Jump to content
HybridZ
RTz

Scratchbuilt L6 EFI Intake Manifold

Recommended Posts

One of the current projects is a pair of L6 intake manifolds. One of them is for Clint Barnts and his beautiful 280z. The other is for an anonymous 240z owner...

.

Man2.jpg

.

Man3.jpg

.

Man1.jpg

.

The completed plenum shape will resemble this red outline...

.

Man4.jpg

.

Runner ID is 1.5”.

Length, for this manifold, is 6”.

 

After testing, a decision will be made on lengthening/shortening them on the second manifold.

 

A Wolf3D EMS will be installed along with a 6-coil Denso ignition system...

.

WolfV4.jpg

.

coils.jpg

.

There is still a large portion of work to be done, along with a couple of necessary adjustments.

 

I’m expecting to test late March or April. Pending favorable test results and enough interest, I may build a small batch of them.

 

I’ll try to update the progression periodically.

Share this post


Link to post
Share on other sites

This manifold is heavily influenced by its throttle body.

 

A scenario I wanted to avoid....

 

Picture9.jpg

 

...When a TB is attached to a 'flat plate' the incoming air has a tendency to separate directly after the TB, leading to turbulence (which can lead to uneven cylinder distribution).

 

My decided approach was to use an oversized TB to attack this two fold. First by slowing the air down, and second by allowing the walls of the TB to closely match the inner walls of the plenum.

 

Using a 65mm Ford TB (‘95 Crown Victoria) I was able to keep the taper after the TB within 10 degrees...

 

AirFlow.jpg

.

FordTB.jpg

 

 

Another 'feature' I wanted to avoid is protruding velocity stack's.

 

As demonstrated by "turbobluestreak" in this CFD....

 

Picture8.jpg

 

...there is a high pressure area that develops between the TB and cylinder one.

 

Certainly this does not happen with all internal velocity-stacked manifold’s. My point is simply... without laborious testing and expensive test equipment, it can be risky.

Share this post


Link to post
Share on other sites

Prior to designing this manifold, I consulted with a few people. One of them, not surprisingly, was Braap. I asked him to list the items he’d like to see addressed. Near the top of his list was raised ports.

 

I recently contracted Braap's services to properly blend the ports into the larger, raised runners, and also asked him to summarize his work....

 

 

Today after church I was able to focus a little effort on matching up the custom N-42 head to the intake manifold. When Ron started the design phase of this intake, he coordinated with me concerning specific details in an effort to meet his goal of building an intake that would be a definite performance improvement over stock taking into account several details that we both feel need to be addressed in a true performance induction system. One of which is an effort to straighten the port. To do that, we remove material from the roof of the port opening, leaving the floor alone. With that, Ron designed this intake manifold with a raised port centerline. This aligned the runner and port floors with each other so when port matching, material will be removed from the roof of the port effectively straightening the port a little aiding in air flow not only in port size but in port shape AND approach.

In these pics you can clearly see the concept and end result.

 

The manifold runners and plenum base mocked up on the head in the engine bay…

Intake1Medium-2.jpg

 

 

 

These next two shots show the port mismatch and centerline bias of the port. In the top shot, note the port offset bias… in the lower shot you can see the virgin port with scribe line on the left and the roughed in port on the right.

 

 

Mis-match1Small.jpg

.

.

.

Mis-match2Medium.jpg

 

 

 

This would probably be a good time to mention runner size....

 

1) N42 manifold runner area is approximately 53% of the intake valve area (stock 280z).

 

2) In my observations, OE manifold equipped Z’s generally run out of breath shortly after 5500 rpm or so, depending on the build.

 

3) I’ve seen enough data to suggest that number 1 and number 2 are related.

 

4) A priority of the design was to build a modular manifold so that runner diameter and length could easily be tailored to the specific engine.

 

The manifold pictured above uses 1.5” ID runners. This puts runner area at 74% of valve area. The runners are also 6” long (OE averages around 7.25”). This should compliment the intended use of the first engine. Dyno testing will follow to confirm success or failure.

 

We’re looking into building adjustable length runners for the second manifold and, of course, we’ll share the results.

Share this post


Link to post
Share on other sites

The good news is that the manifold is fully welded and fits pretty well.

 

The bad news is that the plenum warped a bit. Next plenum will be of a heavier gauge... lesson learned.

 

Aside from having to fix my screw-up, its well on its way to being tested...

 

WeldedManifold.jpg

Share this post


Link to post
Share on other sites

Several people offered advice for fixing the warped plenum (thanks!). A close friend of mine had an unusual solution... dry ice.

 

The warp is caused primarily from the expansion of the material with no place to go. Dry ice contracts the metal considerably, and nearly instantly, I found.

 

A video illustrating the rough idea... http://www.youtube.com/watch?v=4rIKo6241vg

 

To aid the process I capped and pressurized the manifold to 10 PSI to help the 'valleys' come up. This seemed to help a little but probably wasn't entirely necessary (air was off for the 'peaks')

 

The larger 'dents' towards the center came out very readily. The wrinkles at or near the welds took substantially more effort and I wasn't able to get them out completely. I did find though, through a tip from Roostmonkey, that a large portion of the stress is in the weld itself and, oddly enough, using ice on the weld helped a surprising amount.

 

Although not perfect, here are results...

 

Before...

 

BarntsManifoldWarped.jpg

 

 

After...

 

Manifold_pressurized.jpg

 

 

Although I wouldn't do this on a painted surface, I would recommend giving it a try. Pretty slick.

 

Some clean-up work and its off to be ceramic coated.

Share this post


Link to post
Share on other sites

Picked up the manifold from Finish Line Coatings in Milwuakie this morning, Cerma-chromed...

 

 

ManA.jpg

.

.

.

ManB.jpg

.

.

.

 

ManC.jpg

 

 

 

Manifold is now mechanically finished. There are several revisions in the works for manifold #2. What's that saying?... Experience is that thing you receive immediately after you needed it.

 

The car is scheduled for a dyno session immediately following the All-Datsun Canby Meet... I'll post before/after results with a few more specifics.

Share this post


Link to post
Share on other sites

Fulfilling requests for more info on the EMS system...

 

Glovebox mounted Wolf3D (V4) provides the engine management, driving the coils through a J&S Safeguard. The Safeguard uses 'smart knock' technology to hold each cylinder below knock threshold, delivering a quasi-custom timing map on a cylinder-to-cylinder basis.

 

BarntsWolf_JS.jpg

 

 

 

 

Denso ignition system...

 

BarntsCoilsB.jpg

.

.

BarntsCoilsA.jpg

 

 

 

CAS sensor, pried from a Z32, and adapted to fit the L6....

 

BarntsCAS.jpg

Share this post


Link to post
Share on other sites

Fired the car up last weekend. After some initial tuning, the car runs strong. With a mild Electramotive cam, 265 @ .465, the car pulls cleany through 7000rpm. A bonus is that low rpm performance is only slightly compromised. Dyno results in a couple weeks...

 

running.jpg

Share this post


Link to post
Share on other sites

"Turbobluestreak" was kind enough to produce some CFD work on this manifold (Thanks!).

 

One of the goals of the design was to reduce the turbulence immediately after the TB. Looks like success...

 

CFDa.jpg

 

 

Another high priority was equal cylinder distribution. *Almost* accomplished that. The generous short side radius was an effort to promote this. Seems I still ended up with a small stagnant area in the corner. Also, the unexpected detachment on the outboard perimeter wall seems to cause enough of a low pressure to effect cylinder's 3 & 4 to some degree. The consequences of the reversion in the low pressure zone I suspect to be pretty negligible... the least of my concern's anyhow.....

 

 

CFDb.jpg

 

 

I found this to be pretty enlightening. However, I must add that static testing like this is a serious over-simplification, completely disregarding Hemholtz, acoustics, valve events, etc... Interesting nonetheless.

 

In response to this, I've made some plenum revisions and TBS has offered to test them...

Share this post


Link to post
Share on other sites

Finalizing a revised plenum shape and construction method...

 

PlenumRevision.jpg

 

Compared to...

 

barntsmanifoldb.jpg

 

 

The overall footprint is largely the same. A slightly revised TB position and angle, coupled with a different build technique, will allow a better entry into cyl. 1 and *should* help reduce the high pressure 'ahead' of 3 and 4.

 

The strategy is to form the upper and lower plenum halves around an aluminum CNC'd buck, putting the only weld seem on the horizontal perimeter.

 

Stepping up to .125" material, reducing the flat surface area, and better weld placement will allow this manifold to take boost.

Share this post


Link to post
Share on other sites

Saturday, Braap, myself, and the owner (Clint Barnts) met at TJ Wongs Performance in Vancouver, WA for some dyno tuning (Mustang Dyno). After putting in a fair amount of windshield time in this car, I was a bit surprised by the numbers. Car runs exceptionally well... a bit contradictory to the peak numbers.

 

Cutting through the crap, peak wheel hp was [email protected] with peak torque [email protected] rpm....

 

BarntsDyno.jpg

 

Looking at the average power, it torque peak's at 5400rpm and still makes far more low end grunt than stock. That partly reflects the cars broad powerband and demeanor. Broad, streetable power was at the top of the list.

 

You'll notice a couple dips in the torque curve around 2800 and 4600rpm. The AFR's went a bit rich in the same regions. This implies that airflow is down at these points. In a nutshell, I believe this is a function of mismatched parts. Let me clarify... the headwork is pretty extensive (built by Braap), with little left for improvement, outside welding up the chambers. The intake manifold is also very aggressively sized, with liberal runners. Prior to getting it up and running, I had calculated the manifold with a torque peak around 5500rpm (which is almost exactly what we got). The cam, an Electramotive L7 grind ([email protected]",) is far more docile. The headers are the larger 1 5/8" primaries, enough to support well over 300hp... attached to a smallish 2.25" exhaust system. These compromises were made to keep the car tractable enough for an everyday commute, but produce enough power to be fun when the 'itch needs scratching'. From that perspective, it was successful. However, I am confident the right cam and exhaust would amplify the peak number's substantially. The sacrifice would be drivability.

 

Some side notes...

 

Some things came to light from this session, particularly regarding heat. Not surprising is the location of the air filter, next to the radiator. It was a warm day to begin with, around 85 degrees. I was recording 160 deg. intake temps. Ouch. My best guess is that this cost us close to 7 or 8 hp. If I'm right, we would have seen ~180whp. Not bad, but still about 15 or 20hp less than I was betting on. The owner was a contributing factor in the location of the air filter. After seeing the results empirically.... lets just say, before day's end, a resolution was in the works.

 

The one that did surprised me a little was water temps, compression ratio, and detonation. First, the compression ratio of this engine is modest, at about 8.1 to 8.2. Never expected any detonation. Running a 160 deg. thermostat, we never got any when the water temps were 'normal'. However, even with premium fuel and low compression, we experienced SEVERE detonation once the water temps reached about 225 deg. Obviously, dialing back the timing, based on temp, 'solves' the problem, but that's not really the point. Backing into it.... we got the best numbers at 160 deg. Every consecutive run, brought temps up... and power down. Let the car cool down and the power reappeared. When we pushed much beyond 210 deg., we would get detonation, progressively worse with temp.

 

At 160 degrees, the car made the same power with total timing anywhere between 37 and 43 degrees. Running less than 37 cost power. With temps above 200, 40 degrees caused a power reduction.

 

This, in itself, is not all that surprising... the surprise is the knock propensity at such modest comp. ratios. How about all those L6's, running 10+ CR, with lackluster cooling systems? My point is this... don't discount a cool T-stat and a cooling system to back it up.... the car is receiving a Ron Davis radiator and an upgraded fan.

 

The other related conclusion is that this is inspiring me to install a cylinder head temp sensor near cylinder 5, as this appears to be directly related to chamber temperature. Wolf will allow additional sensors to manipulate the mapping... I will be taking advantage of that, even if for no other purpose than data logging... I think it will be interesting to see the correlation between cyl. head and water temp sensors.

 

In the end, this car is proof that a dyno sheet doesn't tell the whole story.

 

The question for me at this point is runner size. I'm of the opinion that a slightly smaller runner, say 1.4" to 1.450", wouldn't have cost us any peak power, but could in fact, bolster the low end (and quite possibly smooth out the torque curve). I need to make a decision shortly.

Share this post


Link to post
Share on other sites

We took the off-season opportunity to repair the warped plenum.

 

Cut the old top off, weld on a new .125" top, and re-coat. Nothing to it :wink:

 

Turned out spiffy methinks...

 

 

FixedManiC.jpg

 

 

 

 

FixedManiB.jpg

 

 

 

 

FixedManiA.jpg

Share this post


Link to post
Share on other sites

Monzter was kind enough to run a CFD analysis on the proposed plenum revision. The goal was to improve three areas... entry into #1, 3 and 4, while reducing the swirling mid-plenum. It seams reducing the swirling was the only item that became 'reality'... it was also of my least concern.

 

 

Velocity chart at 20psi...

 

 

20psi_velocity.jpg

 

 

 

Thanks again Monzter!

Share this post


Link to post
Share on other sites

Monzter was able to calculate individual runner CFM.

 

1. 129

2. 170

3. 156

4. 141

5. 136

6. 138

 

These correlate well with the pictorial view in the previous post.

 

A runner-runner disparity of nearly 32%. Ouch.

 

Here's the crux... these are static flow numbers and an intake is anything but static. Valve events and acoustics play a significant role in manifold function.

 

For example, a friend of mine was relaying a story of intercooler flow testing, using an inlet 90 degrees to the core such as below....

 

Backflow.jpg

 

 

They found that the air passing by the first row's actually created enough negative to cause backflow through the core. The CFD on my manifold shows a tendency to do this as well.

 

However, a large number of 'successful' manifolds are built with a 90deg. entry, both OE and aftermarket (Ferrari, Porsche, Lingenfelter, etc, etc). That, in itself, isn't definitive proof, but it does build suspicion that static flow is only a small part of the picture... in fact, it may be so skewed as to be nearly useless.

 

Maybe some testing would shed some light...

Share this post


Link to post
Share on other sites

After a heap of procrastination, I decided to have a go at a carbon fiber plenum...

 

 

CFPlenumRearMini.jpg

 

 

CF has a number of attractive qualities that make it a nearly ideal candidate for manifold use. Aside from the obvious weight and thermal advantages, it has a nearly unlimited fatigue life. Fatigue resistance is one of my primary motivations as a boosted plenum is subject to significant stress. A well built CF plenum should survive many lifetimes. Add in nearly infinite shaping possibilities and an audible resonance that can't be duplicated by any other material.

 

Of course CF comes with its own drawbacks, mainly cost, manufacturing complexities, and a limited heat range.

 

Anyhow, the first step is to build a plug that mimics the final shape. A mold resembling the 'silhouette' was constructed, filled with 8lb expanding foam, and the plug extracted...

 

 

1.jpg

 

2.jpg

 

3.jpg

 

4.jpg

 

5.jpg

 

6.jpg

 

 

 

Next up, hand shaping...

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×