Manning

That is incredibly admirable of you George.

Geo

Yeah, the $1,100 is a lot. But that is what a set of cams and a header cost for the Sentra SE-R and they sell pretty well. Not all Japanese cars have an over abundance of cheap parts like Honduhs. The biggest difference between the SE-R and the 944 when it comes to these parts is the SE-R parts are extremely well known quantities as far as power production.

DanG

You're right Manning, the flow is likely turbulent throughout the entire intake system, except maybe the expansion chambers like the stock airbox and possibly the plenum. But its not the turns or bends that determine whether the flow is laminar or turbulent.

[Fluid Mechanics 101][/Fluid Mechanics 101]

I started working at a major Tier 1 supplier's air induction systems core engineering department in January. I've held off on typing stuff to this thread cause I'm most often browsing the forums during work and I don't know who's watching I can add some more tidbits in another couple of days.

DanG

Oooo! Resonance! That's even more fun than calculating resistance.

Short answer is you're both right. Yes, a big portion of the positive gains due to resonance tuning are controlled by the individual intake runner characteristics. Good thing is, Porsche knew this. The diameter/length of our runners are probably darn close to optimal for the stock torque profile. Sure you could shift the curve around with different lengths, but any gains at one RPM should be offset by losses at another. And to get optimum lengths throughout the rev range requires variable length runners. (VLIM link)

Geo's notion that pre-manifold plenum resonance tuning can be significant is right. This was most likely NOT optimized by Porsche, or any other OEM, as they have a lot more things to worry about than one of us, such as cost, packaging, warranty... you get the idea. Until you reach the throttle body, I imagine some compromises were made. Where we can put in a nice mandrel bent steel system with silicone tube joints and constant tension clamps, OEM's are stuck with rubber and plastic in easy/cheap to manufacture shapes and sizes with the minimal number of parts.

While most of what I look at for my job is noise related, the same acoustic reasoning can be used with power improvements in mind.

I haven't had the time to run the 944's intake through my company's flow and resonance tuning software, but I should be able to get around to it soon. It should tell us where the major sources of restriction are, and what we can do to fix them.

Matt O.

Pretty cool, however, don't hit a large puddle.

I think a better place to spend your time routing it would be under the nose panel like on an S2. Then cut holes in the nose. 100% "cold air."

Campeck

again. i dont care about cold air.

and under the nose panel isnt getting me anymore length on the pipe.

also. with this filter Ive noticed a VERY LARGE hesitation when first accelerating in the car for the first 30 sec of running. after that it goes away.

you can be cruising in third gear....punch it and the car SLOWS doiwn like crazy..then goes really fast.
again this is only for like a minute after the car has been started.

Campeck

also.
our digital camera just broke.
so I can't get pics of the mounted fiberglass bottom cover.

FSAEracer03

Oh, isn't it

But are you trying to say that pressure waves don't care that there is a massive "barn-door" AFM between the source and intake? The pressure waves getting through the AFM are completely muffled, mixed and reduced. This can be seen by pressure loss from the manifold to the intake piping. Another can be a more accurate digital pressure gauge at either point showing a near-steady pressure beyond the AFM (WRT the motor). I'd argue the gains from intake repiping is from the turbulence reduction. Engines love turbulence in the chambers and want the exact opposite beyond the valves. Smooth piping can do a lot to help the stock J-boot and box's rough interiors.

Big words... maybe I should get my nose out of the books *cough* High-Performance Piston Engines class *cough* and get into the real world *cough* I design racecars for my school and work on Porsches at a specialty shop in northern VA to support my expensive habits*cough*

That's good advice.

See above for my take on intake power gains.



Thanks... I was lost.

944CS

OT......fsaeracer...look for me at this years competition...i'm on drexels squad

Manning

I was of the impression that the barn door on these cars pretty much swings out of the way as significant air flows and is more of a switch than a variable meter.

DanG

OT again: I'll be in Pontiac this year (I live about 10 miles south of the Silverdome) with the UMich-Dearborn team. Its our school's first car in a couple years. B&S 570cc V-Twin. Thats right, a Briggs and friggin Stratton. Don't laugh until you feel the torque.

Back on topic:

I wasn't discounting any ill effects on resonance from the barndoor. To be honest, I have no idea how the AFM actually functions (ie: is it nearly wide open at higher flow levels or not), so again, no idea on its effect. The simple answer would be go to a MAFS or MAP sensor and eliminate that from the equation.

Again, its the reduction in separation that's going to help, not turbulence. I guarantee the flow is 100% turbulent from the moment it enters your intake system. Flow rates are high enough and diameter is large enough that it must be turbulent. It wouldn't matter if the intake was a perfectly strait and polished tube. It'll be turbulent flow through the entire length at this velocity and diameter.

Does the AFM ruin any chance of pre-manifold resonace tuning? I dunno. Does stock intake plumbing have room for improvement? Yes. Which is more important? Not enough info yet.

Geo

For the record (and stated previously in this thread), I do think that it's possible the barn door could damp the effects. However, there will be a resonance between the open end of the tube (yes it has a filter, but it will still be a point of reflection) and the AFM. Gains should still be possible even if the barn door damps them some.

TRWright

I only read the first page of this thread, so I am sorry if this has already been said, but I would suggest getting a water sock to cover the filter. Dinan makes a nice one. Another route would be a valve that I know AEM makes that won't let you suck up any water.

Geo

My point Kevin, is you can think about things in theory all you like. But sometimes theories don't hold up in the real world. That holds true whether you are an engineer, an accountant, a ditch digger, or whatever. In this case, people already have found significant and repeatable gains from CAIs with longer tubing. So any theory saying it won't work has a missing hole somewhere.

Danno

Kevin, there are effects in real-world conditions that are not cleanly modeled in ideal text-book engines. What are the equations governing flow around an AFM? What effect does the angle of the flapper-door have? I'm sure it has a significant effect as with the door in the way, say at partial-throttle, you've got a dual-chamber resonance, and with the flapper fully open, you're looking a a single long chamber. How have you modeled that and come up with some accurate calculations that actually fit real-world conditions to verify that your model is actually accurate? In a lot of cases, real-world examples have too many factors and variables to accurately model, you just have to do emperical testing and examine the results. A lot of times, the results may be unexpected. but at least you've compiled a set of data of cause & effect and can just use that without really being able to explain why it happens.

"I was of the impression that the barn door on these cars pretty much swings out of the way as significant air flows and is more of a switch than a variable meter."

Again, an example of real-world conditions not fitting wtith textbook models. It usually takes less than 125-175gm of force to open that flapper door (how many of you have actually measured this, eh?) No AFM I've ever tested has required more than 200gm to fully open. However, this was force tangent to the swing of the door and not parallel to actual flow-path, so I'm not sure what the actual relationship would be. I suspect we have an asymptotic function with force rising exponentially as the door becomes more parallel to the direction of flow.

However, we don't have to muck with the numbers and try to come up with mathematical models to explain this. There's plenty of test data to show that removing the AFM from a 951 and replacing with MAF/MAP sensing will automatically give you +30-50hp at exactly the same boost-level as before. What usually happens is boost will increase by 2-3psi when you remove the AFM, and you have to turn the controller down to the same boost as before. Then the +30-50hp increase is more accurately compared to the control, rather than the +50-70hp increase with swapping out the AFM alone.