Individual Throttle Bodies with Motronics???

[911st]

I am wondering if there is a way to convert our Motronics 3.2 Carrera system to work with Individual Throttle Bodies like the kit that PMO makes.

See: http://www.pmocarb.com/products.htm#EFI_Throttle_Bodies_for_Porsche®_911® _Engines

It is very expensive and time consuming to convert to an aftermarket injection system to support an ITB conversion.

I am thinking that some type of interface controller could be developed that takes as input the existing sensor inputs available at the motor. Different sensors could then be installed with the ITB conversion and there signals could be adapted or conditioned by the interface controller to be compatible with the Motronics ECU architecture.

As you may know sensors at the motor include the AFM, WOT switch (triggering acceleration map), Idle switch(triggering idle map), and I think an oil & or head temp in addition to the other normal sensors.

With a change to ITB's there is no longer any way to measure air flow.

The main question I have is working with in the constraints of the architecture of our Motronics system what kind of inputs would best support a good running motor.

We have RPM, if we put a manifold pressure sensor in place of the AFM, is this enough to control a motor well?

My thinking is put a Throttle Position Switch in place of the AFM input and add a vac switch in place of the WOT switch to trigger the acceleration map when manifold vacuum drops with acceleration.

This would require reprogramming of the stock chip I would think however, the interface controller could have some mapping ability.

What I am thinking of is something where the factory plugs at the motor can plug into the interface controler that will adapt the signal output that is compatible with the existing Motronics inputs.

Could this work?

[vreference]

No measurement of air means speed-sensity. And yes, a TPS and IAT will be required at a minimum. A lot of the mustang guys like Speed desnity but it does mean you need to spend some time on a dyno whenever you make any changes to the way the engine breathes.

I say you'd spend a lot less time putting some random production speed density computer in the car. Other than getting the tables made the project would practically be free.

[911st]

What is the IAT and please tell me more about the speed density computer.
Is it a piggy back system?
What inputs dose it need?
Will it work with the Motronics?

I think of it as a "reference system" somewhat like MFI. If the RPM's and throttle is X, send Y fuel. Is Speed Density the proper term?

Thinking when running on the cruse map could keep the O2 in the loop like stock for efficiency. Then as soon as the the throttle is cracked the Vac switch could close triggering the Acceleration map.

Above about 5k, the AFM stalls and we are basically running without it anyway.

This is all a bit out of my knowledge so your help is appreciated.

I wonder if there is an opportunity for a Joint Venture between PMO, Split Second Timing or other, and SW for a turn key ITB plug and play system.

PMO kit is $2600, a interface controller at say $500, and chip at say $500. Could be a $3600 kit.

Thx!

[911st]

Why is this worth doing?

First it just looks plain cool!

ITB's should make for faster throttle response and more of that early 911 sound and feel.

It should reduce breathing restrictions and release some power.

It might allow more aggressive cams.

Saves the expense and work of going with a full injection change over.

[Flieger]

IAT= intake air temperature

ITB= individual throttle bodies

TPS= throttle position sensor

MAF= mass sirflow sensor

[vreference]

SD is just another way to determine how much fuel is required. Engine speed is used to look up an engine airflow rate (determined experimentally - on a dynamometer) for a given RPM. This value is used in conjunction with manifold absolute pressure and Intake air temperature to determine Air mass entering the engine. I believe most SD setups use a TPS as well to improve part-throttle drivability.

Right, this SD is a general term for computers that calculate air mass with this method. If you are looking to homebrew I would think it would be easier to get a computer that already uses SD rather than trying to convert the Motronic box. Part throttle drivability is the key and where you will run into issues.

[Bart_dood]

Originally Posted by 911st

I am wondering if there is a way to convert our Motronics 3.2 Carrera system to work with Individual Throttle Bodies like the kit that PMO makes.

See: http://www.pmocarb.com/products.htm#EFI_Throttle_Bodies_for_Porsche®_911® _Engines

It is very expensive and time consuming to convert to an aftermarket injection system to support an ITB conversion.

I am thinking that some type of interface controller could be developed that takes as input the existing sensor inputs available at the motor. Different sensors could then be installed with the ITB conversion and there signals could be adapted or conditioned by the interface controller to be compatible with the Motronics ECU architecture.

As you may know sensors at the motor include the AFM, WOT switch (triggering acceleration map), Idle switch(triggering idle map), and I think an oil & or head temp in addition to the other normal sensors.

With a change to ITB's there is no longer any way to measure air flow.

The main question I have is working with in the constraints of the architecture of our Motronics system what kind of inputs would best support a good running motor.

We have RPM, if we put a manifold pressure sensor in place of the AFM, is this enough to control a motor well?

My thinking is put a Throttle Position Switch in place of the AFM input and add a vac switch in place of the WOT switch to trigger the acceleration map when manifold vacuum drops with acceleration.

This would require reprogramming of the stock chip I would think however, the interface controller could have some mapping ability.

What I am thinking of is something where the factory plugs at the motor can plug into the interface controler that will adapt the signal output that is compatible with the existing Motronics inputs.

Could this work?

Something you might want to consider instead is fabricating a much larger single throttle and a custom air filter arrangement using a very large flat air filter element (perhaps from a modern 911?). This would give you most of the benefits of individual throttles but far less of the headaches.

[911st]

Thx guys.

Do we have Intake Air Temperature input on a stock Motronics?

Or just head and/or oil temps as inputs.

[911st]

I think we have the following set up with Motronics using my terms.

Idle map with O2 correction.

Cruse map with O2 correction.

Wide Open Throttle map.

If we put a TPS in place of the AFM we then change the map from AFM position by RPM to Throttle Position x RPM.

This alone should equal the sophistication of an MFI system.

Having the O2 correction on the idle and cruse maps should make for an improvement over MFI.

Then, substituting a vac switch in place of the WOT we can operate between the corrected cruse map and the Acceleration maps.

On top of this we still have the factory modifications to the fuel delivery and ignition depending on operating temperature.

Thus at idle and cruse we should be ok. WOT should be controllable.

As noted, it might ge the part throttle acceleration drivability that could be an issue.

Sure, this will not be an equal to a full Motec conversion.

[DW SD]

Originally Posted by Bart_dood

Something you might want to consider instead is fabricating a much larger single throttle and a custom air filter arrangement using a very large flat air filter element (perhaps from a modern 911?). This would give you most of the benefits of individual throttles but far less of the headaches.

EDIT - your use of the word "most" is what I'm questioning below.

Not true. Throttle response should be greatly improved with 6x throttle bodies even if the overall combined areas is equal to one large one. overall HP is likely improved, too.

It is more economical to arrange one throttle body and less effort to calibrate. Look at high performance naturally aspirated engines and they are mostly all configured this way. New Porsche GT3, Ferraris, included. They use one or two mass air flow sensors dumping to a manifold which lets air into one throttle per cylinder. I would have to dig out the thermodynamics books to support from an engineering perspective as to why. I would guess it has to do with flow momentum acceleration and the related velocities.

Doug

[911st]

Another possibility would be to put throttle plates at each intake port and still breath through the AFM.

Might get better response but not much gain in HP.

[EarlySport]

There are boxes out there to convert an AFM to MAP ( speed density ). Look up MAP-ECU ( made in NZ ), or the old HKS-VPC. You can run ITB's with MAP, somewhat dependant on how big a cam you run. So yes, you can use MAP-ECU or similar to allow you to use ITB's and a stock motronic computer. MAP-ECU is laptop programmable to control the fuel curve.

Most ITB's though are run with big cams where the MAP signal is less than ideal, and hence they are mapped on throttle position only. I don't believe the factory motronic computer can be reprogrammed to function this way.

At the end of the day, an aftermarket computer with full programmability ( Haltech, Microtech, SDS etc ) are under $1K. When you consider you're going to have to re-map your motronic (or your $400 MAP-ECU box) dyno time won't be much different between the two solutions. All the factory motronic sensors can be used with an aftermarket ECU too. So in my mind it's really a no brainer to spend the extra $500 or so and go full aftermarket ECU.

[Bart_dood]

Originally Posted by DW SD

EDIT - your use of the word "most" is what I'm questioning below.

Not true. Throttle response should be greatly improved with 6x throttle bodies even if the overall combined areas is equal to one large one. overall HP is likely improved, too.

It is more economical to arrange one throttle body and less effort to calibrate. Look at high performance naturally aspirated engines and they are mostly all configured this way. New Porsche GT3, Ferraris, included. They use one or two mass air flow sensors dumping to a manifold which lets air into one throttle per cylinder. I would have to dig out the thermodynamics books to support from an engineering perspective as to why. I would guess it has to do with flow momentum acceleration and the related velocities.

Doug

I'm not sure what the exact definition of "throttle response" actually is?

Do you mean in neutral when revving the engine? Or when in gear and going from cruise etc to flooring the throttle?

If you mean in gear then I doubt individual butterflies would make that much difference, perhaps a few milliseconds? not even measurable.