Intake pipes, straight, tapered or bellmouth
Discussion
As the title really what is the PH collective and considered opinions to the pro's and cons of the three main intake pipe types. My limited understanding is:
Straight has a good reflective coefficent, so is best for pulse tuned lengths but suffers from poor flow capacity.
Tapered has good flow and pretty good reflective coefficent so a provides a balance.
The Bellmouth has the highest flow but poorest reflective coefficient, but as outright flow is the dominant effect may be the best out of teh box so to speak.
What is the best configuration for a road engine intake and what can be done to improve them
Straight has a good reflective coefficent, so is best for pulse tuned lengths but suffers from poor flow capacity.
Tapered has good flow and pretty good reflective coefficent so a provides a balance.
The Bellmouth has the highest flow but poorest reflective coefficient, but as outright flow is the dominant effect may be the best out of teh box so to speak.
What is the best configuration for a road engine intake and what can be done to improve them
If you can taper the entire intake then terrific, however you do need a small straight section ( usually enforced by the cylinder head in any case) before you start tapering out again to meet the valve seat. Flow is the be all and end all of port design for power, the tuning pulse will reflect of large changes of area, so sharp steps will give a partial reflection and diminish the tuning effect.
If a port is well designed then you'll only start to get partial reflection just as the pulse approaches the bellmouth
So to sum up intakes should be have a large radiused entry, with a long tapered section becoming striaght for a short period (approx 15-20% of the port length) then tapered out again to meet the valve seat
Matt
If a port is well designed then you'll only start to get partial reflection just as the pulse approaches the bellmouth
So to sum up intakes should be have a large radiused entry, with a long tapered section becoming striaght for a short period (approx 15-20% of the port length) then tapered out again to meet the valve seat
Matt
OK following it so far ... good bellmouthed intake (with a 180 roll), a long taper straightening out for a bit towards the valve before flairing a bit for the valve seat .
Matt when you say you 'only get partial reflection' as it reaches the bellmouth I understood this to beone of the minor down sides to bellmouths i.e. they do not aid tuned lengths as well as a straight pipe which has a high coefficient of reflection (sudden change in diameter) and hence a good strong reflective pulse to aid pulse tuning at the desired length (rpm). I've seen plates put above bell mouths, (cossie/bda) to try to get the 'lost' reflect pulse back in the runner but never understood how it doesn't compromise the air being sucked in by the bell .
Matt when you say you 'only get partial reflection' as it reaches the bellmouth I understood this to beone of the minor down sides to bellmouths i.e. they do not aid tuned lengths as well as a straight pipe which has a high coefficient of reflection (sudden change in diameter) and hence a good strong reflective pulse to aid pulse tuning at the desired length (rpm). I've seen plates put above bell mouths, (cossie/bda) to try to get the 'lost' reflect pulse back in the runner but never understood how it doesn't compromise the air being sucked in by the bell .
You're better off using a nice set of ram pipes inside an air box to aid the pulse tuning. You get the benefits of both that way, and still use a sensible air filter. You can tune the ram pipe lengths to aid torque at one rev point, and then the airbox to do it at another. The benefits can be added together to help remove flat spots, as they are both rpm specific.
HarryW said:
I've seen plates put above bell mouths, (cossie/bda) to try to get the 'lost' reflect pulse back in the runner but never understood how it doesn't compromise the air being sucked in by the bell .
Most of the air drawn into the trumpet actually comes from around the side which is why a well-radiused bell mouth is so desirable.I've run the F1 flow rig up to full speed and you can put your hand pretty close to the trumpet without it effecting flow, its not until you get about 25-35mm is starts to drop off
Matt
Always thought the bells breathed from the side more, good to hear it first hand though. So that leaves space above them to try and catch/reflect the escaped reversion pulse. Any idea of distance for the stand off I've heard 1.5 D as a good starting point banded about, problem being what is D is it the throat post bell or is half way up the bell or the end of the bell (stop sniggering at the back ).
The way I think about this is to consider the streamlines at the engine and then see where they come from as you get further away. The area change versus distance from the valve gives you the basic pulse/reflective characteristics. As long as the flow is reasonably smooth and contracts smoothly on the way in (i.e. no sudden contraction, and no expansion) it doesn't make much odds what shape the air flow follows. But from the area changes you can see how it's likely to tune in terms of pulse reflection.
If the air flow follows the shape of the bellmouth then the air entering the bellmouth at 90 degrees to the trumpets needs a cylindrical area equal to the trumpet csa in order to have no area change. You can do the arithmetic for yourself, the height of the lid above the bellmouth needs to be rather less than the radius of the trumpet (not the bellmouth) to give equal area.
In order to get reflection at the bellmouth you would need an expansion there, but there's not much point increasing the area by more than about 50% so this gives a rough order of magnitude of the maximum useful height above the bellmouth. Reducing the height here can reduce the area change which would have similar effects to lengthening the intake tract. I don't think you would ever want to reduce the height so far that the total entry area was less than the trumpet csa.
I think to go beyond this simplistic understanding you need to start think in terms of pressure pulses reflecting within the plenum, which is a seriously difficult problem to get to grips with.
If the air flow follows the shape of the bellmouth then the air entering the bellmouth at 90 degrees to the trumpets needs a cylindrical area equal to the trumpet csa in order to have no area change. You can do the arithmetic for yourself, the height of the lid above the bellmouth needs to be rather less than the radius of the trumpet (not the bellmouth) to give equal area.
In order to get reflection at the bellmouth you would need an expansion there, but there's not much point increasing the area by more than about 50% so this gives a rough order of magnitude of the maximum useful height above the bellmouth. Reducing the height here can reduce the area change which would have similar effects to lengthening the intake tract. I don't think you would ever want to reduce the height so far that the total entry area was less than the trumpet csa.
I think to go beyond this simplistic understanding you need to start think in terms of pressure pulses reflecting within the plenum, which is a seriously difficult problem to get to grips with.
I assume this is for your cerbera Harry ?
In my experience (which is fairly limited i have to say but i tried to so at least *some* research) there's no substitute for actually getting something on the car cand trying it. It took no computer modelling, no degree in fluid flow dynamics and not even gut instinct to arrive at the design for my mk2 whirlwind for example .. it just took time on the dyno trying different shapes out until I thought i'd gone as far as i could in the space restrictions. And it works better then the early "thin tube" versions, so that's good
Also when i was doing the pulse plates I didn't think about anything like 1.5D the throat of the trumpet .. I made some plates and tried them on the car at different distances from the trumpets until i found the best place. No fancy equations involved there either, just time.
One thing you will find is that pulse plates work only over a very small part of the rev range, so i use them for filling in the worst hole in the power curve .. if you look at NickCC's last dyno chart you'll see it's almost perfectly flat like a roverV8 .. I didn't use them for finding more hp at peak I used them for filling in the lowest part of the hp graph for a beautiful straight line hp curve.
Of course every time you make a change you have to remap and make sure you've not lost hp somewhere else important but that's relatively easy. all you really need is time on the dyno
In my experience (which is fairly limited i have to say but i tried to so at least *some* research) there's no substitute for actually getting something on the car cand trying it. It took no computer modelling, no degree in fluid flow dynamics and not even gut instinct to arrive at the design for my mk2 whirlwind for example .. it just took time on the dyno trying different shapes out until I thought i'd gone as far as i could in the space restrictions. And it works better then the early "thin tube" versions, so that's good
Also when i was doing the pulse plates I didn't think about anything like 1.5D the throat of the trumpet .. I made some plates and tried them on the car at different distances from the trumpets until i found the best place. No fancy equations involved there either, just time.
One thing you will find is that pulse plates work only over a very small part of the rev range, so i use them for filling in the worst hole in the power curve .. if you look at NickCC's last dyno chart you'll see it's almost perfectly flat like a roverV8 .. I didn't use them for finding more hp at peak I used them for filling in the lowest part of the hp graph for a beautiful straight line hp curve.
Of course every time you make a change you have to remap and make sure you've not lost hp somewhere else important but that's relatively easy. all you really need is time on the dyno
Thanks for the replies peter/joolz
Agree that empirically is the best way to achieve the results but using the 'theory' gets you in the right area to begin with imhe.
Read and interesting bit about bells and why the tuned length of an intake isn't quite where you expect it to be on the rpm's. The effective intake length is not measured from the output of the bell to the back of the valve but approx 0.5/0.6 D above the bell so for a 50mm intake that's 25mm (1 inch ) above the bell, thus assisting/tuned roughly 500rpm (obviously differs dependant on engine/intake) less rpm than expected. So perhaps thats where I got the 1 1/2 D from i.e. if you want to reflect a full wave back then add 1/2D to it.
I shall experiment, obviously the constraints of the air box will have the final say in what can be done, in particular No's 7&8 cylinders as they are at the end of the tapered air box.
Agree that empirically is the best way to achieve the results but using the 'theory' gets you in the right area to begin with imhe.
Read and interesting bit about bells and why the tuned length of an intake isn't quite where you expect it to be on the rpm's. The effective intake length is not measured from the output of the bell to the back of the valve but approx 0.5/0.6 D above the bell so for a 50mm intake that's 25mm (1 inch ) above the bell, thus assisting/tuned roughly 500rpm (obviously differs dependant on engine/intake) less rpm than expected. So perhaps thats where I got the 1 1/2 D from i.e. if you want to reflect a full wave back then add 1/2D to it.
I shall experiment, obviously the constraints of the air box will have the final say in what can be done, in particular No's 7&8 cylinders as they are at the end of the tapered air box.
Edited by HarryW on Friday 29th June 13:48
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff