Full Version: I need some advice
I'm slowly designing the intake for my LT1 914.. I calculated the area of the thottle body to be about 5 square inches. Am I correct in assuming that as long as the rest of the intake tract has a cross sectional area of greater than 5 square inches, it won't be restrictive? (provided the interior surfaces are smooth with no sharp bends)
I forget the specifics but the loss of flow do to length of a tube is significant, I'll try to look it up. In short, No. It has to be a fair bit larger.
Basically you need one of the attached scales, it all depends on length, volume and velocity. See below.
The majority of practical problems fall within the Turbulent Flow range and this calculator determines the required pipe size, flow or pressure loss for any liquid and also for any gas (at low pressures) flowing under turbulent conditions.
Pressure loss, flow or pipe size are found from the front of the calculator by a simple setting of two dials to the known details of length, viscosity, specific gravity, etc. Scales on the reverse side permit an instant determination on whether the flow conditions are Turbulent or Laminar. Viscosities of over 50 liquids at various temperatures are also given on the reverse side along with their specific gravities. Further scales are incorporated for viscosity conversion, including Redwood, Saybolt, Centistokes, and on the metric model, Engler degrees.
Scale Ranges:
Diameter: 1/4" to 40" Liquid Flow: 5 to 1,000,000 g.p.h.
Length: 5 ft. to 5,000,000 ft.
Gas Flow: 50 to 1,000,000 c.f.h.
Pressure Drop: .01 to 500 lbs./sq.in.
The majority of practical problems fall within the Turbulent Flow range and this calculator determines the required pipe size, flow or pressure loss for any liquid and also for any gas (at low pressures) flowing under turbulent conditions.
Pressure loss, flow or pipe size are found from the front of the calculator by a simple setting of two dials to the known details of length, viscosity, specific gravity, etc. Scales on the reverse side permit an instant determination on whether the flow conditions are Turbulent or Laminar. Viscosities of over 50 liquids at various temperatures are also given on the reverse side along with their specific gravities. Further scales are incorporated for viscosity conversion, including Redwood, Saybolt, Centistokes, and on the metric model, Engler degrees.
Scale Ranges:
Diameter: 1/4" to 40" Liquid Flow: 5 to 1,000,000 g.p.h.
Length: 5 ft. to 5,000,000 ft.
Gas Flow: 50 to 1,000,000 c.f.h.
Pressure Drop: .01 to 500 lbs./sq.in.
like this
Royce,
The other restriction you need to consider is the filtration. You should see how large Mike's air cleaner is, dam near blocks vision out the rear window.
Are you changing the plumbing of the intake to find a colder air source or because the LT-1 inlet is right at the firewall?
The other restriction you need to consider is the filtration. You should see how large Mike's air cleaner is, dam near blocks vision out the rear window.
Are you changing the plumbing of the intake to find a colder air source or because the LT-1 inlet is right at the firewall?
Ok, this stuff is WAAAAY over my head 
Ok, here's the deal.. I turned the intake around so it will face the trunk, near where the distributor cutout is on a "normal" V8 conversion. I have a couple options..
1. Elbow and long tube going out the side to the wheel well, to some sort of filter.
Pro: cool air
Cons: long pipe, possible targa top storage conflict, and it'll pick up anything thrown up by the rear wheels.
2. I picked up a nifty trans am "ram air" aircleaner that takes two square filter elements. The elements would sit in the floor of the trunk on either side of the transmission. It has a wide flat duct between the two, that leads up to a round duct fitting that would pretty much go straight up to the intake.
Pros: shorter duct, no targa interference, nifty Dzus fasteners, and I already have it.
Cons: Came off a V6, so it may be undersized, pretty much the hottest place I could get intake air.
Ok, here's the deal.. I turned the intake around so it will face the trunk, near where the distributor cutout is on a "normal" V8 conversion. I have a couple options..
1. Elbow and long tube going out the side to the wheel well, to some sort of filter.
Pro: cool air
Cons: long pipe, possible targa top storage conflict, and it'll pick up anything thrown up by the rear wheels.
2. I picked up a nifty trans am "ram air" aircleaner that takes two square filter elements. The elements would sit in the floor of the trunk on either side of the transmission. It has a wide flat duct between the two, that leads up to a round duct fitting that would pretty much go straight up to the intake.
Pros: shorter duct, no targa interference, nifty Dzus fasteners, and I already have it.
Cons: Came off a V6, so it may be undersized, pretty much the hottest place I could get intake air.
Ok, I'm trying to calculate the velocity of the intake air, based on the 5 square inch area of the throttle body, but I've hit a roadblock.
I started with: Engine = 300 HP @30% efficiency
That means I need 1000 hp of gasoline.
1 gallon of gas has 1.3 x 10^8 joules of energy
1 HP-hour is 2.7X10^6 Joules
1 gallon of gas weighs 6.2 lbs
Air mass/fuel mass ratio is 14.7:1
air density, etc..
Unless I screwed up I need 973 cubic inches of air drawn through a 5 square inch area per second. What's the velocity?
Perhaps it would have been easier to work from engine displacement and RPM..
I started with: Engine = 300 HP @30% efficiency
That means I need 1000 hp of gasoline.
1 gallon of gas has 1.3 x 10^8 joules of energy
1 HP-hour is 2.7X10^6 Joules
1 gallon of gas weighs 6.2 lbs
Air mass/fuel mass ratio is 14.7:1
air density, etc..
Unless I screwed up I need 973 cubic inches of air drawn through a 5 square inch area per second. What's the velocity?
Perhaps it would have been easier to work from engine displacement and RPM..
I always wanted to docustom 2-3" snorkels off both Targa panels with a nice slow radius into a reversed intake like you have. I'd do more of a sheetmetal plenum than pipe though.
Hmm, based on displacement and RPM I get 16,625 cubic inches per second.. something's not right. (I did take into account that only half the engine is on an intake stroke per revolution)
sounds about right
Keep in mind any bends or air re-directions impact the flow too. I really don't think for the short distances you are talking about you will see much of a decline in volume. When in doubt overengineer though.
I think I'm looking at 250 feet per second or so, of air. Is that enough to be able to tell if it's laminar or turbulent? (I'm assuming laminar behaves predictably and turbulent gets complicated)
Laminar has more to do with smooth transitions, no eddys like log jams in a river. The folowing is some fluid related stuff, but the principle is the same as length of pipe increases pressure must increase or flow will decrease, the math to figure out exactly how much is up to you. I don't have one of the nifty lillte slide rules.
Viscosity
Earlier we stated that liquids do not maintain a shape because they cannot sustain shear forces and the molecules slide easily by one another. That does not mean there are no attractive forces between molecules, but simply that those forces are insufficient to maintain a shape. The molecules' mutual attraction to one another is called cohesion and is the cause of viscosity. Viscosity in common language refers to the "thickness" of the liquid. Salad oil is more viscous than water, and honey is more viscous than salad oil. (Viscosity is also function of temperature. In fact, as most liquids cool, they become more viscous and eventually become solid. But this behavior is not simple and we won't focus on it in this course.)
In fluid dynamics, the viscosity is denoted by the greek letter eta, h. It is defined by the fluid's resistance to flow in a way that is reminiscent of how we defined Young's Modulus for solids in the first module. Here's the idea. A thin layer of the liquid of thickness t is trapped between two flat plates of area A. The bottom plate is fixed and the top plate is pulled to the right with a velocity v. A certain force F will be required to maintain this velocity. The more viscous the fluid, the greater the force needed. The definition is
h = Shear Stress / Strain rate = F / A /V / t
Viscosity has the rather odd SI unit of Pa.seconds. But the most commonly used unit is the Poise = .1 Pa.s. There are a few things to note about the behavior of the fluid. First, there is a thin layer of the fluid (perhaps only a molecule thick) that is "stuck" to each plate. It is dragged along with the plate by an attractive force between the fluid and plate called adhesion. In the diagram, the velocity of the fluid is zero at the bottom plate and v at the top. The fluid velocity increases linearly from the zero at the bottom to v at the top. Think of the fluid as many layers sliding across one another, exerting cohesive forces on each other. This internal friction has a marked effect on how fluid flows through pipes.
The velocity profile for water flowing through a pipe is similar to that of the two plates, except the velocity is zero at the walls of the pipe and the maximum value v is at the center of the pipe. So how does the viscosity effect the flow rate of a fluid through a pipe? A Frenchman name Poiseuille derived an expression for the flow rate through a horizontal pipe of radius R and length L with a pressure difference of DP from one end to the other. The derivation is challenging even for our physics majors, so we'll just state the result and discuss what it tells us about installing plumbing.
Flow Rate = p / 8 (R4 / h ) (DP / L )
We want to understand the implications of this equation. The viscosity for a given fluid is a fixed value. What are the variables that can we control? They are the radius of the pipe, the pressure difference, and the length of the pipe. What does Poiseuille's equation say about each of those?
The flow rate depends upon R4! This result may seem suprising, but remember that the Equation of Continuity told us that a pipe with a 1" diameter will have four times the flow rate of that of a 1/2 " pipe for the same flow velocity. In addition, the drag decreases as the pipe size inceases. Poiseuille's Eq. says that for a given pressure difference and given length, the flowrate through a 1" pipe is 16 times that for a 1/2 " pipe!
The pressure difference is what drives the fluid through the pipe. It is determine by your water pump and the elevation difference. Clearly a stronger pump can provide a greater pressure gradient. Note that our previous calculation for checking whether a 20 psi pump could push water up 6 meters to the second floor, did not consider what flow rate could be maintained with such a pressure. We revisit this problem below.
The effect of the length of the pipe is easy to understand. A pipe with twice the length will produce twice as much drag and cut the flow rate in half. So we don't want to make our pipes any longer than we have to. And if there is a particularly long run (for example, out to the workshop located a hundred feet from the pump), then perhaps we should invest in a larger pipe!
There are two ways to look at P's Eq. The pressure difference DP maintains the flow rate OR for a given flow rate, there will be a pressure drop DP. Plumbers generally look at it in the latter way. They estimate the flow rate that will be needed and from that calculate the pressure loss along a given length of pipe to determine whether or not they will have use a larger pipe.
Turbulence
All of the equations we have introduced for fluid flow apply only for laminar flow. Laminar flow is flow in which the fluid "layers" slide smoothly over one another. In laminar flow there are no vortices (whorls) or cavitation. Vortices are where the fluid rotates back upon itself and cavitation is where bubbles and voids are created. Such turbulence often occurs at barriers or wherever there are abrupt changes is the direction of the fluid flow. Turbulence introduces frictional energy losses into the system that are very difficult, if not impossible to predict. Even the seemingly simplist of problems typically must be solve with computers. As the velocity of a fluid increases, so does the onset of turbulence. (The Reynolds number provides an estimate of the transition between laminar and turbulent flow. It is a number calculated from flow rate, pipe size, fluid viscosity and density.) The diagram below shows how the shape of a junction between two different sized pipes can effect the onset of turbulence.
In laminar flow, the streamlines converge (or diverge) smoothly. In turbulenct flow, the streamlines become erratic and the increase in friction can become very significant. Despite its complex nature, there are a few simple things you can do to reduce turbulence in your plumbing.
Viscosity
Earlier we stated that liquids do not maintain a shape because they cannot sustain shear forces and the molecules slide easily by one another. That does not mean there are no attractive forces between molecules, but simply that those forces are insufficient to maintain a shape. The molecules' mutual attraction to one another is called cohesion and is the cause of viscosity. Viscosity in common language refers to the "thickness" of the liquid. Salad oil is more viscous than water, and honey is more viscous than salad oil. (Viscosity is also function of temperature. In fact, as most liquids cool, they become more viscous and eventually become solid. But this behavior is not simple and we won't focus on it in this course.)
In fluid dynamics, the viscosity is denoted by the greek letter eta, h. It is defined by the fluid's resistance to flow in a way that is reminiscent of how we defined Young's Modulus for solids in the first module. Here's the idea. A thin layer of the liquid of thickness t is trapped between two flat plates of area A. The bottom plate is fixed and the top plate is pulled to the right with a velocity v. A certain force F will be required to maintain this velocity. The more viscous the fluid, the greater the force needed. The definition is
h = Shear Stress / Strain rate = F / A /V / t
Viscosity has the rather odd SI unit of Pa.seconds. But the most commonly used unit is the Poise = .1 Pa.s. There are a few things to note about the behavior of the fluid. First, there is a thin layer of the fluid (perhaps only a molecule thick) that is "stuck" to each plate. It is dragged along with the plate by an attractive force between the fluid and plate called adhesion. In the diagram, the velocity of the fluid is zero at the bottom plate and v at the top. The fluid velocity increases linearly from the zero at the bottom to v at the top. Think of the fluid as many layers sliding across one another, exerting cohesive forces on each other. This internal friction has a marked effect on how fluid flows through pipes.
The velocity profile for water flowing through a pipe is similar to that of the two plates, except the velocity is zero at the walls of the pipe and the maximum value v is at the center of the pipe. So how does the viscosity effect the flow rate of a fluid through a pipe? A Frenchman name Poiseuille derived an expression for the flow rate through a horizontal pipe of radius R and length L with a pressure difference of DP from one end to the other. The derivation is challenging even for our physics majors, so we'll just state the result and discuss what it tells us about installing plumbing.
Flow Rate = p / 8 (R4 / h ) (DP / L )
We want to understand the implications of this equation. The viscosity for a given fluid is a fixed value. What are the variables that can we control? They are the radius of the pipe, the pressure difference, and the length of the pipe. What does Poiseuille's equation say about each of those?
The flow rate depends upon R4! This result may seem suprising, but remember that the Equation of Continuity told us that a pipe with a 1" diameter will have four times the flow rate of that of a 1/2 " pipe for the same flow velocity. In addition, the drag decreases as the pipe size inceases. Poiseuille's Eq. says that for a given pressure difference and given length, the flowrate through a 1" pipe is 16 times that for a 1/2 " pipe!
The pressure difference is what drives the fluid through the pipe. It is determine by your water pump and the elevation difference. Clearly a stronger pump can provide a greater pressure gradient. Note that our previous calculation for checking whether a 20 psi pump could push water up 6 meters to the second floor, did not consider what flow rate could be maintained with such a pressure. We revisit this problem below.
The effect of the length of the pipe is easy to understand. A pipe with twice the length will produce twice as much drag and cut the flow rate in half. So we don't want to make our pipes any longer than we have to. And if there is a particularly long run (for example, out to the workshop located a hundred feet from the pump), then perhaps we should invest in a larger pipe!
There are two ways to look at P's Eq. The pressure difference DP maintains the flow rate OR for a given flow rate, there will be a pressure drop DP. Plumbers generally look at it in the latter way. They estimate the flow rate that will be needed and from that calculate the pressure loss along a given length of pipe to determine whether or not they will have use a larger pipe.
Turbulence
All of the equations we have introduced for fluid flow apply only for laminar flow. Laminar flow is flow in which the fluid "layers" slide smoothly over one another. In laminar flow there are no vortices (whorls) or cavitation. Vortices are where the fluid rotates back upon itself and cavitation is where bubbles and voids are created. Such turbulence often occurs at barriers or wherever there are abrupt changes is the direction of the fluid flow. Turbulence introduces frictional energy losses into the system that are very difficult, if not impossible to predict. Even the seemingly simplist of problems typically must be solve with computers. As the velocity of a fluid increases, so does the onset of turbulence. (The Reynolds number provides an estimate of the transition between laminar and turbulent flow. It is a number calculated from flow rate, pipe size, fluid viscosity and density.) The diagram below shows how the shape of a junction between two different sized pipes can effect the onset of turbulence.
In laminar flow, the streamlines converge (or diverge) smoothly. In turbulenct flow, the streamlines become erratic and the increase in friction can become very significant. Despite its complex nature, there are a few simple things you can do to reduce turbulence in your plumbing.
Wow... Ok. overengineering definitely wins over trying to calculate this stuff. Thanks for explaining it though, the parts I understood help alot. This bit here...
is something I can wrap my brain around. It gives me an idea on how big of an effect diameter will have.
Thanks for explaining it though, the parts I understood help alot. This bit here...| QUOTE |
| a pipe with a 1" diameter will have four times the flow rate of that of a 1/2 " pipe for the same flow velocity |
is something I can wrap my brain around. It gives me an idea on how big of an effect diameter will have.
That is about what I get out of it too. I know somewhere I have a very simple equation that someone gave me but I can't find it. That is as close as I could find. Sorry not to be of more help.
| QUOTE (MikeP @ Aug 4 2005, 01:45 PM) |
| I always wanted to docustom 2-3" snorkels off both Targa panels with a nice slow radius into a reversed intake like you have. I'd do more of a sheetmetal plenum than pipe though. |
There are some nice snorkels for MR2's
Ya, but go through the sail pannel with the inlet facing forward, those look sort of like they face down. And the plennum would swoop around a little higher and be larger.
WOW , WHAT A LOT OF OVERENGINEERING!
For my 2 cents: your original premise was good: if the feed tube is larger than 5" , smooth and short as possible
you should be good. I'm assuming this is NOT a 750HP race engine, so all the BIG FLOW issures only come into play in the last 500 RPM of your rev range....it's a streetcar, right?
What I would suggest is to vent the filter/intake to a HIGH pressure area. For my car I made my own fiberglass side scoops: 1 for intake and the other for the oil coolers. High pressure feed will overcome a bunch of the other engineering imperfections!
I have seen a photo of a 914 with the manifold turned around and sucking air from the trunk....of course then you would need an air intake for the trunk so you won't suck the trunk lid IN!
Best of luck,
Terry Stewart
For my 2 cents: your original premise was good: if the feed tube is larger than 5" , smooth and short as possible
you should be good. I'm assuming this is NOT a 750HP race engine, so all the BIG FLOW issures only come into play in the last 500 RPM of your rev range....it's a streetcar, right?
What I would suggest is to vent the filter/intake to a HIGH pressure area. For my car I made my own fiberglass side scoops: 1 for intake and the other for the oil coolers. High pressure feed will overcome a bunch of the other engineering imperfections!
I have seen a photo of a 914 with the manifold turned around and sucking air from the trunk....of course then you would need an air intake for the trunk so you won't suck the trunk lid IN!
Best of luck,
Terry Stewart
| QUOTE (Terryst1 @ Aug 5 2005, 07:01 PM) |
| WOW , WHAT A LOT OF OVERENGINEERING! For my 2 cents: your original premise was good: if the feed tube is larger than 5" , smooth and short as possible you should be good. I'm assuming this is NOT a 750HP race engine, so all the BIG FLOW issures only come into play in the last 500 RPM of your rev range....it's a streetcar, right? What I would suggest is to vent the filter/intake to a HIGH pressure area. For my car I made my own fiberglass side scoops: 1 for intake and the other for the oil coolers. High pressure feed will overcome a bunch of the other engineering imperfections! I have seen a photo of a 914 with the manifold turned around and sucking air from the trunk....of course then you would need an air intake for the trunk so you won't suck the trunk lid IN! Best of luck, Terry Stewart |
Yup, street car (aiming for a quick daily driver) and around 300 HP. I've basically come to the conclusion that it really doesn't matter much.. I can always make a g-tech run with and without the ducting bits and see if it matters.
I want to avoid scoops because I don't really like the look of scoops on a pretty much stock bodied 914.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.