Curved runners from stock 2.0L intake manifold mounted to adapter plates.	Curved runners from stock 2.0L intake manifold mounted to adapter plates.	40mm air horns wire tied to the ends of the curved runners.

With this induction system temporarily held together well enough to drive the car I went out for a series of test drives.  I made no other changes to the engine or the tuning.  I would be able to see any change via the change in VE and any extra fuel required to maintain mixture settings.  The initial tune was within 3% of ideal prior to the intake changes. 

Results of the Experiment

I was not sure if I should expect to feel any difference but as soon as I was out of my neighborhood (open intake is very loud), I could tell this was going to be good!  The car was pulling noticeably better starting at about 3500 RPM all the way up to about 5500 RPM.  I had my EGO correction set to a maximum of 10% and I was running very lean in this region of the tune telling me I had made more than a 10% improvement in VE.  When I analyzed the datalog from my test drive I calculated the following improvements to VE at wide open throttle:

The datalogs from the test drives show that the longer intake starts to hurt performance right around 6500 RPM.  This agrees with the simple calculations provided from the Jenvey FAQ page.  Keep in mind that the 40mm ITBs were not port matched to the 42mm curved runners and that I had 40mm air horns wire-tied to the runners as well!  These results were obtained with two separate 1mm deep sharp edges within the intake runner.  When I get rid of that turbulence the VE improvement should increase.

I re-tuned my VE tables and verified the improvement.  This was more than double the improvement I had expected in my most optimistic estimates.  I would have been happy with a 5% to 10% improvement in VE but 20% was completely unexpected.  I guess the folks at Jenvey knew what they were talking about with the kind of power you are throwing away with runner lengths that are too short.

After the tune was dialed in, I went to a long straight flat road near by that I use for 3rd gear pulls   I compared a 3rd gear pull on this same road from prior to the intake changes using MegalogViewer's HP calculation.  This type of HP estimate is not accurate for a true reading on horsepower but it works pretty well for relative comparison.  The results were between a 10% and 15% increase in horsepower within the 3500 to 5500 RPM range at wide open throttle.  Based on my last dyno run, this should translate into an additional 10 to 15 HP!  Not a bad gain for simply adding 9 inches to the intake runner length.

Making This Real

So now I need to take this temporary setup and turn it into something that will work for daily driving.  This will require the fabrication of a new intake plenum to tie the intake runners to my air filter and cold air intake box.  To make room for the new intake plenum I need to clear out some space above the ITBs.  The specific changes I need to make are:

• Relocate the throttle linkage
• Design a new vacuum manifold and idle air system to fit under and around the plenum
• Relocate the MS coolant temperature sensor
• Fabricate a custom plenum

Relocating Throttle Linkage

The Jenvey throttle linkage is designed to be mounted in any of four different locations.  It can be mounted to the front or rear throttle body and above or below each of the throttle bodies.  There is a different mounting bracket required for mounting the linkage above or below the throttle body.  When I built my 2.2L stroker motor I decided to upgrade to the S14 starter which is smaller than the E21 starter.  Because of this decision, I had just enough room under the rear throttle body to mount the throttle linkage.  I fabricated the required mounting bracket out of 12ga galvanized steel that I picked up at a hardware store.

Test fit of relocated throttle linkage.	I will need to fabricate a shield for the starter solenoid. I don't want the throttle cable to ever touch the terminals on the back of the solenoid.	The space that will be used for the intake plenum.

Relocating the Coolant Temp Sensor

I installed a coolant temperature sensor from an E30 318i for use with my Megasquirt install.  This sensor interferes with the area needed for the plenum.  To make room for the plenum I relocated this sensor from where I had it installed to a new port that I placed at the end of the coolant diverter casting.  There is just enough material in the casting to allow the M12 hole to be drilled and tapped to accept the sensor.

I drilled and tapped a new M12 x 1.5 hole in the end of the casting to relocate the MS coolant temp sensor. Coolant temp sensor relocated to add clearance for the plenum. I still need to cap off the old location.

New Vacuum Manifold and Idle Air Design

The new plenum design required a complete re-design of my idle air circuit.  I pretty much allowed the plenum to dictate where I had room for the idle air valve.  The best location ended up being along side the plenum.  One enhancement I made was to locate the idle valve in the middle of the vacuum manifold.  I noticed this was how BMW did it on their engines running ITBs and figured it might help more evenly distribute the idle air to the throttles.

I followed similar construction techniques as my first vacuum manifold including the built-in MAP sampling port and using a combination of copper and brass and regular plumber's solder to build up the manifold.

Testing the clearances around where I want to place the idle air valve. You can see the unfinished copper vacuum manifold running next to the valve cover.	New vacuum manifold installed. You can see the hoses running from the manifold to each throttle body as well as the centrally located idle valve port on the top center of the manifold. The MAP sampling port is on the far right end.	Idle air valve installed on completed plenum. The idle air intake is pressed through a grommet in the plenum cover.

 

Designing the Plenum

My intention is that the plenum volume will not have a large effect on the tuned induction system.  My testing was done without any plenum (infinite volume).  The goal will be to make the internal plenum volume as large as possible to approximate the same running conditions I had during my test drives.  With the foam form almost complete for the plenum, the volume works out to be somewhere between 4.5 and 4.9 liters.  This is well over 2x the engine's displacement and should not limit the engine's performance.

The plenum will be designed in two pieces similar to the TWM plenum I had been using.  A removable cover will provide access to the air horns for assembly as well as throttle balancing.  The plenum will be made using a foam form and laying fiberglass over the form.  I think I will try vacuum bagging the form after laying up the fiberglass.  Once the fiberglass has set the foam will be dissolved using acetone.  This is a pretty common DIY method for making complex shapes out of fiberglass or carbon fiber.  One of the negative points of this method is that the foam form will be destroyed so I will only get one chance to get this right.

I built up the foam form using 1 inch and 1/2 inch thick sheets of foam glued together and shaped with a combination of razor blades, files, and sandpaper.  The shape of the plenum close to the air horns is somewhat complex to provide clearance for the fuel rail and injectors that are immediately below and behind the air horns.  I used some wood dowels as pins into a couple of the runner mounting holes so I could easily position the form on the engine during fit checks.

Rough shape of the foam form for the plenum base. You can see the taper and the cut-outs necessary to clear the fuel rail and fuel injectors. The dowels help locate the form into the intake runners.	Test fit of the plenum base form. At this point the form clears the injectors and the strut bar. I still need to work on the front where the air intake will attach.	Another view of the base form. The taper on the front is to clear the hood.
This is as deep as the plenum base will be. The plenum cover will be about the same depth so the total plenum will be about twice as deep as the base shown here. (about 5 or 6 inches deep)

I continue to refine the plenum shape as I go.  I simplified the shape around the fuel rail to make it easier to lay-up and increase the chances of success.  I was able to build up a shape to allow the 3 inch air intake to run around the top coolant hose and align pretty well with the cold air intake.

Completed plenum form showing air inlet and simpified fuel rail area.	Plenum base and cover. I still need to add a 1 inch lip to the cover that will overlap the base.	Plenum test fit on engine.
View of the air intake routing around the coolant hose and alternator.	You can see approximately where the idle air valve will be located in this picture. Total plenum volume is well over 4500cc.

The last details added to the base were a mounting feature for an aluminum bung for the intake air sensor and some notches for the  plenum cap mounting latches.

Plenum Fabrication

This was my first attempt at fabricating a part of this complexity.  I have modified my old TWM plenum a couple times but have never fabricated a part from scratch.  I spent a fair amount of time researching how to do this on the web and assembled all the materials I would need for the lay up and the vacuum press.  I purchased most of the materials from Fibre Glast.  There is a nice introduction to vacuum bagging on their web site here.  There are many resources on the web for purchasing these supplies, this is just the place that I ended up using.

Materials used, product numbers are from FibreGlast web site:

• System 2000 epoxy resin with 2120 hardener (120 minutes pot life)
• 43-B black pigment
• 1094 Bi-directional E-glass 9oz/sq yd
• 1678 Stretchlon 200 bagging film
• 583 Polyester release peel-ply
• 579 Breather-Bleeder
• 581 Grey sealant tape
• 891 Vacuum connector
• 909 Two way shutoff valve
• 893 vacuum tubing
• 1/4 inch Plexiglas for mold base
• car wax for mold release
• vacuum pump
• Mixing cup, stirrer

I used a sheet of 1/4 inch thick Plexiglas that I picked up from my local hardware store for the mold base.  I treated the surface with several coats of car wax so the part would not stick.  I was careful to not wax the edges where the sealant tape needs to go.  Next I placed the sealant tape around the edges and stuck the foam form on the base with some double sided tape.

I used between 6 and 8 layers of glass for the base.  6 layers on the rectangular region and about 8 layers around the round air intake region.  I tinted the epoxy resin black just for looks.  The part will be painted but when the paint chips, the black will still look pretty good.  Resin was brushed into each piece of glass, I wrapped the foam form with packing tape to keep the epoxy from being soaked up by the foam.

Base and foam form ready for lay up. There is a layer of packing tape over the foam.	First layer of glass test fit over the foam. I used two main pieces per layer. One around the intake at the front, and one over the rectangular portion.	Tinted epoxy was then brushed onto the glass one piece at a time.
Alternating layers. A total of between 6 and 8 layers were used with more layers around the circular intake.	Lay up ready for peel-ply.

After the lay up was complete, I placed a layer of peel-ply over it.  The peel-ply had been cut to size and test fit before the lay up was started (notice arrows etc on the peel-ply).  The 120 minute pot life of the epoxy is plenty long enough so I did not need to rush but you also don't want to waste time.  After the peel-ply, several layers of breather cloth were placed onto the part with extra breather around any sharp edges.  Next the bagging film was placed over the part.  Several pleats were made in the film to allow it to drape over the part better.  The bagging film stretches but you don't want to rely entirely on this stretching, pleats are always a good idea on a part that has multiple heights like this one.

First piece of peel-ply fabric placed on part. Notice cuts in fabric.	Second piece of peel-ply placed on part. Notice arrow and cuts. These pieces were pre-cut and test fit prior to lay-up.	Breather fabric places over part. Notice extra breather that will be under the vacuum fitting.
Bagging film installed with vacuum fitting and pleats.	I picked up this Gast vacuum pump on ebay for $45.	Vacuum being pulled on part. Notice the extra resin being drawn into the breather fabric.

I held the vacuum on the part for about 6 hours and then allowed the part to cure for 24 hours before removing the breather and peel-ply.  Removing the breather and peel-ply was not too difficult.  The part popped off the base easily as well due to the coat of car wax.

I trimmed the edges and removed the packing tape from the bottom of the foam form.  The foam was then dissolved using some acetone.  The acetone quickly dissolves the foam turning it into a messy white paste.  Once the foam was gone, the layer of packing tape was pulled from the inside of the part.

After 24 hours cure time the part was removed from the base.	Picture of the foam being disolved with acetone.	What's left of the foam form. This took less than a minute to happen once the acetone was added. You can see the packing tape remaining inside the part.
The foam and packing tape removed from the inside of the part.	Finished part after some rough sanding and trimming.	Inside of the part. The packing tape surface leaves a nice smooth finish.

 

I followed exactly the same procedure used for the plenum base in fabricating the plenum cover.  The cover was easier to lay up since its shape is simpler.

Plenum cover layed up and under vacuum. The procedure followed was exactly the same as for the plenum base. 4 layers of glass were used for the cover. Plenum cover after rough sanding.

I purchased an aluminum bung for mounting the intake air temp sensor.  I used a file to make several grooves in the outside of the bung and then epoxied it into the gusset that I had created in the bottom of the plenum base.

Intake air temp sensor threaded into the bung that I placed into the plenum base.

 

Complete plenum, top view. Complete plenum, bottom view.

Here are a couple pictures of the assembled plenum, test fitting the cover.  I still need to do a lot of finish sanding and filling before I paint it.  I also need to shape the grooves that will accept the cover mounting latches.

 

Here are a few pictures of the completed plenum placed into position on the engine.  I epoxied a piece of the 1/4 inch fiberglass plate into the plenum base.  This plate will provide the material thickness needed to countersink the mounting hardware for the intake runners.

Plenum placed into location on the engine. There is enough clearance between the plenum and valve cover to remove the plenum cover and to locate the vacuum manifold and idle air control valve. Another view of the plenum test fit. And one more view... You can see the intake air temp sensor and the 1/4 inch plate in this view with the cover removed.

Completed Project

I ordered some 1/4 inch aluminum plate and fabricated the final version of the adapter brackets.  The only tools used were a drill press with a hole saw and basic hand tools to make the adapters.  After about 6 passes with wet sanding, filling, and painting I got the surface finish on the plenum where I wanted it.  I ordered a set of stubby cast air horns from Jenvey for the inside of the plenum.  I used my dremel to shape the 1/4 inch fiberglass plate inside the plenum to blend the transition from the air horns to the curved runners.

Aluminum adapter brackets installed on the throttle bodies.	A view of the inside of completed plenum showing the stubby air horns installed.
The completed intake.	Top view of engine with completed intake.