S38 Plenum ?

[M5W]

Hey guys just swapping out my injectors and searching for a vacuum leak and found that if i made my remote mount oil filter setup a reality, i could fit a pretty decent sized plenum in my engine bay. Ive only seen one that was made for the S38 that has shown gains, but even this is debated over on M5Board.

My engine bay as of today.



The debated plenum (claims 430 HP with shrick cams, this plenum and alpha-n tuning) (Plenum is supposedly 2.6 times larger than stock)







He is using the factory velocity stacks, and im thinking they are to small for this plenum or any plenum larger than factory. This would give a good top end because they are short, but it would be gutless down low compared to what it could be. My thoughts are to try using the velocity stacks from an S54 Plenum.

These:





And adapt them to the S38 throttle body housing using an adapter that i designed







Which would bolt right up to the throttle body and allow a half inch of clamping to mount a 2.00 to 2.50 reducing coupler at 45° to mount the S54 Velocity stack.

[pspassos]

Man, if you do it, make two, I'll buy one from you, just give the price... I'm searching for something like this for ages, never found a good one.... seems like you are in the right path, keep on going

[xatlas0]

But then don't you lose the variable runner length system that the stock intake has? That will make your torque will drop considerably on the low end. The Alpha-N style you showed will do better at high RPMs, but in terms of daily drivability, I think it would be considerably worse, given that the stock S38 doesn't exactly wake up until 3.5k RPMs anyway.

[PorscheDude]

I think you will loose allot of low end torque by increasing plenum volume.

[M5W]

The variable runner length system will be lost yes. My theory is that the S54 has the same amount of valves and less displacement yet it produced 262 ft·lbf @ 4900 and the S38 produced 266 ft·lbf @ 4750 without the variable runner length, so BMW must have changed something in the intake manifold to account for this loss of torque. The only real difference is that the Plenum is larger and the velocity stacks are shaped far more aggressively on the S54. I will post pictures tonight after i destroy the M3 Plenum to get them out.

Just so im clear, i dont like the design of the box i showed in my first post. Im just using it as a reference as to why im pursuing this.

[Pooch]

Why must you do this to me?


I want an S38. I'd also want to clean your engine bay up.

[M5W]

lol The engine bay will be gettin cleaned as i go along. Searching every vacuum hose and every port for the leak. I think in found it. I will have side by side comparisons of the velocity stacks as soon as i find a way to take apart the S54 plenum without damaging the stacks inside

[xatlas0]

Originally Posted by M5W

The variable runner length system will be lost yes. My theory is that the S54 has the same amount of valves and less displacement yet it produced 262 ft·lbf @ 4900 and the S38 produced 266 ft·lbf @ 4750 without the variable runner length, so BMW must have changed something in the intake manifold to account for this loss of torque. The only real difference is that the Plenum is larger and the velocity stacks are shaped far more aggressively on the S54. I will post pictures tonight after i destroy the M3 Plenum to get them out.

Just so im clear, i dont like the design of the box i showed in my first post. Im just using it as a reference as to why im pursuing this.

The point of the variable length intake runner system isn't to maximize the peak torque, it is to improve the area under the curve. Larger intake runners provide more low-end torque at the expense of high end horsepower. The inverse is also true. (hence why race cars typically have very short velocity stacks, as they spend almost all their time above 5k RPMs.) You will still get the same peak torque with or without the system, but your overall torque across the rev range will be higher. (the peak torque hits there due to the resonance of the intake manifold volume, head VE, and flame front properties, as well as the standard timing (both ignition and cam) and AFRs) The S54 (as far as I know) doesn't use the variable length runner system because it instead modifies the cam timing, which can somewhat replicate the effect. In the race cars that require a large dynamic effective RPM band, they use both systems for maximum effect. The peak will stay constant, but you will be closer to the peak for a broader rev range.

[M5W]

My variable length runner system is not in place right now since i am running standalone with Alpha-N tuning. I am pretty satisfied with the setup but the intake manifold always seemed like the only place i could improve things. The velocity stacks have the same diameter that the S38 ones have on the end, the entrance is the difference. So by using these i will have good torque but lose my top end ? Also, i am running a 296 intake cam and a straight thru exhaust, would that create more suction thru the intake tract to use the bigger stacks ? Im just throwing around ideas because as of yet no one has been able to make a plenum that makes alot more power. I know its possible. Im just trying to find a way.

[xatlas0]

Originally Posted by M5W

My variable length runner system is not in place right now since i am running standalone with Alpha-N tuning. I am pretty satisfied with the setup but the intake manifold always seemed like the only place i could improve things. The velocity stacks have the same diameter that the S38 ones have on the end, the entrance is the difference. So by using these i will have good torque but lose my top end ? Also, i am running a 296 intake cam and a straight thru exhaust, would that create more suction thru the intake tract to use the bigger stacks ? Im just throwing around ideas because as of yet no one has been able to make a plenum that makes alot more power. I know its possible. Im just trying to find a way.

Hm. Then it really boils down to the goals of the work. The original engineers de-tuned their race M88 (used in the M1 and E9 Gr2 and Gr5 race cars) to make them have a wider useable range. It seems like you are undoing the majority of their work. Back in the late 70s, they got ~450 hp out of the same engine, although they were likely using race gas. The turbo version got ~800 hp. If you look at the old race engines, you will see the way to get maximum peak values, but it comes at the expense of lower performance at non-optimized RPMs.

It sounds like you want peak values more than standard driveability, so the best way to go about that is a flared conical intake manifold, short runners, tubular equal-length headers and a pretty direct exhaust. Your cam and cam timing will also be very, very important.

The velocity stack length determines the RPM at which they will give the best performance. Longer=lower RPM. Shorter=higher RPM. Larger intake manifold volume does not always mean better. The intake manifold volume and shape in a racing situation is determined by the flow streams required by the engine at the desired operating RPM, that way you can get a slight ram effect from the intake manifold air's momentum coupled with a resonsance effect from the combustion causing a pressure wave in the manifold. When these events are timed correctly, you can get as much as a 10-20% increase in intake charge, as both effects will cram more into a cylinder at a given RPM. Outside that RPM, things will be worse, as the optimal runner length and timings required to use these effects change.

There are a multitude of books on intake manifold design. It is arguably one of the most researched aspects of an internal combustion engine, with actual combustion taking the most-researched title. I'd suggest you take a look through some of the papers and you will see what I am talking about.

[93FIM5]

When it comes to losing the variable resonance setup you know my stance, but I for one would like to know what benefits or draw backs the E46 velocity stack design has. Mainly the tear drop shaped inlets compared to the S38 standard trumpet design.

So my suggestion is to first maintain the S38 plenum (if possible) and instead build/have built a set of velocity stacks that mirrors the s54 inlet but maintains the S38 length and diameter. This should allow some sort of comparison between the different velocity stack designs.

Another thing, take me some pics of your filter assembly and pre-plenum air intake setup maybe that has room to be optimized.

[M5W]

Im not trying to make a race car, im just trying to do something new. The carbon airboxes that use the factory stacks make usually 10-15 horsepower more . They are slightly larger. Im looking at trying a new stack design. If it dosent work so be it, but i still think its worth a try seeing as how no one has yet. They are larger, but they have a aggressive taper.

Originally Posted by 93FIM5

When it comes to losing the variable resonance setup you know my stance, but I for one would like to know what benefits or draw backs the E46 velocity stack design has. Mainly the tear drop shaped inlets compared to the S38 standard trumpet design.

So my suggestion is to first maintain the S38 plenum (if possible) and instead build/have built a set of velocity stacks that mirrors the s54 inlet but maintains the S38 length and diameter. This should allow some sort of comparison between the different velocity stack designs.

Another thing, take me some pics of your filter assembly and pre-plenum air intake setup maybe that has room to be optimized.

The S54 inlet is very large. Looking at it i might be able to make something, but is would not be cheap as it would have to be a molded item, like the BMW s38 trumpet. I will take side by side comparisons tonight with dimensions.

[xatlas0]

Originally Posted by M5W

Im not trying to make a race car, im just trying to do something new. The carbon airboxes that use the factory stacks make usually 10-15 horsepower more . They are slightly larger. Im looking at trying a new stack design. If it dosent work so be it, but i still think its worth a try seeing as how no one has yet. They are larger, but they have a aggressive taper.



The S54 inlet is very large. Looking at it i might be able to make something, but is would not be cheap as it would have to be a molded item, like the BMW s38 trumpet. I will take side by side comparisons tonight with dimensions.

BMW already did this with their older race cars. That's why I was suggesting looking there first. Those older manifolds will be the most effecient bar none for the engine at high RPMs. The best option would be to copy the old manifold design, pretty much bolt for bolt. Plus, it saves you the trobule of having to design everything yourself.

Alternatively, you could look at the S54 racing car's intake manifold, as that would have very similar frequency responces to the S38. The newer manifolds will likely be a bit better than the old, so the way you are approaching it certainly isn't a bad plan.

[2DYE4]

How about individual TB cones or stacks??

[M5W]

Originally Posted by xatlas0

BMW already did this with their older race cars. That's why I was suggesting looking there first. Those older manifolds will be the most effecient bar none for the engine at high RPMs. The best option would be to copy the old manifold design, pretty much bolt for bolt. Plus, it saves you the trobule of having to design everything yourself.

Alternatively, you could look at the S54 racing car's intake manifold, as that would have very similar frequency responces to the S38. The newer manifolds will likely be a bit better than the old, so the way you are approaching it certainly isn't a bad plan.

Iv looking into buying the M88 racing manifolds, but they wont fit because of the brake booster i believe. Im just going off of what the seller told me. Id love to get an M88\3 manifold and use that but the only ive been able to track down is in germany on ebay for 790 euros. Thats alot of coin for something that "might fit".

With the oil filter housing gone i will gain about 5 inches of clearance for the plenum.



My engine bay room



Comparison

















Also, the S54 stacks do not fit inside the S38 plenum





My engine bay setup as current







M3's were harmed in this

[xatlas0]

Nice documentation!

The main advantage of those velocity stacks is the increase in the opening area. Since the air is (sorta) an incompressable fluid at the speeds you are talking about, the initial cross-sectional area directly limits the maximal amount of air you can have pass through the individual plenum, given the velocity limitations. So, in effect, like a nozzle, the opening tapers to a smaller circular cross-section, which increases the velocity of the air as it goes through the stack.

Math:
(Q)=AVp
Q=Mass flow rate
A=cross-sectional area of current slice
V=fluid velocity
p=density of fluid

So as you can see, if you've got a set Q (which you have as soon as the air enters the velocity stack) if the area decreases and the density remains constant, you get an increase in velocity of the fluid. In this case, that also causes a larger pressure drop immediately in front of the stack when it is operating flat out, improving the inlet flow characteristics and reducing the drag at the opening.

Overall, this should be better than the original stack. The overall volume of the stack is probably less, as it actually starts at the sorter length, not the full length, as it isn't enclosed at that point. Props to the CFD guys at BMW, it is an elegant solution, from a fluid dynamics standpoint.

If you want to fully maximize your intake manifold performance for the top end, I can see two ways to go: tune it so you get the Helmholtz frequency charge boost, or go for a ram-air style intake internally. The former uses the total plenum volume and the combustion noise (which is actually a pressure wave, remember, noise is a mechanical wave in a fluid) to make the return of the wave to the stack coincide with the open intake valve. The latter method is more like making an individual plenum for each stack then bolting the individual plenums together and calling it a manifold. Think long-tube headers in reverse. The collector of the headers would be where your air filter bolts to. However, the big downside of this is that it will only work really well at the design point.

I have seen a hybrid of the two, which is when you have a single large plenum, but inside are baffle channels to direct the incoming air into each stack when flat out. However, the channels do not go all the way to the stack, and there is a gap to allow mixing of each cylinder's flow streams at non-optimal RPMs. Overall, I'd probably go with one like this, but I'd also strongly suggest doing more research into intake manifold design before really building much. You don't want to have to build the think many more times, right?

Oh, if you want another good example of a "tuned" intake, look at HPF's turbo S54 manifold. It won't be as good for a NA application, but it certainly wouldn't hurt to ask them about it and how they designed it.

[M5W]

HPF took out the velocity stacks in there manifold didnt they ? I watched a youtube video of them showing it. Also, my idea was to use the factory size panel filter and weld in a bracket for it , them have a cover that would clip over it much like the CSL box, and have that cover hooked up to my highbeam 3.5 diameter velocity stack. I still have alot of figuring out to do, and i have to see how to make this bolt up to my throttle bodies. I got a quote for my adapters, still to much but im working on getting it down.

[rodgers]

verrry interesting thread. M5W, have you dyno'ed your car yet the way it is now w/MS?

[Elekta]

have you thrown this up on the M5board wall? I bet you'd get some decent analysis over there

[DefactoM6]

Yeah, I'm rather curious to know what you laid down as well. Is your head just a MM race head on a stock B36 bottom end? Interesting thread, I'd be curious to see how much of a bottleneck the intake plenum is.

[M5W]

Ive tried joining M5Board twice. They just dont want me lol. I get a username and password but it never works. I will be dyno testing this summer. I was going to last summer but my crank hub key sheered. Didnt bend any valves, but refreshed the bottom end and built the top end. The throttle bodies ported, and i put in a straight pipe this past december (check out my exhaust vid at the bottom of my sig) Running Alpha-N and wasted spark. With my tune from before i went inside the motor with a factory setup and just exhaust and Alpha-N I was sprinting away from Charger SRT8's, and that was from a dig! I cant wait to see what this year brings. Im hoping for close to 325 whp. I think its attainable.

[///Manuel]

I must have missed something but what engine management are you currently using and how much performance did you get out of it?

[Binjammin]

Originally Posted by DefactoM6

Yeah, I'm rather curious to know what you laid down as well. Is your head just a MM race head on a stock B36 bottom end? Interesting thread, I'd be curious to see how much of a bottleneck the intake plenum is.

Look at you all slumming it in the e34 land! How's the car, done done or any loose ends?

[M5W]

Originally Posted by Binjammin

Look at you all slumming it in the e34 land! How's the car, done done or any loose ends?

Car is as you see it in the first pic of the post. Got it home, then went on vacation, then came back and started modding again. Right now trying to figure out the best way to redo that oil catch can at the bottom of my manifold. I think BMW over did the size of hose, its 1" in diameter. i have an idea, and im pretty much gunna re-make the little bracket that bolts to the valve cover to fit a 1/4 NPT fitting and run a 3/8 hose to an air compressor filter, and use the port on the bottom of the ITB's to use as suction for it. The bung was used for the fuel tank ventilation filter in the front of the car underneath the power steering reservoir. Any thoughts ?

[DefactoM6]

Originally Posted by Binjammin

Look at you all slumming it in the e34 land! How's the car, done done or any loose ends?

Mega loose ends. After all that time down restoring it and stuff, going to have it tuned, i find out that I was handed a bogus dyno sheet 2 years ago, that they tuned it to mask a very large issue with the motor, and that the car won't make anything more than 255whp. I had originally thought i was making 290whp, and thought that I should be able to hit 325 easily because I thought had gigantic chokes in other parts of the system...nope, something was wrong with the motor. Mind you, this is a motor that has barely 2k on it. And when I confronted MM about it AGAIN (I had confronted them about really weird running issues and very very low power when i first got the motor in '05 and had it running on a Jim C chip, which they said it would run well), and surprise surprise, after spending over 20 hours in total on the phone, I can't even get Jim Rowe to acknowledge the possibility that he might have gotten something wrong with my motor...the chief suspect is cam timing. He has the nerve to suggest that my engine management system, which was double checked by 3 very well known specialists, including the chief engineer at Electromotive, was the cause of all this, and that I needed to buy his chip and MAF setup to cure all my ills. Not to mention the fact that he expected me to handle the costs of specifically diagnosing and solving issues directly stemming from his incompetence, since the motor has remained untouched since I received it.

I just decided that it simply wasn't worth my time to keep fighting with him, trying to shove logic down an illogical donkey's throat, and to cut my losses and find a way out of the problem that didn't involve MM. I know that it's going to cost me MORE money, but if it means sparing myself the pain of having to fight with Jim, for results that might never materialize, so be it...I have better things to do with my time than fight with him.

Based on all of the symptoms, Mario@VSR recommended that I get rid of the MM head and cam setup, since the surface turbulence stuff is BS anyway (I simply didn't know any better when I bought the motor), keep the built bottom end, buy a stock head, and get some proper cams to run it on. This seemed like a reasonable solution, but at that point (early September) with another season wasted after much anticipation, I was already fed up with the car, and couldn't bear looking at it, let alone figuring out a way to get it fixed.

I finally started to emerge from my funk last week, and I acquired a clean S38B35 head, cam box and lifters off an '88 M5 that posted 200, 205, 200, 205, 200, 200 on compression before being pulled off the block, no cracks, burnt valves, and has never been machined. It does have minor warpage (it's off by 4/1000"), but this would be fine if I want to simply slap the head on the bottom end currently in the car. I currently find myself with the difficult decision of whether or not I should just get over myself, put the head on as is with a decent cam combo, and deal with the lower power and choke on an otherwise hotted-up motor, or if I should venture down the road of modding yet once again, and have the head aggressively PnP'd, with bigger valves, etc, built to rev hiiiigh, with a set of aggressive cams. If I'm having headwork done, I'm going to send it to someone in the area who has an awesome reputation; I'm not sending it halfway across the country to a BMW "specialist" to get headwork done at 2x the cost and 2x the hassle, without any guarantee of being a superior product. I know that the S38 is more complex than an SBC, but headwork is science, not practice. I think the whole online mentality of "OMGZ NOE U NED SEND 2 BMW BILDER OR ELSE U HAZ BAD MOTORZ!" is a bunch of bs. I tried that once already, and look where it got me.

Anyway, that's my newest story of heartache. I do sometimes wonder if I'll ever be able to actually drive the car. Jeez, my fears of becoming Carmen are becoming more and more real. Input from the audience is appreciated, since I'm still undecided as far as whether or not I will get the head built, since this will also determine what cams I get. Weigh in, everyone!


PS, Here's the heartache, whoops, I mean car, for reference:

YouTube

YouTube

M5W, your car sounds great, how high are you winding it out in that vid?