Gain 100HP with an intake manifold change?? - Cross post from Ferrari Chat
#781
Captain Obvious
Super User
Super User
#782
Archive Gatekeeper
Rennlist Member
Rennlist Member
I could swear I saw one like that on Facebook recently, painted (PC'ed?) red. Colin's handiwork, maybe? Can't find it.
#783
Rennlist Member
#785
Pro
Upfix --
I've got a couple of questions about your modifications.
First, did you by any chance fix anything, like a knock sensor or Hall sensor, between the before and after dyno runs?
The reason why I am asking is that I don't understand how simply improvements in manifold flow coefficients could give you the same percentage increase in torque at low and high rpms. The flow losses due to bad bellmouths etc. are proportional to the air velocity squared. At low rpms of an engine sized for high rpms, the air velocity is low and the bad vs. good bellmouths etc. don't matter and there's no torque increase that one can detect from improving runner flow coefficients. For example, the curvy runner #5 is filling about as well at low rpms as all the other short runners (based on some fuel trim data).
There are a couple of methods that would boost torque proportionally across the rev range. The first is increasing compression. The thermal efficiency gain is similar at low and high rpms. The second is increasing ignition advance from the base table. That's why I think that your gains could be consistent with fixing a sensor and taking out the error mode retard from the ECUs.
Resonance effects could boost the low-rpm torque. So one thing that could theoretically be happening in your manifold is that the resonance effect improvements improved the low rpm performance and better flow coefficients improved the high rpm performance.
I am having a hard time understanding how your modifications would increase the low rpm torque due to resonance effects, however. Reducing the plenum volume by filling the wall cavities will move the Helmholtz resonator torque peak to higher rpm, lowering the low rpm torque. Inserting the flow divider in the throttle body casting (which may help high rpm flow) will make the ram tube look more like two, independent short tubes to pulses, instead of like a single long u-shaped tube. This effective shortening of the ram tubes will also increase the torque peak rpm of the flappy closed mode (possibly insigificantly since you're not literally sealing off the two tube ends).
All the effects that I see in your manifold make me think that low rpm torque should be reduced and the first torque peak move to the right, if there's any significant effect on resonance tuning. My understanding is inconsistent with your results, so one of those has to be wrong (my understanding or your results).
So my question to you is _why_ do you think your modifications resulted in a torque increase that is proportional across the rpm range? Clearly, you're going by a theory here, but does that theory line up with the results?
I've got a couple of questions about your modifications.
First, did you by any chance fix anything, like a knock sensor or Hall sensor, between the before and after dyno runs?
The reason why I am asking is that I don't understand how simply improvements in manifold flow coefficients could give you the same percentage increase in torque at low and high rpms. The flow losses due to bad bellmouths etc. are proportional to the air velocity squared. At low rpms of an engine sized for high rpms, the air velocity is low and the bad vs. good bellmouths etc. don't matter and there's no torque increase that one can detect from improving runner flow coefficients. For example, the curvy runner #5 is filling about as well at low rpms as all the other short runners (based on some fuel trim data).
There are a couple of methods that would boost torque proportionally across the rev range. The first is increasing compression. The thermal efficiency gain is similar at low and high rpms. The second is increasing ignition advance from the base table. That's why I think that your gains could be consistent with fixing a sensor and taking out the error mode retard from the ECUs.
Resonance effects could boost the low-rpm torque. So one thing that could theoretically be happening in your manifold is that the resonance effect improvements improved the low rpm performance and better flow coefficients improved the high rpm performance.
I am having a hard time understanding how your modifications would increase the low rpm torque due to resonance effects, however. Reducing the plenum volume by filling the wall cavities will move the Helmholtz resonator torque peak to higher rpm, lowering the low rpm torque. Inserting the flow divider in the throttle body casting (which may help high rpm flow) will make the ram tube look more like two, independent short tubes to pulses, instead of like a single long u-shaped tube. This effective shortening of the ram tubes will also increase the torque peak rpm of the flappy closed mode (possibly insigificantly since you're not literally sealing off the two tube ends).
All the effects that I see in your manifold make me think that low rpm torque should be reduced and the first torque peak move to the right, if there's any significant effect on resonance tuning. My understanding is inconsistent with your results, so one of those has to be wrong (my understanding or your results).
So my question to you is _why_ do you think your modifications resulted in a torque increase that is proportional across the rpm range? Clearly, you're going by a theory here, but does that theory line up with the results?
Hey Hans. :-)
No mate, I didn't fix anything. inadvertently or otherwise.
I thought I would get some flow improvement at mid to high rpms, but couldn't believe the oomph in the back when I touched the accelerator at about 2-2500 rpm in second.
Consider this: I'm using a 50 profile rear tire instead of the 45. That's 7% taller gearing. The A/C was on. Yet the throttle response was better than with 45 rears and the A/C off.
It got better. A 50-75 mph 3rd gear roll on went
I think the splitter pre TB, removing the turbulence shadow behind the butterfly, has made a large contribution. But my engineering intuition is also saying the next two downstream airflow mods can't have been all bad either. Only question is the extent of each contribution. It was really a case of as the additions were so easy to do, and the manifold was out, then why not?
Whilst I hear what you're saying about resonance effects, I made my changes in accordance with CFD airflow analyses, reasoning that if I apply these wherever applicable then I've given the manifold its best chance for top performance. The upper plenums lost 600cc volume per side.
I can't offer any more than that, and must say nobody could have been more surprised than I was. As for which of us are correct, maybe we both are to varying extents. I could not/would not claim to have sufficient experience in this area to make such a judgement.
Accidentally or otherwise, here is a low cost, really whoo-hoo performance gain, especially if you have hobby-level hand skills.
A key point is that the benefits any downstream flow mods will be enhanced by the improved airflows.
I must also mention, that while high flow volumes are needed for performance, so are air flow speeds, as it's the momentum of the column of moving air carrying it (air) into the cylinder and building pressure (like hammer blow in water pipes) before the intake closes, that results in cylinder pressures.
I'll make the observation that the first thing I hear (everywhere) from ppl on their way to better performance, is "let's hog out and polish those intake ports and runners".
Runners and ports need be just large enough to get to your HP goal. That way you'll keep maximum low end.
As for polishing, do you realise the fastest creatures in the air or water do not have smooth shiny skins or equivalent? Eg, Peregrine Falcons can fly at 240 mph. Mako Sharks swim at 50 mph.
Paint on the hull of a modern sub has a bad "orange peel" finish to reduce hydrodynamic drag for that reason. (I have a sample of paint flakes from the fin of the USS Chicago). As well as having accoustic anti-reflective properties, it mimics the texture of whale/dolphin skins. Hey - if it's good enough for those guys ... :-) :-) :-)
Nature has been at the game for a good while so it would make sense for us to take note of what it's been up to all these years.
Cheers
Upfixen.
#786
Pro
Good to see someone else do some cfd and incorporate divots as well. I remember discussing this and incorporating it into my SC intake and manifold design back in 2003...
The divots work miracles for laminar flow and a smooth boundary layer!!
The entire plenum roof of my sc is divoted for this reason.
Anyway, didn't mean to hijack, just pleased that others are becoming aware of aero as well.
Cheers,
The divots work miracles for laminar flow and a smooth boundary layer!!
The entire plenum roof of my sc is divoted for this reason.
Anyway, didn't mean to hijack, just pleased that others are becoming aware of aero as well.
Cheers,
You didn't hijack Richard. To the contrary. :-)
IMHO, only a relatively few shark owners want to go down the gung-ho cubic-dollars cubic-horsepower road, but I'll happily bet a huge number wouldn't mind just a little bit more, particularly if they didn't have to risk putting their car on a track or getting on the wrong side of the law to enjoy it.
I can promise any S4 owner who faithfully carries out my mods will get a performance gain, particularly a kick-in-the-back instantaneous throttle response gain, that will have him/her coming home every time with a big (bigger, sillier) silly grin on his/her face, after even just a short trip to the corner shops. . Makes the car 'feel' like a featherweight.
Did I mention "point and shoot"? Gone, is that long, slow "who? Me?" accelerator delay, replaced with racehorse-type readiness. Did I mention the 50-80 mph roll-on? Muahhh haaa haaa.
As you can see, I didn't just stumble upon this by accident. When I had the manifold out several years earlier I had no idea I was looking at such a dog's breakfast of a design.
In the intervening time I happened to discover the subtleties of Aero Stuff at all levels (Re Dimples: did you notice that the higher the airspeeds need a smaller dimple size to delaminate the airflow? That's why they work on both the golf ball the Mythbuster dimpled car).
My mods are not any sort of a complete fix (need a ground-up re-design for that), BUT the bulk fun-factor increase will easily do it for most owners. If it doesn't then you're not easily pleased.
Cheers
UpFixen.
#787
Rennlist Member
...
I can promise any S4 owner who faithfully carries out my mods will get a performance gain, particularly a kick-in-the-back instantaneous throttle response gain, that will have him/her coming home every time with a big (bigger, sillier) silly grin on his/her face, after even just a short trip to the corner shops. . Makes the car 'feel' like a featherweight.
My mods are not any sort of a complete fix (need a ground-up re-design for that), BUT the bulk fun-factor increase will easily do it for most owners. If it doesn't then you're not easily pleased.
I can promise any S4 owner who faithfully carries out my mods will get a performance gain, particularly a kick-in-the-back instantaneous throttle response gain, that will have him/her coming home every time with a big (bigger, sillier) silly grin on his/her face, after even just a short trip to the corner shops. . Makes the car 'feel' like a featherweight.
...
My mods are not any sort of a complete fix (need a ground-up re-design for that), BUT the bulk fun-factor increase will easily do it for most owners. If it doesn't then you're not easily pleased.
#788
Pro
Unfortunately, ten days ago I managed to drop my Goldwing, so progress has been slow. Bodies don't bounce as well as they used to when they're unfit to the n'th and overweight. Ohhh groannnn.
Cheers
UpFixen.
#789
Nordschleife Master
UpFix -- You didn't answer my question How is it possible that improving intake manifold's flow coefficients caused approximately the same percentage torque increase at all rpms? Flow losses being proportional to velocity squared...
#790
Rennlist Member
Hey Hans. :-)
No mate, I didn't fix anything. inadvertently or otherwise.
I thought I would get some flow improvement at mid to high rpms, but couldn't believe the oomph in the back when I touched the accelerator at about 2-2500 rpm in second.
Consider this: I'm using a 50 profile rear tire instead of the 45. That's 7% taller gearing. The A/C was on. Yet the throttle response was better than with 45 rears and the A/C off.
It got better. A 50-75 mph 3rd gear roll on went
I think the splitter pre TB, removing the turbulence shadow behind the butterfly, has made a large contribution. But my engineering intuition is also saying the next two downstream airflow mods can't have been all bad either. Only question is the extent of each contribution. It was really a case of as the additions were so easy to do, and the manifold was out, then why not?
Whilst I hear what you're saying about resonance effects, I made my changes in accordance with CFD airflow analyses, reasoning that if I apply these wherever applicable then I've given the manifold its best chance for top performance. The upper plenums lost 600cc volume per side.
I can't offer any more than that, and must say nobody could have been more surprised than I was. As for which of us are correct, maybe we both are to varying extents. I could not/would not claim to have sufficient experience in this area to make such a judgement.
Accidentally or otherwise, here is a low cost, really whoo-hoo performance gain, especially if you have hobby-level hand skills.
A key point is that the benefits any downstream flow mods will be enhanced by the improved airflows.
I must also mention, that while high flow volumes are needed for performance, so are air flow speeds, as it's the momentum of the column of moving air carrying it (air) into the cylinder and building pressure (like hammer blow in water pipes) before the intake closes, that results in cylinder pressures.
I'll make the observation that the first thing I hear (everywhere) from ppl on their way to better performance, is "let's hog out and polish those intake ports and runners".
Runners and ports need be just large enough to get to your HP goal. That way you'll keep maximum low end.
As for polishing, do you realise the fastest creatures in the air or water do not have smooth shiny skins or equivalent? Eg, Peregrine Falcons can fly at 240 mph. Mako Sharks swim at 50 mph.
Paint on the hull of a modern sub has a bad "orange peel" finish to reduce hydrodynamic drag for that reason. (I have a sample of paint flakes from the fin of the USS Chicago). As well as having accoustic anti-reflective properties, it mimics the texture of whale/dolphin skins. Hey - if it's good enough for those guys ... :-) :-) :-)
Nature has been at the game for a good while so it would make sense for us to take note of what it's been up to all these years.
Cheers
Upfixen.
No mate, I didn't fix anything. inadvertently or otherwise.
I thought I would get some flow improvement at mid to high rpms, but couldn't believe the oomph in the back when I touched the accelerator at about 2-2500 rpm in second.
Consider this: I'm using a 50 profile rear tire instead of the 45. That's 7% taller gearing. The A/C was on. Yet the throttle response was better than with 45 rears and the A/C off.
It got better. A 50-75 mph 3rd gear roll on went
I think the splitter pre TB, removing the turbulence shadow behind the butterfly, has made a large contribution. But my engineering intuition is also saying the next two downstream airflow mods can't have been all bad either. Only question is the extent of each contribution. It was really a case of as the additions were so easy to do, and the manifold was out, then why not?
Whilst I hear what you're saying about resonance effects, I made my changes in accordance with CFD airflow analyses, reasoning that if I apply these wherever applicable then I've given the manifold its best chance for top performance. The upper plenums lost 600cc volume per side.
I can't offer any more than that, and must say nobody could have been more surprised than I was. As for which of us are correct, maybe we both are to varying extents. I could not/would not claim to have sufficient experience in this area to make such a judgement.
Accidentally or otherwise, here is a low cost, really whoo-hoo performance gain, especially if you have hobby-level hand skills.
A key point is that the benefits any downstream flow mods will be enhanced by the improved airflows.
I must also mention, that while high flow volumes are needed for performance, so are air flow speeds, as it's the momentum of the column of moving air carrying it (air) into the cylinder and building pressure (like hammer blow in water pipes) before the intake closes, that results in cylinder pressures.
I'll make the observation that the first thing I hear (everywhere) from ppl on their way to better performance, is "let's hog out and polish those intake ports and runners".
Runners and ports need be just large enough to get to your HP goal. That way you'll keep maximum low end.
As for polishing, do you realise the fastest creatures in the air or water do not have smooth shiny skins or equivalent? Eg, Peregrine Falcons can fly at 240 mph. Mako Sharks swim at 50 mph.
Paint on the hull of a modern sub has a bad "orange peel" finish to reduce hydrodynamic drag for that reason. (I have a sample of paint flakes from the fin of the USS Chicago). As well as having accoustic anti-reflective properties, it mimics the texture of whale/dolphin skins. Hey - if it's good enough for those guys ... :-) :-) :-)
Nature has been at the game for a good while so it would make sense for us to take note of what it's been up to all these years.
Cheers
Upfixen.
then, you start talking about the dimples.... a smooth surface will always have less drag than a dimpled surface. why do you think airplanes are perfectly smooth? intakes in F1 engines are smooth... the only time they make rough surfaces is when they want turbulence to get better fuel and air mixing
golf ***** have dimples for control, not for reduced drag. dont confuse incompressible fluid flow( water), with theory of compressible fluid flow (air)
the principles of each are in some cases, direct opposite .
you are right about the splitter (inlet guide vanes) they are useful, anytime you have a bend or a turn in an intake air path. a tremendous reduction of turbulence and flow efficiency.
#791
Pro
There would have to be a benefit at all rpm, but as you say, One would not expect it to be so even.
All I can point to is my earlier comment, re that of the pre-TB splitter eliminating much of the butterfly plate turbulence shadow (flow restriction).
As for me? Well right now I'm not too worried if my analysis is not quite correct; I'm treating it as a work in progress. In the meantime I'm not tiring one bit of the instant 'go'.
:-D
UpFixen
#792
Rennlist Member
That I can't definitively answer Ptuomov.
There would have to be a benefit at all rpm, but as you say, One would not expect it to be so even.
All I can point to is my earlier comment, re that of the pre-TB splitter eliminating much of the butterfly plate turbulence shadow (flow restriction).
As for me? Well right now I'm not too worried if my analysis is not quite correct; I'm treating it as a work in progress. In the meantime I'm not tiring one bit of the instant 'go'.
:-D
UpFixen
There would have to be a benefit at all rpm, but as you say, One would not expect it to be so even.
All I can point to is my earlier comment, re that of the pre-TB splitter eliminating much of the butterfly plate turbulence shadow (flow restriction).
As for me? Well right now I'm not too worried if my analysis is not quite correct; I'm treating it as a work in progress. In the meantime I'm not tiring one bit of the instant 'go'.
:-D
UpFixen
#793
Nordschleife Master
But the benefit is approximately proportional to velocity squared. Velocity is approximately proportional to hp. So the benefit should scale up with the square of hp. For example, if you make extra 10 kW at 6000 rpm (230->240kW) from better flowing throttle body, then you'd make only 1.2kW at 2000 rpm (80->81.2kW). SWAG, of course, but illustrates the point. So how can a pre-TB splitter produce the constant benefits across the rpm range that you're seeing on the dyno?
#794
Former Vendor
But the benefit is approximately proportional to velocity squared. Velocity is approximately proportional to hp. So the benefit should scale up with the square of hp. For example, if you make extra 10 kW at 6000 rpm (230->240kW) from better flowing throttle body, then you'd make only 1.2kW at 2000 rpm (80->81.2kW). SWAG, of course, but illustrates the point. So how can a pre-TB splitter produce the constant benefits across the rpm range that you're seeing on the dyno?
The discrepancies of the dyno results were pointed out in the original thread....linear horsepower increases across the rpm range from increased airflow, simply not possible.
Pretty obviously the engine got more timing....that's the only explanation that fits, short of increasing the displacement or increasing the compression, which obviously did not happen.
Knock sensor contact, hall sensor contact, full throttle switch contact....one of them wasn't working and then it was.
#795
Nordschleife Master
With these old cars, multiple things can change without any knowledge of them happening....or changing any parts.
The discrepancies of the dyno results were pointed out in the original thread....linear horsepower increases across the rpm range from increased airflow, simply not possible.
Pretty obviously the engine got more timing....that's the only explanation that fits, short of increasing the displacement or increasing the compression, which obviously did not happen.
Knock sensor contact, hall sensor contact, full throttle switch contact....one of them wasn't working and then it was.
The discrepancies of the dyno results were pointed out in the original thread....linear horsepower increases across the rpm range from increased airflow, simply not possible.
Pretty obviously the engine got more timing....that's the only explanation that fits, short of increasing the displacement or increasing the compression, which obviously did not happen.
Knock sensor contact, hall sensor contact, full throttle switch contact....one of them wasn't working and then it was.