New MAF system with ITB's Options

[904svo]

Here are pictures of the old and new MAF air flow system.
Injectors mounted on ITB's
Old way



New way, need some tig welding but it runs great now
Injectors mounted on manifold close to valves

[utah914]

Wow, that is really cool! (IMG:style_emoticons/default/aktion035.gif)

I've been thinking about putting this setup together for quite a while. MAF is a nice direct signal. So, how was it to set up and tune? And how does it run? I'm looking for a buttery smooth idle. How's yours?

Oh, and more details please on what components you used. (manifolds, throttle bodies, MAF, computer, etc.)

[VaccaRabite]

Aren't you worried about cylinders getting uneven amounts of air? Seems like 2 and 4 would get most of the air.

Zach

[JFJ914]

Aren't you worried about cylinders getting uneven amounts of air? Seems like 2 and 4 would get most of the air.

Zach

The short answer is no. The engine is a pump, it sucks air in, there are no flow impediments to keep the cyls from sucking all the air they can.

[utah914]

I'm wondering the same thing about even airflow. A well designed larger airbox around the ITB's would even things out. Kind of like BMW did on their E46 M3 CSL intake. They use ITB's with a LARGE airbox and a single MAF.

On a standard 914, seems like one could use the standard EFI hardware and mount a MAF sensor before the single throttle body and get rid of the MPS. Obviously ITB's are the only way to go for 904svo's car.

And while you're at it, once you've got the MAF and crankfire ignition, might as well go sequential injection. Hmmm.... maybe this would cure my emissions woes. (IMG:style_emoticons/default/idea.gif)

[rohar]

It theoretically could be a problem depending on primary plenum size. Compared to the throttles under there, the plenums are massive so there should be absolutely no issue.

[JFJ914]

Wow, that is really cool! (IMG:style_emoticons/default/aktion035.gif)

I've been thinking about putting this setup together for quite a while. MAF is a nice direct signal. So, how was it to set up and tune? And how does it run? I'm looking for a buttery smooth idle. How's yours?

Oh, and more details please on what components you used. (manifolds, throttle bodies, MAF, computer, etc.)

Yes it idles like butter!

[utah914]

John, are you running MAF too? Can you post some video? I'm REALLY interested.

[JFJ914]

Nice job, Larry. I've watched this project develop over the last 2 years and Larry has made almost every mistake possible listening to "Internet Experts" about how Megasquirt works. My advice to anyone about to embark on their own project is that 99.5% of expert advice is wrong. You are really on your own and like Larry you need to study theory as if you are getting a PHD. My only advice, and it's not tested yet, is start with stock D or L-Jet manifolding and intergrate a 2.0L VW MAF sensor. Don't worry about "it's too restrictive" theory. Do not waste your time with MAP or Throttle Position, it will never work well especially with hotter cams. Do one system at a time, do not do the cooling, fuel and ignition at the same time, it could add years of development time! Good luck!

[JFJ914]

John, are you running MAF too? Can you post some video? I'm REALLY interested.

No, I've only been a sounding board for Larry to bounce ideas off of. He has put in over 2 years of almost 7 days a week work on this entire project. I cannot but stand in awe at the laser like focus and dedication he has brought to this project. The car has passed emissions and is street licensed.

[biggy72]

Do your throttle bodies have bell mouths on the tops? Even if they didn't it would probably run fine, but they help the air turn the corner more evenly. The plenum volume has more to do with the acoustic dynamics than it has anything to do with whether the cylinders will get air evenly.

In college we had three different intakes over three years on I4 engines all with side entry and they all ran well (when tuned properly) and the plugs always looked even.

[utah914]

John, are you running MAF too? Can you post some video? I'm REALLY interested.

No, I've only been a sounding board for Larry to bounce ideas off of. He has put in over 2 years of almost 7 days a week work on this entire project. I cannot but stand in awe at the laser like focus and dedication he has brought to this project. The car has passed emissions and is street licensed.

Very impressive! So, Larry, I guess we'll look to you for some video if you get a chance. Can't wait to hear it run.

EFI really only does one thing. Tells the injectors when to turn on and for how long. The rest is just airflow data collection and analysis. MAF is the direct way to measure airflow. Funny so many people shy away from it. As long as your MAF is well calibrated and you've got a nice wideband setup, tuning should go pretty smoothly. In theory anyway.

[904svo]

Here are the specs on my engine.

Manifolds are modified 914 to hold the TBI's.
Throttle body's are from a Honda motorcycle 600CBR
MAF is Hot wire type from a late model Ford Mustang
Computer is a Megasquirt running MS1 Extra code
Injectors are from Honda motorcycle 600CBR
Spark is from VW Cops running in wasted spark mode.
Using a 36-1 tooth wheel for timing mounted on the crank using a Ford VR Sensor

Different injectors location were tried, the best location was the stock location
I try MAP, vacuum signal not stable enough for the VE table
TPS was tried next, but due to inspection requirements for emissions requirements
in my state. The VE/TPS table at idle would cause it to fail.

After reading about all the trouble you with have with MAF. I figure it was worth a try after doing some test. I could not see where MAF was causing air flow problems
that were reported by people, Yes a Trap Door Sensor has air flow problems.

I then rework the air intake system for MAF operation, DAM if it start up on the
first try. I then used TS VE Live to tune the VE tables the more I drove it the
better it got. I then rework the cold start table and got the engine to start and idle
at 900 RP M's. After that I work on the Acceleration pump set up and now have a smooth transition from idle.

As John Jentz said I would use the standard 914 intake system and change it to use a Ford MAF sensor and Megasquirt MS1 Extra code.

It's engine is nice a smooth now, thru all gears and RPM's ranges.

Would I do it again YES! I'm building a 2056 type 4 engine now. Which I'm planning
to run with MAF now that I know how it runs much better then when it was equipped
with Weber's Carbs ( which did not pass emission testing at all)

[utah914]

Thank you for the great info. So, how does the smoothness compare to the stock EFI? And how's the exhaust smell? Strange question, I know, but a well-tuned engine (not rich or lean) doesn't smell strongly. Also an indication that emissions are low.

I think I'm going to go for it with the stock hardware and use the Ford MAF. Did you make your trigger wheel yourself?

[904svo]

The throttle response is crisp and smooth and has no smell out of the tail pipe.

[FourBlades]

Looks great! That is a beautiful car.

I remember the ride you took me for at Docs tranny rebuild class. (IMG:style_emoticons/default/smilie_pokal.gif)

Is it a lot quieter with this intake system?

Did you make the intake pipes long to help engine tuning or more for packaging
within the engine bay?

John

[904svo]

I can hear a valve that needs adjustment now, the intake runners I made up to fit
the ITB's. The only problem I have was the Cone air filter got wet when I wash the car. Here is my new air filter ( Thanks to Al Meredith ) box mounted in the car
and the deck still closes.