The Top Mitsubishi Galant VR-4 Resource

Join the best E39A 1991-1992 Mitsubishi Galant VR-4 community and document your GVR4 journey.

  • Software Upgraded - Reset Your Password to Login
    In order to log in after the forum software change, you need to reset your password. If you don't have access to the email address you used to register your account, you won't be able to reset your password. In that case, follow the instructions here to regain access to the forum.

Build your own intake manifold


Well-known member
May 15, 2005
Rochester, New York
Build your own intake manifold


Last year at Watkins Glen my car was fast but it really just fell on its face in 4th and 5th. The car was using a Cyclone Intake manifold with a waste gate actuator fabbed up to operate the secondaries. As much as I love it I think its holding me back. I normally take the car up to 8500 rpms and at those rpms the intake plenum is just too small. With a limited budget and the price of intakes going up I decided to build my own.

Background theory-

For the design I am following some basic manifold building rules. First the runners should be 9" from the plenum to the intake guide seat. The distance on a 4g63 head from the gasket surface to the valve seat is just over 4". We all know that the longer the runner the better a manifold performs at lower rpms. But there is another rule to follow. That is the plenum should be at least 2.5 times the volume of 4 runners (gasket to base of plenum). The tricky part is that you can move the whole power curve and peak horsepower numbers either up or down depending on plenum volume. Bigger the plenum the higher the peak horsepower.

I want to make good power all the way up to 8500 rpms but I want some grunt down low too. My best design so far uses a 6" runner with a plenum volume just under 3 liters. I am planning on building the manifold by machining a 1/2" plate to act as the plenum floor that would feature velocity stacks (radiused entries) built in. That means they would be milled into the stock rather than rolling the end of a tube. I found ready made velocity stacks but they are $40 each and that's just too much money for me. I am planning on a plenum length of 16-17" inches with it tapering from 4" to 2.5".

I got most of my design ideas from Google including this site.


The plan is a single plenum just under 3L in volume with 6" runners. That would give me about 10.1" from the beginning of the runner to the edge of the valve seat. From what I've seen I thought the 6" runners would be longer than what others have done and the .125 thick runners yield a 2" ID which is also smaller than what others have done. I think the slightly longer and smaller runners would help low end power when combined with a slightly lower than 2.5 times plenum volume. I think making more power up top is easy and I really don't think I can mess that up. Making power down low is another thing all together.

What I think is that intake charge velocity controls cylinder filling at lower rpms and that like the 2G and EVO head design, smaller is better. At what point (rpm) does this smaller runner become a bottle neck I don't know. I am betting the farm that combining 1G port size dimensions with a big plenum would give me the best of both worlds. Much in the same way it would if I were to merely cut the plenum off a stock manifold and just weld on a bigger plenum. From my research that has been done with success.


Table saw with carbide metal cutting blade
Would router with ½” round-over bit and dado bits
Drill press with bits
Metric taps


This is where I sourced the materials for the build. I had some ½’ aluminum fixture plate that I got from the scrape bin at work but that can be purchased online as well.

For the plenum I am using 4x4" OD 6063 T52 aluminum square tube with a .125" wall thickness. The runners are going to be made out of 2.25" OD 6061 T6 aluminum tube also with a .125" wall thickness. That would give me a 2" ID on the runners and 3.5" on the plenum. I chose the thicker .125" wall thickness to ensure maximum strength while giving the welder lots of meat to weld with. The cost between thinner stock was minimal.



Building the manifold-

Lube the blade with metal cutting fluid (Home Depot) and have at it.


This next photo shows how nice the of an edge the blade left.


Then I milled a lip on both sides of the 4" strip I just cut out. This will receive the box top after its cut.


Then I cut the bottom of the plenum box off so it would fit on the milled base plate.



Later I'll cut an angle so the plenum will have a taper to it. For now it’s just dry fitted to see how it fits.


Then I cut the 2 1/2" hole for the TB in another piece of 1/2" plate I had. I just randomly cut the hole so I could practice milling the velocity stacks.


The plan is to have a velocity stack on the back side of the TB flange as well. I cut the hole in a separate sheet so I would have a wide base for the router which is what I'm going to use to round over the edge forming the stack.

Here are the pictures of the TB radius exit. This was my trial run with the wood router on aluminum and I have since improved my technique.


Next up was cutting the velocity stacks for the runners. I am using a ball bearing guided bit for the router which needed more than 1/2" of depth to cut the full radius so I cut another hole out of the plate material and then clamped the two pieces together. That way I had a full 1" for the bearing to ride on.




The next operation was to mill the backside of the plenum floor to open up the 2" holes to 2 1/4" so the runners would slightly fit into the plate. This is going to help properly align them and allow for a very strong weld. It should also make smoothing the entry a snap during the welding and assembly. I did that with my hole saw and my drill press.



These next photos show the transition between the velocity stack in the plenum floor and the intake runner.



Finally the money shot showing the plenum floor with the box and one runner mocked up.


Laying out the flange



Here's the flange. The bottom is notched to provide some overlap with the base plate and it has a 5 degree slope on the top to match the taper in the plenum. The idea is to build the intake in a way that allows the welder to crank up the power and really get a good weld. I think that when combined with the extruded plenum box will make it super strong.




Here's a close up of the TB flange. It looks like the overlap is only 1/8" inch but both the base plate and the TB flange are from 1/2" plate stock which should allow for a strong weld.


Here's a money shot showing the partially assembled plenum box/floor/and TB plate.


Building the end plate from 1/2" stock.


I also had a bit of an accident while milling that plate on the table saw. It was getting to small to guide with the cross-cut guide so I free-handed most of the cuts. After I the last pass I held the piece just beyond the blade and I reached down and turned the saw off. As it spun down I took my hand off the piece and the vibration of the saw sucked it back into the blade. Man that makes a sick sound and the back plate went flying past my head. Good thing I've been wearing my full face shield.

Once I found the plate the only damage was this gash. Could have been worse.


Here are a few shots of the plenum with both the TB and end plate.




Working the flange by hand with the dado cutting bit in the router


Here it is after I've cleaned it up a bit.



Here's a close-up after a bit of drum sanding


Crushing the round tubes to match the oval port shapes


Here they are mocked up in the flange.


The intake and TB flanges.



Here are the runners mocked up with the intake base plate and intake flange.



As you can see in the close up photo's of the runner to intake flange shots there is a small gap between the port shape and my crushed tube. That can be easily filled with weld and then ported smooth on the inside after everything is done.

Counting the thickness of the top plate and flange the runners will be 6". Here is a couple of mock up shots.



Believe it or not but making the throttle cable bracket was a royal pain. Getting that angle right and then spacing the cable bracket just so the cable lines up and has enough adjustment and accounting for the taper in the plenum. Much sailor talk was needed....


Here is a 3/8" plate that will be welded to the bottom of the intake to give me lots of meat to tap my manifold boost/vacuum sources.


For good measure I've stepped up to a 63mm NT TB with no FIAV or ISC. I figure the car stalls enough already that why not ditch em when I can. Any reduction in water lines is going to be a plus in my book. There's also a polyethylene intake gasket too. That's supposed to provide a heat barrier to keep the intake cooler.


Welding it all together. Please note that I am not a welder and that I hired a guy to weld everything up for me.

This is the jig we built to hold everything together flat and true.


onto the welding











All done!




and finally, where any good race part belongs...the coffee table shot


Mounted on the car




Here's the new IC piping.



The last step in the process is to have the flange milled. Even though I built a jig to hold the flange flat while it was welded some warping was unavoidable. For $30 my local machine shop milled it flat for me much in the same way you would mill an aluminum head.


Like any custom race part some fabrication work after the build is needed to install it. The upper IC pipe needed to be redone as well as the vacuum hoses. I drilled and tapped the end of the manifold to mount the transistor to and since I use COP there was no need to mount the coil. I was worried that the 6” runners and box design would interfere with the brake master but it cleared with no issues.

Performance wise I am very happy with the results. The car pulls strong from top to bottom and it will make power all the way to 8500rpms for me. I do not have the $$ to put this on a dyno so I have no real numbers to back my claims. I can tell you that last year I could hardly keep up with the GT3’s on the back strait at Watkins Glen and with this manifold I can match both the GT3 and GT2 all the way to the buss stop.

The best dyno I could come up with is a friend with DSM link. The test car was a 2G with a 2.4L stroked motor running a BT28 turbo. Needless to say this car is a purpose built autocross monster. He agreed to do his best scientifically wise but everyone should know these are not very scientific results. To start with he did some baseline tests with the stock 2G manifold. Then he installed my manifold and went out and collected more data. It should be noted that this is a 1G manifold on a car with a 2G head so a adaptor plate had to be made so the bigger 2G ports would seal on the smaller ports of the 2G head. It should also be noted that there was no way around the ¾” lip at the base of the junction between the head and the manifold. Everyone can agree that for flow reasons this is not the best idea.

The results were very good. Drivability was not hurt and the car made more power throughout the entire rpm range. The car made 28hp and 35 lbs/ft more power switching to my manifold.

Lessoned learned-

This was cheap. The parts to build were less than $50 and the welding and milling were under $100 so the total cost was less than $150. The whole project took almost three months of fooling around before it was finished. If I had to I could knock the parts out in a day and have it milled and welded the next day. Also, the biggest time killer was the flange. A band saw would speed things up some but in honestly if I could have bought a CNC'd flange for $50 I would do it.

Its been on my car for the whole summer and personally I don't miss the cyclone one bit. I've had it off the car a few times to inspect it and so far so good with cracks. Now go build you own.
Last edited:


Well-known member
May 15, 2005
Rochester, New York

The first plot is from the OEM 2G manifold while the second is with my manifold.


Well-known member
May 15, 2005
Rochester, New York
It's hard to believe that its been nearly 4 years since I posted this thread and its almost to 20,000 views. If you type "how to build an intake manifold" or some similar combination into Google, this thread is at or near the top of the results. Since its been so long I thought I would post an update.

First I have to say that the price of aluminum has gone through the roof. When I first posted this it was much, much cheaper. The number one thing I get asked is "How did you build it so cheap?" Well, I couldn't duplicate that price if I tried. The real big ticket is the 1/2" plate. If you can find some in a scrap metal store or other salvage type operation you could save yourself a lot of money.

This manifold is durable. I have thousands of miles both on the street and on the track and this manifold, both design and performance wise, have been flawless. And its been tested! This exact manifold has been on the 2009 and 2010 RochesterDSM One Lap VR4 competing in the Cannonball Run One Lap of America race. The most grueling test of man and machine we have in the States. But after 4 years of abuse I finally ran into some durability issues that I think expose a minor flaw in the design. When I crushed down the round tube to form the runners I was not able to get the exact oval shape of the ports so I just welded it up and ported it. That worked fine for 3 1/2 years but this spring I found cracks where I had thinned the weld down too much. Seeing how I now have my own TIG welder I just re-welded it with lots and lots of fill and then gently ported it again. This cost about $20 in TIG gas and fill rod along with $50 for the local machine to mill the flange flat after I warped it from welding.

I also get a lot of emails regarding how I made the transition between the runners and the intake flange so here you go.

This manifold performs. After 4-years I have had endless chances to put this to the test with both big and small turbo's, autocross, road racing, and drag racing and it's done everything I could have asked of it. In 2009 it made 355whp on Street Tuned Motorsports Mustang dyno (24psi with a FP3052). In 2010 with the same set-up it made 408whp at around 31psi and when it was de-tuned a bit for the actual race it made 385hp at 27psi. With an old-school Spearco 2-216 core fMIC and the FP3052 (GT30R) it could make 23-24psi by 3800 rpm. Keep in mind that we made 408whp with a bunch of knock up top and a blown head gasket. If we had spent more time tuning I would guess we could get somewhere in the neighborhood of 435whp. Not too shabby for a intake build with a wood router and a table saw!

Now go out and build your own!
Support Vendors who Support the GVR-4 Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications RixRacing RockAuto RTM Racing STM Tuned

Recent Forum Posts