ford maf

Cory9584 · 02-10-2005, 07:31 PM

I have a ford maf sensor with all the wiring laying around my garage and was wondering if I can adapt it to my car and if so whats involved in the swap. I know I will need a controller and thats my next question, whats a fairly reasonably priced good controller. Whats your opinions on this idea?

d_roedel · 02-10-2005, 08:26 PM

i have been playing with the idea also,
split second sells and arc-2 controller for our purpose and
pro-m also sells a programmer for it,

you can find their websites easy on google,

and suposevely maf is the best gain for our cars,,

derek

Clowd · 02-10-2005, 08:35 PM

I was under the impression that you couldn't do a MAF setup on an N/A 944?

will944 · will944's Garage

Quote:

Originally posted by Clowd
I was under the impression that you couldn't do a MAF setup on an N/A 944?

You can do a MAF...Huntley use to sell one for the NA. In fact, fast924s has one on his car. Not much in the way of performance gains like on the turbo though.....

d_roedel · 02-11-2005, 10:31 AM

it is more of gain than a $600 exhaust, and i think more worthy of spending money on,

fast924S · fast924S's Garage

Hey WILL944, Long time no talk. Anyway back to the question, I have heard of many people thinking about running a ford setup but I havent heard of any one doing it. If FR-wilk was around he has some good info on it. Yes you can run a MAF setup, I am running one on my 87 924S just wish it sucked in air some where else but that will get work out during my rebuild. The MAF setup I have is from Huntley Racing, Its uses a fuel controler ARC-2 from split second. I love the setup, Huntley claimed 15hp and 13 ft/tq, I havent run my car on a dyno yet so Im not sure, but seat of paints dyno said it made a difference. I would ony convert to MAF if your really looking to go the performance rout, If not then just a chip and a good exhaust should be fine.

here are some pics of my MAF setup

[img]http://forums.pelicanparts.com/uploads4/
DSCN48971108148698.jpg[/img]

here is one of the controler

Also its a good idea to have a A/F ratio gauge

DannoXYZ · 02-12-2005, 02:53 PM

"I was under the impression that you couldn't do a MAF setup on an N/A 944?"

Sure you could. Just that it wouldn't give you as much of a gain as with the turbos, a +5-15hp increase as opposed to +40-70hp. Also I can map chips to directly match any MAF sensor, thus no need for the signal-interceptors/massagers. While these will skew the look-up point on the chips to adjust fuel, it will also mess with the ignition values. Most often times in the opposite direction as you want to go for increased performance.

Cory9584 · 02-12-2005, 09:26 PM

Is there really a way to just change the chips to get a maf to work being i want chips anyway. I really dont want something i have to manually adjust anyway like the 4 knob controllers. Maybe some custom chips?

Chriz75 · 04-08-2005, 05:39 AM

Im currently installing a Pro M Racing MAF to my 944 S2. Im going to use a converter of my own. Unfortunally Pro M went out of buisness before I had to buy a harness and a VAT sensor, but I will get these components from someone else.
Maybe one of you knows which pin is the 12V and Ground and Signal and Signal Ground for the Pro M ?

The converter Im going to make is with a PIC processor and one net of resistance(two components and one 5V stab.)
To calibrate it the first time Im going to run the AFM and MAF in series and then make a 256 point converison table and load it into the PIC processor.

I have some plans also to supercharge the S2 with a eaton M62 supercharger but that is a later project.

I can send pictures of the project if someone is intressed in this!

/Christian

operationcougar · operationcougar's Garage

^^^^ YES ^^^^

Zero10 · 04-09-2005, 11:36 PM

Chriz75, you stole that idea right out of my head!!
I'm working on the same thing right now, using a PIC16F87 to try to do the dirty work for me, I have some voltage-frequency / frequency-voltage converters, DAC's, all sorts of crap to make this work, just having some issues with the programmer...
Crazy that you'd think of that same idea too. Great minds think alike?

I was going for 1024 conversion points, or about 1/8 the resolution of the DAC's I'm using. Not sure if that's too fine or not.
I don't know what MAF's to use with it though, so I was just going to pick up a couple cheap ones, similarly sized to stock, so that the intake boots wouldn't have to be changed...

Mind if I ask which controller you planned to use?

For calibration, I was planning to hook up a leaf blower, or something else that moves a lot of air to the AFM, then to the MAF, just so the pressure buildup will be before the AFM, and won't affect my results in any way.

DannoXYZ · 04-10-2005, 03:54 AM

Best way to get accurate calibrations is to run the car bone-stock with the MAF installed (before or after AFM shouldn't make a difference). Then datalog both the AFM & MAFs output to arrive at a CFM vs. voltage cross-reference. No need to go crazy with resolution because the chip-maps are only 12x12 anyway.

This AFM-emulation with a MAF+PIC will only work for an NA or something that will not flow much more than stock. Because once you hit the AFM's max-opening point at say... 250CFM and 4.5v output, you're stuck on the edge of the maps.

The Apexi S-AFC-II is a darn good massager. Better than anything else on the market because it has quasi-3D mapping of LOADxRPM reference points just like an aftermarket EFI system. No other controller on the market does that and after hundreds of hours on the dyno fiddling with all the other units, you'd wish you had a SAFC from the start. The 3D mapping makes very precise adjustments possible and you save a tonne of money in dyno-time. It's even possible to dial in a car with just 2 runs, a before run, print out the dyno-chart, make the adjustment in the S-AFC, do an after run and that's sit!

Zero10 · 04-10-2005, 07:55 PM

At around how many HP do you max out the stock AFM? Should be in the 300's, no?
More than I want to make

The Apexi S-AFC-II is a very nice controller, but in real life, most of us can't afford the pro MAF kits. A PIC controlled Ford MAF or something similar to that would be just peachy for us low-end performance guys. I'm looking to make a bit more power than stock, and learn something doing it.

Am I correct in understanding that the Apexi S-AFC-II does a bit of processing itself, sending (occasionally) ficticious numbers to the DME in order to make it run better? A lot of intelligence would go into that, and that would be pretty neat. A bit above my level of knowledge, but the PIC would be more than capable of handling it, the ones I'm looking at are well over 10MIPS, so they could do many other things at the same time with no issues.

Chriz75 · 04-11-2005, 12:43 AM

I was thinking about using PIC16F88(maybe a bigger one) and run 255 calibration points. I think I will also add four potentiometers to fine calibrate the value after the calibration from the AFM.

When I do the calibration then I will remove the air filter which is located in the front on the S2 and connect the MAF and a small air filter after that. Then I will do the calibration with the AFM and MAF while driving.
When Im finshed with the calibration then I will fit the original air filter and remove the AFM and connect the MAF in that position. It is very important that all the calibration values are written!

If I fit a supercharger then I will do a reprogram of the ECU map but that is a later issue.

Chriz75 · 04-11-2005, 01:28 AM

Why use a PIC16F87 with no ADC ??
If you use a PIC16F88 or PIC16F876 then you can use the internal ADC and the EEPROM as calibration points. And then use a R/2R net and the eight digital outputs from the PIC as a DAC.

Should work I think!

DannoXYZ · 12:54 PM

"At around how many HP do you max out the stock AFM? Should be in the 300's, no?"

Well, it depends upon the spring-calibration as well. I've seen a large variation in AFMs with wildly different calibrations. In stock trim, the 944 maxes out at about 200hp and the 951 around 280hp. A big limitation is the single-row map under WOT with the Motronic. This limits you to a single boost-level under WOT. Tune for one only, all others will be compromised.

"Am I correct in understanding that the Apexi S-AFC-II does a bit of processing itself, sending (occasionally) ficticious numbers to the DME in order to make it run better? A lot of intelligence would go into that, and that would be pretty neat."

You can get an S-AFC-II for around $250. Don't know what your current wages are, but I suspect it would cost more than that in time to build a PIC massager. You also don't get 3D mapping. That is different adjustments to the same voltage at different RPMs. A 2.5v signal at low-load/low-throttle may need to be adjusted up to 2.75v. While that SAME 2.5v at high-load/+80% trottle will need to be adjusted down to 2.25v. This is were 3D mapping works and no type of volt-in -> volt-out massaging can possible accomplish this feat. This is a VERY common adjustment BTW, since the mixtures tend to be too lean at 2.5v under partial-throttle and to rich under full-throttle.

Zero10 · 04-11-2005, 03:33 PM

The reason I was using a PIC 16F87, is because I have a bunch of them sitting around, since I'm building a temperature monitoring system, with about 45 inputs, so I also have lots of external ADC's. If I was starting from scratch, I would use the 16F88.

I'm presently trying to build the el cheapo programmer, but I just roasted a 16F87 and a 2N7000 making a mistake, and reading a diagram slightly wrong. I'm changing my design to the newer layout, using the 2n3906 instead of the zvp2106a, hoping that will eliminate the random data errors I was receiving when it worked.

Interesting, I didn't know the S-AFC was only $250, for some reason I was thinking $400-$500, will it work with any MAF?

I'm considering hitting up a junk yard for a couple decently-sized MAF's, and just trying to invent something.
It doesn't have to be any better than stock, doesn't need fancy mapping, just needs to be fun to do.
I often undertake inexpensive projects like this just to see if I can do them. However, from what I have heard from you Danno, it really does sound like the S-AFC is the best solution if I want to do things right. Thanks for the info.

Just curious, but does anybody know what current the DME draws from the AFM? My DAC's will only put out 100mA, that should be more than enough, but I just want to make sure.

If anybody's interested, I'm trying to build a complete temperature monitoring system for my 951, everything from all vent temperatures, to intercooler temperature (in a couple spots), oil temp, water temp, everything, all going to my PIC microcontroller, using a 40x2 LCD display and a few buttons to navigate through all of the temperatures. If I ever make the thing work, expect a neat-o thread about it.

fast924S · fast924S's Garage

Wow, Here I am thinking I know alot about MAF, MAP and engine managment systems (since I went to school for them) But some one needs to school me on the terms you are using, like ADC's, DAC, Pic15f87 ect ect.... Im assuming the PIC is some type of prosessor lile a E-prom. I would love to site down with you guys and just come up will all kinds of ideas about this stugg, I love playing with it all

Danno how easy is it to insatll the S-AFC, I have heard of some rennlisters having trouble

Zero10 · 04-11-2005, 09:46 PM

Hehe, ADC, Analog-Digital-Converter, DAC Digital-Analog-Converter
PIC - Programmable Integrated circuit, PICMicro's are exactly what we're talking about ( www.microchip.com ) They're made my GE's semiconductor department, or so I have been told. For some of us, terms like this are in our daily vocabulary, for others, they sound like greek or something.

So, in our case, we're using PIC16F87/88's, decoded, this means PICMicro 16 series Flash memory model 87/88, the only difference is that on the 88, most of the input pins can be used for analog - digital conversions, a rather handy feature, for like $0.10/chip more.

I really like talking about this stuff, so don't get me started in this thread, I find it absolutely fascinating all the things that can be done with them.
I nailed down the problem with my PIC programmer... I can't read. After getting some really funny readings, it looked like the programmer kept browning out, so I read my power supply... 16VAC, nice, I thought it was 16VDC, time to build a bridge diode rectifier, but I'm fresh out of diodes... have to wait until tomorrow.

DannoXYZ · 08:17 PM

"Danno how easy is it to insatll the S-AFC, I have heard of some rennlisters having trouble"

That's because they're doing 1st-time R&D and programming. The instructions aren't very clear, giving you only settings for specific car types, like Sentra, Celica, Skyline, etc. But not what kind of sensors those cars uses. Some of them even have twin MAFs and the S-AFC does an averaging function and outputs a simulated twin-output.

A lot of trouble also comes from selecting the incorrect MAF. People didn't select one with the proper CFM->voltage calibration for their use, and thus have to do wild adjustments in the S-AFC. Or they pick one with similar response-curves to the stock AFM, and they run the computer straight into the edge of the maps and end up with the same flow-limit issues as the stock AFM.

Anyway, I've got it all worked out and gonna dyno-tune some 951s with this in the near future. This unit's been the stable foundation of the ricer upgrade aftermarket for over a decade, it's finally catching on with the Porsches. The extruded 3D mapping is simply ingenious as it's much more precise than straight volt-in -> volt-out massaging and it's much faster than a pure 3D map with discrete cells. Kinda the best of both worlds.