Carrera Air Flow Meter upgrade ideas?

[scarceller]

I own a modified 1984 Carrera 3.2L and have been brainstorming on some ideas for eliminating the Air Flow Meter (AFM, Barn Door air device on the intake). The car has the stock Motronic with a custom chip from Steve Wong.

What I really would like is a solution to completely remove the AFM. I know folks have installed modern Hot Wire air meter devices in it's place and while this is do-able it is still pricey.

So, I have another thought but need some fuel/air ratio experts to comment on it. Here it is:

The basics of the stock AFM are this:
The stock AFM sends out a simple analog signal (DC Voltage) between 0.1V to 5.0v based on the amount of air being taken in via the intake. 0.1V would be with no air, engine shutoff. 5.0V is at wide open throttle. At Idle my AFM reads 1.3V

The Idea:
Could it be possible to design a simple solution using a Throttle Position Sensor (TPS) and a Manifold Pressure Sensor (MAP) and a simple circuit? What if we added these 2 sensors and somehow simulated the same Air Flow Curve (0.1V-5.0V) produced by the stock AFM, if this could be done the simple circuit would also have some potentiometer adjustments for fine tunning idle O2 and CO at idle. The idea sure seems simple and would most likely cost under $100.00 in parts.

I got to thinking about this because many modern day fuel systems work via just a TPS and MAP sensor.

Thoughts? am I off the wall here? Let's here from our Air/Fuel ratio experts.

Thanks for reading.

[TroyGT]

Sounds interesting... but I have not knowledge to lend. Someone is bound to chime in on this.

-Troy

[ChrisBennet]

A couple of data points:
(1) The barn door air flow meter (afm) incorporates a temp sensor and the DME calculates the actual air flow using the temp and door position.
(2) Protomotive sells a MAP conversion for the Carrera.
http://www.geocities.com/protomotive/pressens.htm

-Chris

[scarceller]

Originally posted by ChrisBennet
A couple of data points:
(1) The barn door air flow meter (afm) incorporates a temp sensor and the DME calculates the actual air flow using the temp and door position.
(2) Protomotive sells a MAP conversion for the Carrera.
http://www.geocities.com/protomotive/pressens.htm

-Chris

Chris, thanks for the link. This confirms that someone has thought of this and is doing it. However, a price tag of close to $2000.00 seems very high. I would think one could replace the AFM with under $200.00 in parts. Promotive web site does not offer much detail, other than they can do it. Lets keep thinking about a 'simple' solution.

[drl911]

"What I really would like is a solution to completely remove the AFM."

I am all about cost effective DIY improvements, but enlighten me.... why change/replace the AFM?

What are the projected advantages?
Any trade-off disadvantages?

Dan

[ChrisBennet]

Originally posted by scarceller
Chris, thanks for the link. This confirms that someone has thought of this and is doing it. However, a price tag of close to $2000.00 seems very high. I would think one could replace the AFM with under $200.00 in parts. Promotive web site does not offer much detail, other than they can do it. Lets keep thinking about a 'simple' solution.

From the picture on their web site, it looks like they modify the DME board to add a pressure sensor.
-Chris

[scarceller]

Originally posted by drl911
"What I really would like is a solution to completely remove the AFM."

I am all about cost effective DIY improvements, but enlighten me.... why change/replace the AFM?

What are the projected advantages?
Any trade-off disadvantages?

Dan

Here's why: the stock AFM is restrictive to air flow and the motronic engines will take in air more effectively without the AFM restriction. Also, many of our AFMs are around 20 years old and they are mechanical devices, in my case I could not set the CO correctly because the spring tension on the spiral spring for the Barn Door inside the AFM had sagged due to age. In otherwords, the spring had gotton weak. I suspect this may be the case in many older Carrera Motronic cars. Of course this is only my experience. I'm just willing to try a new approach and see what we can come up with. The idea of only using TPS and MAP sensor is not new, many after market Fuel Control systems use this (ie: Simple Ignition Systems 'SDS' as well as MegaSquirt) this is what got me thinking about this new approach.

[ChrisBennet]

Just thinking out loud...
If you use a placed a map sensor below the throttle body and monitored the signal while simultaneously monitoring the signals coming from the the air flow meter (temp and barn door signal) you could (probably) make a transfer "map" that you could could translate the signals into something the DME would recognize. (Think PIC or PSoC processor doing AtoD and DtoA.)
Gotta get back to work now but I could expand on this later if it isn't obvious.
-Chris
Edit for spelling, so shoot me.

[scarceller]

Originally posted by ChrisBennet
Just thinking out loud...
If you use a placed a map sensor below the throttle body and monitored the signal while simultaneously monitoring the signals coming from the the air flow meter (temp and barn door signal) you could (probably) make a transfer "map" that you could could translate the signals into something the DME would recognize. (Think PIC or PSoC processor doing AtoD and DtoA.)
Gotta get back to work now but I could expand on this later if it isn't obvious.
-Chris
Edit for spelling, so shoot me.

Chris, I do understand and it should be easy to add the MAP and record it's ratio against the AFM signal.

As I recall the Air Temp Sensor signal travels on seprate wire to the MCU, I think it is only housed in the AFM assembly. Will need to double check my Motronic diag book. If I'm correct I suggest leaving the stock Air Temp Sensor in the air stream. Sort of replace the AFM with a Alum. tube that would still house the Air Temp Sensor.

So, don't replace to much at once. Just replace the Barn Door Functionality and it's simple 0.1v to 5.0v signal.

What I'm not sure about: is the Air Flow Input a 1to1 relationship to Manifold Pressure? My guess is the answer is NO. For example: say you open throttle to 10% with unloaded engine (car in Nuetral) you will get a one MAP signal. But if you open throttle to same 10% underload (driving) you will have a much lower RPM in the engine than at no load. So I'm sure the MAP signal will be diffrent. What I'm getting at is that I think Air Flow In is more complex than just using a MAP signal, more than likely you will need to look at MAP, TPS and possibly RPM.

We really need some advice from a pro familiar with Air/Fuel ratio and how air flow relates to load.

Maybe I'm wrong here and it would be nice if Air Flow is always 1to1 with MAP.

[89911]

I ran (and sold) a mass airflow system on my car for 3-4 years (3.2l carrera). When I rebuilt my engine using SteveW's chips, the biggest hurdle was getting it to run correctly. Finally took it off, used the OE system, and never looked back. My unscientific spin on the AFM vs Stock intake is this; Unless you redesign the complete motronic system using this (which they did in the 993), your at the mercy of a system that is lacking in its complete integration. Sure, you could add sensors and connections to supposedly improve it, but I have yet to see accurate dyno numbers to prove this. Steve W told me the biggest improvement users had with the Autothority MAF and chip is from the chip itself. In the end I sold it for less the half I paid for it. I'd put the money into bigger P/C's or headers for that kind of money.

[JFairman]

I spent alot of time modifying those airflow meters on BMW's back in the late 90's
It's the exact same heavy alloy casting, just the resistance values and spring tension on the flapper door is different.
There isn't alot you can do to improve the Bosch unit and keep it driveable during transition from idle to just above idle.
You can adjust the CO to an extent and the big clock type spring tension that acts on the door, and you can reset the pointer contact position on the resistance trace while at idle but not much else.
The cool thing is Nippondenso made a copy of the Bosch air flow meter and put it on the Toyota Supra back in the late 80's or so.
This one is made of cast aluminum and is about 4 LBS lighter than the magnesium/zinc alloy Bosch used and you can take the metal case apart because it is put together with silicone glue and small screws while the Bosch unit is riveted and you will destroy it if you take it apart.
I'm not talking about the plastic cap that covers the electronics. That is silicone glued on and you cut the edges of the glue with a razor, then pry it off carefully.

The circut board from the bosch needs to be removed and it fits right into the Toyota unit after just slightly rounding off one small corner of the circut board with a file or sandpaper. The screw holes line right up perfect.
Nippondenso did a good job of copying the Bosch unit and improving it so all the parts are interchangeable, but you only need to swap the circut board.

The Nippondenso unit also flows a little more air than the Bosch one. The person who first discovered this at "Metric Mechanic" has a flow bench and he checked and found this.
Also because the top metal cover is removeable you can port the interior of the flow meter when it is at the full throttle position, and then reassemble it.
Thats impossible with the much heavier Bosch unit.

Go to http://www.metricmechanic.com/ maybe he still talks about this easy, cheap conversion in there still.
I did this on a hot rodded euro 635 motor back in the mid 90's.
I got the air flow meter out of a toyota supra in a junk yard for $15.
They also used the same AFM in some of the big Toyota sedans that used the same DOHC inline 6cyl motor as the supra back then.
Nippondenso copied alot of Bosch parts under license from Bosch and or they changed them just a little bit... usually for the better.

[Bearded Viking]

Expensive, but Steve Wong sells this setup.

http://www.911chips.com/HFM1.htm

[ianc]

I too am very interested in this thread. My AFM is 20 years old as mentioned. While I have set the wiper on a new resistive track, I'm still having jerky transition issues at on\off throttle position. I'm looking for any reasonable alternatives to the $440 replacement Bosch part.

ianc

[scarceller]

Originally Posted by ianc

I too am very interested in this thread. My AFM is 20 years old as mentioned. While I have set the wiper on a new resistive track, I'm still having jerky transition issues at on\off throttle position. I'm looking for any reasonable alternatives to the $440 replacement Bosch part.

ianc

Jerky transition at on/off throttle could be your idle stop switch (the micro switch on the Throttle body that closes when the throttle goes to idle). Double check that it is working and most importantly that it opens at less than 1 Degree off idle. This switch should open as soon as you move the throttle off idle, or as soon as possible. It could be it is opening late.

My car had this set wrong and it caused the car to sort of Kangaroo.

[JFairman]

On all the L jetronic and motronic AFM's I've been into the microswitch that opens at the door stop is the fuel pump switch.
It's there so the fuel pump won't come on till the engine is turning over and sucking enough air to open the AFM door a tiny bit... then that switch closes and the fuel pump comes on.
It's a safety thing so if the car was in an accident and the motor is off but the ignition is on, the fuel pump shuts off too.
CIS injection has a similar microswitch at the sensor plate.

It's common practice to gently insert a screwdriver in the AFM door to hold it open with the ignition on and motor not running to turn on the fuel pump for checking fuel pressure or other motor off/fuel pump on diagnostics.
Thats easier than jumping the fuel pump relays.

[jbrinkley]

J, it really is an off idle switch on a motronic 911. Fuel pumps on carreras start on the ref sensor signal from the flywheel.

GREAT info about the nippon

[scarceller]

Originally Posted by JFairman

On all the L jetronic and motronic AFM's I've been into the microswitch that opens at the door stop is the fuel pump switch.
It's there so the fuel pump won't come on till the engine is turning over and sucking enough air to open the AFM door a tiny bit... then that switch closes and the fuel pump comes on.
It's a safety thing so if the car was in an accident and the motor is off but the ignition is on, the fuel pump shuts off too.
CIS injection has a similar microswitch at the sensor plate.

It's common practice to gently insert a screwdriver in the AFM door to hold it open with the ignition on and motor not running to turn on the fuel pump for checking fuel pressure or other motor off/fuel pump on diagnostics.
Thats easier than jumping the fuel pump relays.

What AFMs have you been into? the AFM on my 84 Carrera 3.2L has has no such switch in the AFM. The fuel pump circuit (relay) is controlled by the DME box. Has nothing to do with AFM. Are you looking at other AFMs (non Carrera 84-89)?

[JFairman]

..yeah, it's older ones I'm more experienced with and figured that part was still the same. Mostly from L-jetronic BMW motors.
The microswitch I'm talking about is under the glued on plastic cover. Maybe they don't use it in the later motronic setups.

It's the same heavy magnesium/zinc AFM casting though.
The nippondenso copy of the bosch AFM used in older toyota supra's is a ALOT better and lighter (made of lighterweight cast aluminum) and serviceable (it can be taken apart) than the heavy bosch casting.
You can still find them in junkyards and convert them over for a little more top end performance.
They are around 4 pounds lighter, and look nicer too.
I went into more detail about it in a post above.

[stlrj]

Originally Posted by 89911

My unscientific spin on the AFM vs Stock intake is this; Unless you redesign the complete motronic system using this (which they did in the 993), your at the mercy of a system that is lacking in its complete integration.

That's precisely why I gutted a complete MAF TPI system from an 85 Chevy 2.8 when I converted my 78 3.0 to EFI. I even got the on board diagnostics working well enough to where I was able to plug into the Chevy dealership diagnostics.

Best part about the experiment was that it cost me less than $250 for the parts at the wrecker.

$2000 for a half baked system? You gotta be kidding!

By the way, you won't believe the amount of free tech support I got from the Chevy dealership technicians. I just can't say enough about them.

Cheers,

Joe

[dmwallace]

Interesting thread!

Is anyone running Steve Wong's AFM replacement - if so do you have any comments about it?