When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
When the exhaust gas cools its density goes up, which means velocity goes down unless you are stepping down the pipe size, which means that for the same loss per foot of pipe you can use a smaller diameter pipe. The maximum pipe size is required at the point where the wastegate circuit merges and the banks are still separate. After pulses get combined between banks, the required cross-sectional area goes down and then as the exhaust gas cools it goes down further. All the high powered factory exhaust systems nowadays have huge cross sectional areas around the cats and then the pipe size is stepped down pretty dramatically towards the tail pipe muffler, before the mostly visual big tailpipe tip.
Although I think 4” at the tail pipe would be enough, I think 4.5” is good for experimentation as we can use inserts in that tail pipe to reduce the diameter and see how that impacts power and sound.
The visual big tailpipe is mostly for bragging.
Åke
All the high powered factory exhaust systems nowadays have huge cross sectional areas around the cats and then the pipe size is stepped down pretty dramatically towards the tail pipe muffler, before the mostly visual big tailpipe tip.
i was wondering about this the other day...
how it might make sense to use larger piping for most of the run (large pipe = lower restriction/pressure drop due to length) and then step down at the muffler/tail pipe.
i like the idea of stepping up in size at restrictions (cats) too...it makes sense to me, am i on the right track here?
also, data point for your brain to chew on...
the Dodge HellCat cars (707hp supercharged 6.2L) from the factory come with dual 2.75" pipes with an X-pipe about mid-point on the car. looks like they just have 1 cat per bank, mounted between the manifolds and the X pipe.
dual 2.75s have about the same CSA as a single 4" (a little less actually).
granted, people unlock nutso-power when fitting a 3" system+cat delete but those have to get a tune at the same time so hard to know how much gain is from which change.
I think we're on the same page in terms of the general size requirements.
In terms of nit-picking, I don't like that sorts of charts, I think they are oversimplifications. What's the loss accepted by that curve? David Vizard I think uses 1% hp loss and says that after the cross-over that combines pulses 3.5" OD exhaust pipe does 500hp. And where is the diameter measured? For my 928, I am computing that a 3.5" OD pipe close to the cross-over flows the same as a 3" pipe at the rear bumper because of the cooling effect.
Many people like short, big runners for a turbo car. We’re thinking about trying another direction. That is, modify the stock manifold to flow as well as possible while tuning it to produce maximum torque at rpms right below where the turbos have spooled to reach the knock limit.
Åke modified a couple of throttle body elements for me for experimenting for that purpose. The better flow capacity will help the top end and then we’ll try a couple of tricks to move the resonance tuning peak for the rpm range that most needs it.
ive got a theory on runner length and boost...
in the 951 world the stock intake manifold is very good for 99% of street cars, past 500hp.
however the subset who really want that extra rush of power 5500+ rpm have got large plenum short runner intakes (a few options on market, plus custom ones).
these intakes have runner lengths (including the head port) way too short to have any meaningful resonance tuning (8.25" should work in the 9-10000rpm range for 3rd harmonic)
i'm thinking the short vs long thing is twofold:
1) simple pipe friction, cutting the length in half with everything else constant cuts the flow drag (pressure loss) in half too. minor, but an extra 0.25psi that makes it to the valve can be 5-10hp depending on the engine.
2) the long runners on a stock intake that are tuned for say 3000rpm reach an "anti-resonance" point at 4500-6000rpm. instead of the air's momentum compressing the air column as it goes towards the valve, the air is being expanded away from the valve (say the plenum has 20psi but at the intake valve there might only be 17 psi due to expansion), similar in concept to the "by design" 997GT2 intake manifold.
part 3) is that short runner plenums tend to just look cool in magazines so that's why all the kids favor them.
so if going to the trouble of making a custom intake for a typical 1000-6000rpm engine, i think the answer is make the runners as long as you can reasonably fit, with reasonably straight, smooth-profiled runners with a built-in taper. no 944 or 928 with stock oiling system will live long in the high 6000rpm+ range anyways, so designing something focused on that narrow band is kind of silly unless for a dedicated racecar where you've accepted the risk of the stock oiling system.
or just optimize the OEM intake to keep its desired characteristics (idle quality, mid range for off-boost driving 90% of the time) as you are doing.
ive got a theory on runner length and boost...
in the 951 world the stock intake manifold is very good for 99% of street cars, past 500hp.
however the subset who really want that extra rush of power 5500+ rpm have got large plenum short runner intakes (a few options on market, plus custom ones).
these intakes have runner lengths (including the head port) way too short to have any meaningful resonance tuning (8.25" should work in the 9-10000rpm range for 3rd harmonic)
i'm thinking the short vs long thing is twofold:
1) simple pipe friction, cutting the length in half with everything else constant cuts the flow drag (pressure loss) in half too. minor, but an extra 0.25psi that makes it to the valve can be 5-10hp depending on the engine.
2) the long runners on a stock intake that are tuned for say 3000rpm reach an "anti-resonance" point at 4500-6000rpm. instead of the air's momentum compressing the air column as it goes towards the valve, the air is being expanded away from the valve (say the plenum has 20psi but at the intake valve there might only be 17 psi due to expansion), similar in concept to the "by design" 997GT2 intake manifold.
part 3) is that short runner plenums tend to just look cool in magazines so that's why all the kids favor them.
so if going to the trouble of making a custom intake for a typical 1000-6000rpm engine, i think the answer is make the runners as long as you can reasonably fit, with reasonably straight, smooth-profiled runners with a built-in taper. no 944 or 928 with stock oiling system will live long in the high 6000rpm+ range anyways, so designing something focused on that narrow band is kind of silly unless for a dedicated racecar where you've accepted the risk of the stock oiling system.
or just optimize the OEM intake to keep its desired characteristics (idle quality, mid range for off-boost driving 90% of the time) as you are doing.
Well, over here in Europe tuned intake ducts (3rd harmonic) have been around since at least seventy years, as long as the Weber side draft carbs have been on the market. No need to reinvent the wheel. If people like Mark Kibort could accept a modified hood it would be easy to design a good working intake system with straight runners.
Åke
Most of the work sounds like it's on the "throttle body shoe" pictured above, are you going to tweak the actual S4 intake manifold at all?
Extrude hone, or maybe just smoothing out casting lines inside the runners a little?
Most of the work sounds like it's on the "throttle body shoe" pictured above, are you going to tweak the actual S4 intake manifold at all? Extrude hone, or maybe just smoothing out casting lines inside the runners a little?
The blue engine already has a mildly ported stock intake. That porting work should help a little at high rpms. In the same way as the lower throttle body casting porting work should also help a bit at high rpms.
However, as far as I know, manipulating the resonance tuning effects is the only way to move the 928 intake flappy closed torque peak to lower rpms. There are three ways that I am aware of. First, add plenum spacers. Second, increase primary runner length. Third, increase the secondary pipe (zip tube) length that is the channel that goes thru the lower throttle body element. All these three changes will move the flappy closed torque peak to lower rpm. The idea is to first see where the turbos spool and then shift the flappy closed torque peak to the rpm point where turbos need help.
Plenum volume can be manipulated with spacers or epoxy. The runner and zip tube length can be manipulated to some extent with spacer plates. Those same spacer plates can be used to smooth out the intake to head interface without expanding the cross-sectional area.
i wonder how much the resonant effects would be altered by shaving the "stubs" in either plenum tank down to bare minimum (flush with the casting, incorporating as best a bellmouth as possible).
looks like a couple inches could be lost, side effect of greatly increasing open plenum volume.
For a street car with turbos, one likely needs more not less primary runner length in an S4, to bring down the flappy closed torque peak to slightly lower rpms.