something out of 1-2mm sheet metal of the same size of the housing. But that'de still have the resistance from the flap. Don't get me wrong, I'de take that over the stock afm any day(if it works), But I really would like to see if SOMEBODY could figure this out so the MAF conversion can be done. Come on guys, we have to show people what our cars can really do. I know there's alot of people smarter than me on this board. I'm just not smart enough to figure this conversion out(nor do I have the equipment or know how, don't even know what half of the technical jumbo is on that site). Can anyone come up with something??? It's not really impossible, is it???!

 

I don't understand why. Any way you could elaborate for me some(without wasting a whole lot of your time )? I really need to learn about computers!

 

any conversion will work... so long as there is a peice of hardware doing the conversion from the hotwire maf sensor response curve over to the afm's response curve... in that case for x amount of air coming in, the ecu will see the same response from the maf as it would from the afm

it has nothing to do with the computer, it just means making a lil black box that will adjust the signal coming out of the maf sensor

what has to be done is to place both sensors inline in the intake stream, with the car still hooked up to the afm, run the car from idle to redline, and datalog the output of both sensors, do this a few times to get a good average

then take the two sets of data, compute the curves for each, and find a conversion between the two (may be some sort of polynomial function, or it may have to be a table lookup, if no good function can be found)

then that function or table lookup needs to be either implemented with some sort of dsp/microcontroler... or if its simple enough, some analog circuit

then the afm can be removed, the maf put in its place, the little black box installed between the maf and the ecu, and then we can enjoy reduced air flow restriction and better throttle response

all and all its not that hard, it just requires the datalogger and then a lil time to make the signal converter

 

Basically talkin' out my *** here, but thought I'd throw in my 2 cents...

I can't see two AFMs or a modified AFM working at all. The thing is (obviously) measuring how much air is moving through, which is converted to an electrical signal and sent to the ECU.

Two AFMs in parallel will split the intake air in two, so each will be receiving around half the total (depending on turbulence and overall resistance). It would take some extreme fiddlin' to get any voltage that the ECU could use.

Making a "bigger flap" is all fine and dandy, but the amount of air moving in will also be bigger, presumably in some non-linear fashion, and the existing set up can't allow for that.

I don't know, but it would seem to me that -- given the fact the maximum voltage put out by the AFM would coincide with the maximum amount of air it can flow -- the stock ECU maps would be incapable of covering larger amounts of air. I suppose you could compensate, to some degree, with an S-AFC and some tuning.

Anyway, like I said, my two cents...

 

agreed, modifying the current afm is for the most part out of the question, because any change to it will change the way it responds and will have a negative effect (the ecu wont be getting the data it wants), but like i outlined in my last post, replacing the afm with a maf sensor would work fine, so long as the conversion was being made

the two sensors side by side thing doesnt make sense to me at least, in order to figure out the difference between them they have to be inline (aka, have the exact same amount of air moving thru each of them)

the issue isnt flowing more air... as the engine is only going to draw a constant amount of air for a given rpm (assuming not turboed, which can change that a lil, based on diff levels of boost), the issue is that any restriction in the airflow robs horsepower, as the engine has to pull harder to get the air in, a more freely flowing maf sensor would give back that lost power (prolly not "that" much, but it would help, and it would definately increase throttle response)

 

Ahh...Now I finally understand. The engine will only suck in "so much" air at any certain rpm. The idea with no restriction on the intake is to make it easier, in otherwords take a "load" off the engine. Thank you. So does anyone have a datalogger?

As soon as I get **** straightend out in my life, even with as stupid as I can be at times, I'm going to try to do this(if no one else has beaten me to it yet)

Oh, did anyone else try to e-mail that dude that posted the sight? I tried twice but never got an answere. Wonder what's up with that?

 

My advice would be to put your efforts into something more worthwhile. There's a very good reason why this is rarely attempted and never actually seen working. It's simply not worth the effort. Even if you could build a "black box" to convert a hot-wire AFM's output into something the ECU could make sense of, the actual gains are going to be minimal.

You have to remember that even hot-wire AFM's present a significant restriction to airflow. Yes they are much better that a vane-type AFM, but you're still only talking about one of the many restrictions the engine has to overcome. Assuming you can get mixtures back to factory settings, I'd be impressed if you got close to a 10hp peak gain. And at the end of the day, you're still stuck with a non-programmable computer that is many generations out of date.

Hot-wire AFM's are now over 15 years old, so if this idea was feasible, it would probably have been done a long time ago. The fact that it is only ever talked about is a big hint that even if the electronic hurdles could be overcome, it wouldn't necessarily be worth the hassle. If your car is modified enough for the AFM to have come an obstacle, you should be considering aftermarket programmable EFI anyway.

 

about a year ago there were two sites that had a how to on this for our cars. I had them bookmarked then my computer crashed an I have not been able to find them sence. there was also some one on this board that had something on this. I think his name was ted or tim website was pro somethingor maybe fcpro ?? he had a right up on flaper style to slider style and some thing about changing to hot wire style. there was also a chart on one site showing flow differances on all the styles it was a prety big differance between the door type and the hot wire type. probly a worth while differance on turbo engines.

 

Gee, I wonder who he's talking about?

 

All of your programming and restriction troubles solved, for about $300: http://www.splitsec.com/products/psc1/PSC1005.htm

 

i dont know which style the s5 afm is but it is easy to install one on an s4

 

Out of curiosity what would you guys pay for something like this? Exclude the cost of the MAF.

Henrik
87TII

 

Ok guys, I messed with a Mustang MAF conversion a year ago. You have to invert the signal and adjust it's width for it to work. Easier to do is get a MegaSquirt and eliminate the AFM altogether and go to a MAP sensor. Easier to tune also.