Author Topic: Exhaust Flow and Aero  (Read 125764 times)

0 Members and 1 Guest are viewing this topic.

Offline ChrisLenahan

  • Jr. Member
  • **
  • Posts: 48
Re: Exhaust Flow and Aero
« Reply #165 on: October 16, 2011, 04:03:58 PM »
Rex,  That is beautiful metal work!  I have aspirations of being able to do work like that.  I guess maybe I am using the laminar in the wrong context.  By laminar I mean the airflow is attached, and in my understanding keeping it attached or reattaching it is the goal of aerodynamic as applied to LSR.  My thought is on a scoop like that to have  an inner and outer skin with suction holes near the front of the scoop body to get the airflow to reattach sooner.  After watching some of the videos from A2 it seems this is an area that causes drag in most applications.   Chris

Offline interested bystander

  • Sr. Member
  • ****
  • Posts: 997
Re: Exhaust Flow and Aero
« Reply #166 on: October 16, 2011, 10:12:22 PM »
OUTSTANDING, REX,

Looks like some inner workings for air control.

Re:Your semi self- critique - as a hafas scoop builder myself -making sure the air TURNS the corner made me look dumb a few years ago when I partnered with a cylinder head wiz - big wakeup call on the TURN part.

I learned you can't wish air to do your bidding without helpin' guide it.
5 mph in pit area (clothed)

Offline superford317

  • Full Member
  • ***
  • Posts: 67
Re: Exhaust Flow and Aero
« Reply #167 on: November 02, 2011, 01:59:36 PM »
A lot of research and tax payer money went into NACA submerged inlet development, in the beginning there were 5 separate scientific investigations undertaken over a period of about 10 years, by some of the best scientist and engineers in the country. The name of the NACA submerged air inlet comes from the organization that first developed it, National Advisory Committee for Aeronautics, the predecessor to NASA. One of the NACA scientific investigations was completed and came up with the NACA duct with the divergent parallel walls, wanting a higher pressure recovery they launched another scientific investigation and came up with the curved divergent ramp NACA duct, that we know today, that has higher pressure recoveries. The first investigations were done small scale, in later investigations performed on full scale models; pressure recoveries were even higher, because of the difference in the boundary layer thickness, from small scale to full scale. NACA ducts are efficient and have little or no drag. For the best results the NACA duct should be located where the boundary layer is the thinnest. That will normally be near the front of a surface it is placed in, as the boundary layer will become thicker, the further it travels over a surface.  The engineers that designed the NACA duct, wanted the lowest drag with the highest pressure recovery they could possibly get.
 Velocity ratio is the amount of air that can flow through a given size at free stream speeds as opposed to actual real world, operational airflow through a similar size. The NACA duct forms a vortex on each wall that will force more outside air to flow into the inlet than normally would, generating a small ram effect. The two vortices improve the pressure recovery and improve air flow through the duct. The vortices are formed and work best when the air flow through the duct is at a continuous high speed, at lower inlet air speeds the vortices will break down and the duct will lose part of its pressure recovery. If a NACA submerged inlet were used to supply air to the engine intake manifold, it would be more suited to a situation like LSR than most other forms of racing, because the throttle is normally wide open or close to it the entire run, keeping the air flowing through the diffuser, preventing the vortice structures from breaking down. In racing where the driver is continuously on and off the throttle starting and stopping the airflow, the vortices would never get a chance to form and start there ram effect.  A NACA submerged duct with curved convergence walls will have a 20% to 30% higher pressure recovery than a NACA duct with straight parallel convergent walls and have velocity ratios 20% to 50% higher also, due to the twin vortices functioning more efficiently with the curved convergence walls than with the straight parallel convergent walls. If you look at the curved convergence wall NACA duct closely, you will notice that the walls follow the contour of a vortice, that is why they are more efficient, they promote the vortice structure and increase the efficiency of it. The air flowing over the top and down the walls of the NACA duct is what forms the vortices, so all the corners should be left sharp to promote the creation and structure of the vortices. The early engineers didn’t realize that the vortices were being formed inside the duct, they found out by accident, after the straight walls were reshaped to follow the airflow streamlines along the sides of the duct, thus giving us the curved convergence walls we all know today.
The NACA duct has an advantage over other types of inlets for certain inlet and airflow requirements. The placement of the NACA duct is more critical than for other types of inlets because of boundary layer thickness and local air velocity. As the speed of the vehicle increases boundary layer thickness will increase also, decreasing the effectiveness of the NACA inlet. A lot of thought should go into where to place the NACA inlet beforehand. A well designed NACA inlet placed at an ideal location had a pressure recovery of .97 at a speed of 245MPH and a pressure recovery of .95 at a speed of 400MPH. The small differences between the pressure recoveries were due to the boundary layer thickness increasing with the speed, reducing the efficiency of the twin counter rotating vortices. A NACA duct has good pressure recovery when used in a region of low air velocity and a thin boundary layer. NACA ducts are best for systems that need only a small amount of air flow diffusion. Pressure recovery will be better if placed in a thick boundary layer but the velocity ratio will decrease. NACA ducts have reduced internal ducting and fewer bends, with less weight and also have a large reduction in drag compared to conventional external scoops. The divergence of the walls of the duct gradually lets the air expand and reduce the chance of separation. NACA ducts do not have as good a pressure recovery or velocity ratio as normal external style scoops, which can have a velocity ratio of over 100 percent, creating a positive pressure in the scoop and diffuser system, having a ram effect, and thereby increasing power at high speeds. The NACA duct external drag can actually be a negative number in certain situations, because of the removal of the boundary layer behind the inlet. 
The NACA duct should have a width to depth ratio of between 3 and 5. That is the ratio of duct entrance width to entrance height. The NACA duct should have a ramp angle of between 5deg and 7deg. If the duct entrance ramp angles become too steep, the airflow can separate from the surface of the ramp and reduce airflow through the duct dramatically. The ramp angle can be up to 10deg without a significant pressure loss, less wall divergence should be used for smaller ramp angles. Ramp angles of 15deg or more will result in a large loss to the pressure recovery characteristics, a steeper ramp will make a shorter duct but as ramp angle increases efficiency decreases. All corners and edges should be left sharp to aid in the generation and structure of the vortices. The curved divergent ramp walls improve the pressure recovery, because the twin counter rotating vortices function more efficiently than a straight parallel wall NACA duct. The counter rotating vortices generated on each side of the ramp, force the extra airflow down the center of the ramp between the 2 vortices and into the entrance. About 50% of the body of the twin vortices will extend out of the NACA duct, above the surface of the vehicle, through the boundary layer and pull high energy air from above the boundary layer and into the duct. The pressure recovery at the end of the ramp and at the duct entrance can be over 90 percent after diffusion, in ideal situations. Pressure recovery can be improved with higher ramp angles up to 10deg and the divergent walls result in reduced pressure loss. The pressure loss at the ramp inlet will be dependent on the pressure along the ramp and the thickness of the boundary layer the submerged inlet is placed in. If turbulent airflow is allowed to flow into the submerged ramp, it will increase drag and cause pressure recovery to be reduced due to interfering with the formation of the vortices and causing the airflow to become detached from the ramp.
The best pressure recoveries will be with curved divergence walls. Wall divergence is the ratio of the width of the ramp entrance to the width of the submerged entrance.  The divergent walls reduce the pressure loss by the air entering the duct and reduce boundary layer air flow into the duct. Thicker boundary layers reduce the pressure recovery at the submerged entrance, because it will reduce the efficiency of the twin counter rotating vortices, if the boundary layer is sufficiently thick enough so that the vortices will not be able reach above the top of the boundary layer, it would reduce pressure recovery and efficiency by as much as 30%. Pressure recovery will be increased if deflectors are used, regardless of boundary layer thickness. The use of the deflectors will increase the drag somewhat, and the height of the deflectors should be in accordance to the height of the boundary layer they are placed in. With the use of deflectors placed on top of the walls and the use of curved divergence, pressure recovery will be increased. The difference in pressure recovery with duct angle of attack is small. When the duct is located in a curved surface, deflectors will be necessary.
The lip is the shaped protrusion above entrance at the end of the submerged ramp. The shape of the lip must give a high speed at low inlet velocities. The lip should have an airfoil shape. When increasing duct divergence it will increase the angle of attack of the lip and by increasing ramp angle decreases the angle of attack of the lip. Adding curvature to the inner surface of the lip improves flow characteristics, adding curvature to the outer surface only will increase flow loss, by adding curvature to the inside and the outside of the lip surface will benefit flow the most. The lip should have a radius to prevent flow separation. Increasing camber and nose radius will have the best effect on flow. By submerging the lip below the surface in which the entrance is placed in, will be more efficient, but if the lip protrudes slightly above the surface the effect is not harmful. Pressure distribution over the lip will be changed drastically if the ramp angle is changed.
The entrance is the opening at the end of the ramp and is below the body surface, it is the area between the ramp floor and the lip. A square entrance has greater pressure loss than a rectangle entrance; hence the pressure recovery of a square entrance is less than that of a rectangle opening. Because the rectangle entrance is more efficient, that is why we use the width to depth ratio of between 3 and 5. The rectangle air entrance is more efficient because being flat and wide lets the twin vortices function more freely and gives more room between them, in the ramp, to ram the airflow between them and into the entrance at the base of the ramp. The rectangle entrance being more shallow and wider, will allow the body of the vortices to extend above the surface of the vehicle further and hopefully through the boundary layer.  The square entrance will not let the bodies of the vortices stick out above the vehicle surface as far and thus loose part of their ram affect due to the ramp being a deeper depth. If the NACA duct is placed in a situation where the airflow is ramming into the entrance due to circumstances outside, efficiency can be increased with Corner baffles, small triangles added to the top corners above the entrance, where the top of the wall meets the entrance.
The NACA submerged inlet and the rear diffuser under the rear of a lot of race cars, function in almost the same way and have a lot in common. The twin counter rotating vortices inside the diffuser ramp and having the peak negative pressure at the entrance to the ramp from the vehicle surface, the peak pressures will follow the entrance to the ramp as the ramp angles are changed and it moves the ramp entrance locations.
The NACA duct has had a bad rap for a long time by people that were improperly designing it or improperly using it. Designed properly and employed in the correct area and situations, the NACA submerged duct can have pressure recovers and velocity ratios approaching those of a conventional style external scoop with a lot less drag. The reason we do not see NACA submerged inlets in a lot of racing classes is because of their poor performance in on-off accelerator situations, so that would cover pretty much all race classes that involve cornering and braking.
Methods can be employed to further increase the airflow through the NACA duct and through the pipe and ductwork leading from the inlet. That is for another set of post at a later date, as time constraints permit.
 
Constructing a 1000hp turbocharged bellytank

Offline Seldom Seen Slim

  • Nancy and me and the pit bike
  • Administrator
  • Hero Member
  • ***
  • Posts: 13169
  • Nancy -- 201.913 mph record on a production ZX15!
    • Nancy and Jon's personal website.
Re: Exhaust Flow and Aero
« Reply #168 on: November 02, 2011, 04:52:46 PM »
I think we've got a winner -- for the longest-ever single post :-D.  I'd like some input from the rest of the forum folks - - on whether I should sooner or later add this monologue to the landracing.com archive of basic background information.  Is it valid stuff?  I'm not well-versed enough in the field to make the judgement, so will rely upon your (all of you) opinions.  Let me know - and thanks.
Jon E. Wennerberg
 a/k/a Seldom Seen Slim
 Skandia, Michigan
 (that's way up north)
2 Club member x2
Owner of landracing.com

Offline SPARKY

  • Global Moderator
  • Hero Member
  • ***
  • Posts: 6912
Re: Exhaust Flow and Aero
« Reply #169 on: November 02, 2011, 08:49:10 PM »
With what my limited knowledge of NACA ducts is ---I think that this is "MOST likely"  All spot on stuff
Miss LIBERTY,  changing T.K.I.  to noise, dust, rust, BLUE HATS & hopefully not scrap!!

"Security is mostly a superstition. It does not exist in nature, nor do the children of men as a whole experience it. Avoiding danger is no safer in the long run than outright exposure. Life is either a daring adventure or nothing."   Helen Keller

We are going to explore the racing N words NITROUS & NITRO!

Offline Seldom Seen Slim

  • Nancy and me and the pit bike
  • Administrator
  • Hero Member
  • ***
  • Posts: 13169
  • Nancy -- 201.913 mph record on a production ZX15!
    • Nancy and Jon's personal website.
Re: Exhaust Flow and Aero
« Reply #170 on: November 03, 2011, 10:27:08 AM »
Thanks, Sparky, and thanks to the others that told me that this stuff appears to be straight scoop (pardon the pun).  I'll get it into the archives for one and all to peruse whenever.
Jon E. Wennerberg
 a/k/a Seldom Seen Slim
 Skandia, Michigan
 (that's way up north)
2 Club member x2
Owner of landracing.com

Offline Elmo Rodge

  • Hero Member
  • *****
  • Posts: 1654
Re: Exhaust Flow and Aero
« Reply #171 on: November 03, 2011, 10:42:10 AM »
Superford, do you intend to give any hints as to sizing for engine requirements for example? Wayno

Offline SPARKY

  • Global Moderator
  • Hero Member
  • ***
  • Posts: 6912
Re: Exhaust Flow and Aero
« Reply #172 on: November 03, 2011, 11:17:23 AM »
ah these intrepid LSR racers  :evil: and ( OUR) quest to cheat Mother Nature out of her DUE  :cry:
« Last Edit: November 03, 2011, 02:27:31 PM by SPARKY »
Miss LIBERTY,  changing T.K.I.  to noise, dust, rust, BLUE HATS & hopefully not scrap!!

"Security is mostly a superstition. It does not exist in nature, nor do the children of men as a whole experience it. Avoiding danger is no safer in the long run than outright exposure. Life is either a daring adventure or nothing."   Helen Keller

We are going to explore the racing N words NITROUS & NITRO!

Offline WOODY@DDLLC

  • Hero Member
  • *****
  • Posts: 1807
  • ECTA made it to AR-Kansas!
    • Design Dreams, LLC
Re: Exhaust Flow and Aero
« Reply #173 on: November 03, 2011, 12:19:01 PM »
I hope SuperFord317 took a breath in the middle. A good dissertation though. All these comments and posts are pieces of the puzzle!

NACA ducts are typically miss-applied more often than not. Any hole will let air in it's just how efficiently will it work?

Here is another dissertation with some history that will give you another good overview:

http://www.flyingmag.com/scoop-naca-scoop

As for sizing here is another: [I have one but it's not for public use, sorry! :-(]

http://www.melmoth2.com/texts/NACA%20inlet%20sizing.htm

Just connecting a round stub to a NACA duct has some issues.
All models are wrong, but some are useful! G.E. Box (1967) www.designdreams.biz

Offline hitz

  • Sr. Member
  • ****
  • Posts: 332
Re: Exhaust Flow and Aero
« Reply #174 on: November 03, 2011, 12:30:03 PM »
These are by far the best definitions on NACA submerged ducts that I have seen. Superford has a knack of of drawing a good picture with words. I gathered as much information as I could find before I built the one on my lakester and used an on line drawing to scale it. I didn't really have a clue how it really worked even after I gave it the back yard tuft and weedblower test. I could see that it did work though.I just dump it in to a large air air box without the engine running. On the low blower setting every was really smooth except the tufts in the ramp had a different slightly angled pattern. It showed no turbulence anywhere. On the hi blower speed the tufts in the ramp tightened up and angled a little more toward the center at  the air box end and after a few seconds the tufts at the back of the duct (on the outside of the body skin) really tightened up. I took this to mean that without the engine running the air box had built pressure and was dumping air back outside. Somehow it did that without any sign of turbulence. I didn't know why the tufts angled in the throat but thanks to Superford's reference to the vortexes involved in the ducts I think I can "see the air".

"The proof of the puddin" will be during a run when I can incorporate the extra MAP sensor I've installed to data record pressure in the air box.

Thanks to Superford for all his work, to Slim for all the info he's made available here on this forum and to all the entries the members have posted.

hitz

Offline superford317

  • Full Member
  • ***
  • Posts: 67
Re: Exhaust Flow and Aero
« Reply #175 on: November 03, 2011, 01:56:17 PM »
As can be seen, I hate short yes or no answers :-D.
If I answer something it will be with a full explanation, with the how’s and whys presented in a way that can be easily understood.
Currently i am working on 5 different posts for landracing at the same time and I hope to post more of them when I get a little more time, hopefully in a few weeks I will be in Cambodia and will have a lot more time to devote to all of these future posts I have started.
Constructing a 1000hp turbocharged bellytank

Offline Tman

  • Hero Member
  • *****
  • Posts: 3672
Re: Exhaust Flow and Aero
« Reply #176 on: November 03, 2011, 02:20:55 PM »
Holiday in cambodia? :-D

I bet I am one of a handfull that gets that line?
« Last Edit: November 03, 2011, 03:04:24 PM by Tman »

Offline Rex Schimmer

  • Hero Member
  • *****
  • Posts: 2633
  • Only time and money prevent completion!
Re: Exhaust Flow and Aero
« Reply #177 on: November 03, 2011, 02:36:58 PM »
Woody,
Your post are always so thought provoking!!! and you always leave us a little short on what the parameters are and also how we should interpret  them. I assume that the air speed is around 300 mph (440 ft/sec) and from your illustrations the inlet area looks some what larger or very close to equal to the area of the round ducting tube that it is connected to. Looking at the internal velocities it is very interesting that the center line velocity in the duct inlet appears to be zero and that there is an area in the internal duct in which the air has stagnated to zero velocity. Your public awaits your explanation.

A note to all that are following this thread, you need to read both of the attachments that Woody has provided, the first goes over some of the ground that superford posted and also provides some incite as to the proper application of NACA ducts. Because of the nature of land racing vehicles many of the parameters that would govern the use of a NACA duct on an airplane can be avoided as we are really mostly only dealing with air speeds and inlet conditions at or near the vehicles maximum speed. Obviously cars that are very powerful and need to optimize acceleration will have additional needs for an adequately designed engine air inlet.

Rex

 
Rex

Not much matters and the rest doesn't matter at all.

Offline WOODY@DDLLC

  • Hero Member
  • *****
  • Posts: 1807
  • ECTA made it to AR-Kansas!
    • Design Dreams, LLC
Re: Exhaust Flow and Aero
« Reply #178 on: November 03, 2011, 03:25:29 PM »
Woody,
Your post are always so thought provoking!!! and you always leave us a little short on what the parameters are and also how we should interpret  them. I assume that the air speed is around 300 mph (440 ft/sec) and from your illustrations the inlet area looks some what larger or very close to equal to the area of the round ducting tube that it is connected to. Looking at the internal velocities it is very interesting that the center line velocity in the duct inlet appears to be zero and that there is an area in the internal duct in which the air has stagnated to zero velocity. Your public awaits your explanation.
Rex

Rex, the purpose of the posts is to provoke everyone, but thoughtfully!  :-D

We all have our thoughts, experiences and opinions but since much of what we do has interactions it is difficult to find the right combination. Even when we do we can't be sure it's the optimum. What CFD has shown me is that just about everything regarding fluid flow is case specific. That's why we need testing, experiments and nowadays simulation! Like you, when I look at these I wonder why is it doing that? I try to find experimental papers with enough info to model it and see if the model is good enough to use.

The NACA duct shots were a specific mass flow at the round duct (hairdryer inlet) with no external flow just to see what the velocity profiles would look like in the ducts [baseline]. Then we look at it with the exterior flow to see what has changed and what the pressure recovery might be. Beyond that my former Japanese coworkers had a saying when questioned, "Not need know!" (Proprietary!)  :cry:

Now I have to go finish the tire stuff on Buddy's build diary!  :cheers:
All models are wrong, but some are useful! G.E. Box (1967) www.designdreams.biz

Offline Saltfever

  • Full Member
  • ***
  • Posts: 276
Re: Exhaust Flow and Aero
« Reply #179 on: November 04, 2011, 01:55:10 AM »
Woody:
Yeah, ditto what Rex said. I would like to understand what I am seeing. Are the inlet and outlet areas similar? And according to Bernoulli, the zero velocity (blue) areas are the highest pressures, right? The velocities appear to range from 0 mph to about 255mph. The flow appears to be stagnant (highest pressure) in the trough and in front of the inlet but with nominal speeds of about 255mph mid-stream. Then there are other speeds of zero (in the diverging cone area) and 300 mph on the opposite side of that high pressure area. Geeze, a lot is going on in there!  I know Woody, you said CFD is “case specific” and now I finally see the magnitude of what you are saying. You said it was done with a specific mass flow so that would equate to some constant velocity right? The pics we see are at a single speed?

What is interesting about the CFD picture is how it points out how incredibly difficult instrumentation would be. Just look at the widely deviating pressure distribution inside. If you put in only one pressure sensor, and made assumptions from that data point, you would be clueless as to what is really going on! Even a couple of sensors in the wrong spots would be giving you very bad information. That might lead to some really bad decisions on jetting or head-scratching when reading the plugs and looking at scoop pressure in the data log.