TeamSwift

Home of the Suzuki mini-compacts ! Your Home for all things Suzuki Swift, Geo Metro, Holden Barina, Chevy Sprint, Pontiac Firefly, and Suzuki Cultus. TeamSwift is a technical performance oriented community!
It is currently Mon Nov 20, 2017 1:29 am

Underbody braces, turbos and more!

All times are UTC - 5 hours




Post new topic Reply to topic  [ 64 posts ]  Go to page Previous  1, 2, 3  Next
Author Message
 Post subject:
PostPosted: Sun Oct 07, 2007 3:18 am 
Offline
User avatar

Joined: Sat Jun 03, 2006 11:54 pm
Posts: 673
Location: Illinois
Scientifically, bumble bees are supposed to not be able to fly. But yet they can....

_________________
If it ain't broke, fix it until it is. LOL.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Oct 07, 2007 7:14 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
I didn't really want to get into Reynolds numbers and other aerodynamic esoterica because I thought they would be beyond most of the readers here.

A textured surface helps flow adhesion around outside curves. Which is the reason that a golf ball with dimples flies further than one without. It turns out that a golf ball covered with sandpaper will fly even slightly further than one with dimples. Increased flow adhesion around the outside curves means that as the port bends down toward the valve the flow is a little more likely adhere to the lower wall of the port which will help overall flow velocity.

I'm quite familiar with low Reynolds numbers since I used to design my own R/C airplanes. With a small chord and a low speed, Reynolds numbers are quite low in model planes. Turbulator strips on the leading edges of model airplane wings are quite common. On the faster and more aerodynamically efficient model planes like a pylon racer or an F5B though a mirror like finish on the model is desired.

On those low speed models that use turbulators, the turbulator is placed on the upper side of the wing leading edge, in order to increase flow adhesion over the upper side of the wing which enhances lift.

And actually I do know the difference between porting and polishing. The image I showed was just the first one I could find in a cursory search. I would have posted some of my own but I lost all my hundreds (thousands?) of car magazines and photos in a house fire about twenty years ago. I haven't been playing with cars to any great extent since because I had kids and other responsibilities. Model airplanes are just as techy and cost a whole lot less than hot rods.

And in response to a couple of other comments: JBweld is tough but not as tough as metal. A small chunk entering the combustion chamber is unlikely to cause enough damage to harm the efficiency of the engine.

Plus if you have ever tried to remove properly applied jbweld you'll know that it is extremely tenacious. I've literally glued motorcycle engine cases back together with it.

And the point about the bumble bee being unable to fly was made a long time ago, before it was known that at insect scales the air acts in a far more viscous manner than it does at our human scale.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Oct 07, 2007 10:44 am 
Offline
Suzuki Elder
User avatar

Joined: Mon Dec 01, 2003 12:47 pm
Posts: 11674
Location: columbus, ohio
http://www.efm.leeds.ac.uk/CIVE/CIVE140 ... bulent.htm

http://www.exploratorium.edu/xref/pheno ... ffect.html

poiseuille's law, which basically describes flowing fluids under conditions with smooth, laminar flow that allows the calculation of volume flowrate demonstrating radius dependence and length dependence. it involves viscous resistance and is illustrated by comparisons of resistance to flow and resistance in series vessels.

the introduction of discussions of dimples on golf balls increasing lift and aerodynamic effects of airplane wings with a vague reference to bernoulli effect departs from the original and basic concept of flows through a pipe and just muddies the water for the discussion of creating turbulence in an induction system and it's benefit or degradation of air flow. it works well in aeronautical engineering but in general, in engine tuning, minimizing profiles and knife edging leading and trailing edges is much more common to increasing flow. on this point, i will conceed that things like the contours of the arcs of intake manifold runners and passageway profiles in heads are "purpose designed" to achieve flow rates desired by the original engine designers.

in principle, for you to tell people on a car forum that scarring up any part of their intake tract is in any way beneficial without your doing it first and objectively testing it to prove that it will indeed "improve mileage" or "increase efficiency" is not a particularly responsible action. at the least, you'll get some poor guy to spend his time to do it and worse, you could entice him to decrease the efficiency of his engine and cost him mileage.

for the sake of debate, i'll agree that one can usually find some design abberation where a specific introduction of perturbation of air flow will be of benefit in an overall system but that is rare (or it would be in wide practice) and is always the result of well documented experimentation. most of the time, those breakthroughs are wholly owned intellectual property and are rarely achieved by a guy in his garage. more to the point, the guy in the garage will notice a specific quality in the design of a car part he just reverse engineered with a hammer and try to incorporate it into his system.

i wish i had just a penny for every "increase your mileage with my special 'vortex' adapter" scam that was run on somebody using an out of context scientific principle. i urge any teamswifters who are following this thread who don't have a sound basis in physics and engineering - do not scar up your intake manifolds, heads, or valves and expect your engine to run better or get better mileage. i'd also resist the urge to coat passages with a material that can spall off and collect in the cylinder. that can only end up badly.

as always, the scientific proof dispels all bullshit. show me proof with pics and data that any application you have illustrated benefits a suzuki engine in either gain of averaged horsepower or fuel efficiency and i'll eat my hat. the last time anyone cared to do a scientific assessment of flows in the head of a suzuki twincam, they documented everything with cast foam plugs, drawings, measurements, and photos. that was a long time ago but i don't recall anything but shiny polished surfaces.

and the debate rages on....... :lol:

_________________
1991 Blue Geo Metro Convertible highly modified 1.0L Turbo3 5 spd. - 1991 Red Geo Metro Convertible customized with a Twincam 5 spd.

My Turbo3 Project
My Cardomain Page -Ol' Blue
My YouTube Channel
My Photo Gallery
SAAB Sonett II


Top
 Profile  
 
 Post subject:
PostPosted: Sun Oct 07, 2007 12:53 pm 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
If you think you are hurting my feelings by disagreeing with me you couldn't be more wrong. If you look at my OP on this thread you will see that I specifically welcomed disagreement in order that I may improve my understanding of these things.

The fact that I made that statement should lead the intelligent reader to understand that I'm not totally sure of myself, which I'm not. I'm throwing out ideas in the hope of stimulating discussion.

Furthermore consider that Einstein had little experimental evidence when he developed the theory of relativity, instead Einstein depended on "thought experiments" to advance his understanding of the principles of relativity.

Apparently you also missed the post which had this image of a roughened intake port that also stated it was being done with some success.

Image

Everyone seems be of the opinion that the Bernoulli principle completely explains the aerodynamic lift on a wing. This is not so, the Coanda effect has quite a large input on aerodynamic lift, particularly so when high lift devices such as flaps and leading edge slats are deployed.

http://en.wikipedia.org/wiki/Coanda_effect

http://jef.raskincenter.org/published/coanda_effect.html


The Coanda effect is ridiculously easy to demonstrate. Take a spoon and put the bottom of the spoon in a smoothly flowing stream of water from a tap and the spoon will be pulled toward the water flow.

Image

It is the Coanda effect which makes the fluid flow in the intake port adhere to the outside curves of the port such as the curve immediately before the valve seat.

I have a good bit more to say on this issue but I'm busy today, I'll probably log on again tonight.

Peace,

Jon


Top
 Profile  
 
 Post subject:
PostPosted: Sun Oct 07, 2007 3:38 pm 
Offline

Joined: Fri Jun 03, 2005 4:36 pm
Posts: 493
Location: Roscommon, MI
Let's put dimples on the spoon and try again. Or maybe rough it up with some sandpaper. Then we can compare the aerodynamic downforce created by a Top Fuel dragster at speed.

I think we're obfuscating the discussion a bit. Or maybe I just don't see the relevance of spoons to intake porting.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Oct 07, 2007 5:07 pm 
Offline
Suzuki Elder
User avatar

Joined: Mon Dec 01, 2003 12:47 pm
Posts: 11674
Location: columbus, ohio
Soliton wrote:
If you think you are hurting my feelings by disagreeing with me you couldn't be more wrong. If you look at my OP on this thread you will see that I specifically welcomed disagreement in order that I may improve my understanding of these things.


i don't think anyone's feeling are getting hurt and it's a wonderful thing. ask anyone here and they'd probably tell you that i love a wonderfully esoteric debate and can beat a dead horse into a silk sheet.

now, we're getting more people posting in the thread.

rarson, that's what i was going to say, too. :lol: any flow outside of a confined area takes on a whole new set of physics which is why golf balls, airplane wings, bumble bees, and spoons under a faucet have done squat to deter me from the main point of my debate - scarring up a pipe wall is counterproductive to aiding flow efficiency.

_________________
1991 Blue Geo Metro Convertible highly modified 1.0L Turbo3 5 spd. - 1991 Red Geo Metro Convertible customized with a Twincam 5 spd.

My Turbo3 Project
My Cardomain Page -Ol' Blue
My YouTube Channel
My Photo Gallery
SAAB Sonett II


Top
 Profile  
 
 Post subject:
PostPosted: Sun Oct 07, 2007 6:55 pm 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
I note that no one has responded to the point regarding the roughened intake port of which I reposted the photo from the earlier part of the thread.

Image

And people still don't seem to understand that an engine optimized for low rpm torque and efficiency will not be tuned the same as for high rpm torque and horsepower.

An engine optimized for horsepower will be run somewhat rich, so as to utilize all the available oxygen in the intake charge.

An engine optimized for efficiency will be run somewhat lean, so as to utilize all the available fuel in the intake charge.

There are quite a few other parameters that differ considerably from a horsepower engine to a fuel efficient engine.

The laws of physics, specifically aerodynamics, do not change simply because the fluid flow is within a constricted area. That's why Bernoulli's principle and the Coanda effect are relevant to this discussion.

Which also means that dimpled golf balls are also relevant to the discussion since they are an example of enhancing the Coanda effect through surface texturing.

One more thing: The only points that I have noticed disagreement on is the intake porting, the ignition and the combustion chamber coatings.

Is there anything else I have mentioned with which anyone disagrees?


Top
 Profile  
 
 Post subject:
PostPosted: Sun Oct 07, 2007 11:24 pm 
Offline

Joined: Sun Aug 20, 2006 11:31 pm
Posts: 57
Location: Lummi Island
>[quote="Soliton"]I note that no one has responded to the point regarding >the roughened intake port of which I reposted the photo from the earlier >part of the thread.

Actually I incorporated that into my comment about low reynolds number flow benefiting from a rough surface. That texture looks like it would be effective. I would like to see what the actual flow looks like at 1500, 2500 and 3500 rpms.


>And people still don't seem to understand that an engine optimized for >low rpm torque and efficiency will not be tuned the same as for high rpm >torque and horsepower.

I was also referring to this concept, since a low rpm intake translates to low reynolds numbers. If we are looking for efficiency, we won't be turning high rpms and will be optimizing the low end.


>The laws of physics, specifically aerodynamics, do not change simply >because the fluid flow is within a constricted area. That's why Bernoulli's >principle and the Coanda effect are relevant to this discussion.

Relevant, but only in conveying the concept of low speed air flow vs. high speed air flow. We are definitely operating in the realm of rough surfaces, but the question is how rough? I don't think any harm will come from creating a mild surface texture, but I certainly wouldn't coat anything with JB weld!!! Thats asking for trouble... Also, flow through a tube is significantly different that over a free flowing surface.

>Which also means that dimpled golf balls are also relevant to the >discussion since they are an example of enhancing the Coanda effect >through surface texturing.

I would agree that his conveys the concept well, but don't let that suggest that dimples are the solution. Rough sanding or glass beading may be a more appropriate texture.

>One more thing: The only points that I have noticed disagreement on is >the intake porting, the ignition and the combustion chamber coatings.

I'm not sure that was disagreement on porting, maybe more of a clarification between port and polish for high rpm, high flow, high HP applications and rough surfaces or some sort of turbulation for low rpm, high efficiency applications. Hopefully some of these things will be experimented with. I plan on Singh grooves the first time my head comes off and I will probably bead blast the intake, but I am also working on a special carb.....

MM


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 12:30 am 
Offline

Joined: Fri Jun 03, 2005 4:36 pm
Posts: 493
Location: Roscommon, MI
Soliton wrote:
I note that no one has responded to the point regarding the roughened intake port of which I reposted the photo from the earlier part of the thread.


What exactly are you looking for? I saw the picture the first two times you posted it. It's great that someone claims to have this working, but it's not first-hand information coming from you.

Soliton wrote:
And people still don't seem to understand that an engine optimized for low rpm torque and efficiency will not be tuned the same as for high rpm torque and horsepower.


Honestly, I feel you aren't paying attention. No one ever said anything contrary to this at all.

Soliton wrote:
An engine optimized for horsepower will be run somewhat rich, so as to utilize all the available oxygen in the intake charge.


Wrong. An engine optimized for horsepower will run somewhat rich to prevent detonation. Excess fuel is added for a cooling effect, which is why you can simply add water (with water injection) instead and remove some fuel for the same effect. Anything less than stoic is doing squat for "utilizing oxygen" because there's already more fuel available than ideal. There is absolutely no way a cylinder can suck in air and not "use" it (unless you're talking about a Miller cycle or Atkinson cycle engine), perhaps I'm misunderstanding you. If anything, it's the fuel that is going unused.

Soliton wrote:
An engine optimized for efficiency will be run somewhat lean, so as to utilize all the available fuel in the intake charge.


An engine optimized for efficiency runs leaner because it can. It's not pushing a setup to the ragged edge, so detonation, even if it occurs, is less of a problem. Performance is all about making the engine consume more air, while efficiency is about making it consume less fuel. The difference in tuning comes about because of the difference in the amount of air the engine is consuming.

I might also play devil's advocate and mention that detonation is less likely to occur at high rpms, which is one of the reasons many people tend to lean their fuel out as rpms increase, in a power-optimized setup. But the fact that the engine is also consuming more air at those rpms means it's still using more fuel (which serves to illustrate the point that the goal of tuning for efficiency isn't to *tune the car leaner*, but to get the car to *use less fuel*).

Soliton wrote:
There are quite a few other parameters that differ considerably from a horsepower engine to a fuel efficient engine.


Again, I'm not really sure what your point is. We know this. You are the one that brought up performance in the first place; the rest of us are talking about the efficiency merits of the particular intake port mod you posted.

Soliton wrote:
The laws of physics, specifically aerodynamics, do not change simply because the fluid flow is within a constricted area. That's why Bernoulli's principle and the Coanda effect are relevant to this discussion.


I'm not going to go into the whole closed system vs. open system discussion, because I think t3 is a lot more qualified to comment on that. However, from the Wikipedia article you posted about the Coanda effect:

Wikipedia wrote:
The Coandă effect (IPA: ['kwandə]), also known as "boundary layer attachment", is the tendency of a stream of fluid to stay attached to a convex surface, rather than follow a straight line in its original direction.


If this effect deals strictly with convex surfaces, then I hope you can see my questioning of it's relevance. How much of the intake port is a convex surface? The port floor maybe? I don't know, it's been almost a decade since I had a G10 engine to look at.

I'm having a hard time understanding how increasing headloss could increase efficiency. The only thing I can think is that better mixing of the air and fuel would create a more efficient burn and might decrease the amount of fuel required to create the same power. But that's more to do with air/fuel mixing than some kind of fluid dynamics effect. Increasing headloss is also going to add heat to the fluid due to the increased friction. If the head is hot, the textured surface will create more surface area to transfer heat into the fluid, etc. etc. etc. I question whether actually threading the ports might be taking the "rough wall" concept a bit too far. If you create (speaking relatively, of course) huge chasms in the port, I would think that would likely also cause a hindrance to the flow of fuel.

I'm no expert by any means and I'm probably a bit more than rusty on the little fluid dynamics that I used to know, so feel free to correct me, but I have to agree with t3 that it's just not making sense to me. And I still maintain that a mirror-like polish is a bad idea for the runners of a carbureted setup due to the increased likelihood of fuel sticking and forming large drops on the runner walls, in either performance or economy setups (by the way, I'm a fuel injection guy so I'm definitely no carb expert).

It's all well and good to have theoretical discussions about this stuff, but when the argument is indecisive, it's nice to have some actual real-world results to look upon. That's why I suggested that maybe talking about spoons and golf balls was getting a little off the topic of intake porting. Furthermore, I would have to assume from a simplistic standpoint that sucking air through a pipe seems vastly different than pushing a wing through the air.

Oh, and I hope I wasn't coming off as arrogant or anything with my comment, I was just trying to make a point with a little humor in the mix. Tongue-in-cheek, that's all. :)


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 5:10 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
I keep forgetting about you 3play, I appreciate your support. :)


I'm not so sure that it's the volume of flow that counts as much as it is the velocity and resulting more efficient mixing of the fuel and air. I see two mechanisms by which the increased flow velocity will help with mixing the air and fuel.

First, the increased velocity will give the fuel droplets less time in the port in which to combine together into larger droplets. And second, the increased port velocity will lead to more turbulence in the combustion chamber.

Adherence of a turbulent layer to the port wall will increase the velocity of the stream passing down the center of the port. A rough texture will help this adherence. Thus the Reynolds number will be increased, if even only slightly.

Rather than jbweld, use the formagasket that stays (nominally) pliable. Smear a thin layer on the inside of the port and texture with something like an old toothbrush. If some does come off and go into the engine it isn't going to cause any damage since it is less hard than jbweld.

For many years I specialized in repairing the unrepairable, both electronic and electromechanical . Since I charged dearly for my services my clients only used me as a last resort when replacement parts were unavailable. To do this I had to have not only a good knowledge of how things work but also a knowledge of how to modify something to make it work after it had stopped working.

Working on a special carb?

Peace,

Jon


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 5:28 am 
Offline
User avatar

Joined: Sat Jun 03, 2006 11:54 pm
Posts: 673
Location: Illinois
Here's a better article about the Powre Lynz.

http://www.allpar.com/fix/holler/performance-and-economy.html

If any of you care to talk to him, he's MPG Mike at MPG Research.

http://www.mpgresearch.com/profile.php?mode=viewprofile&u=5

_________________
If it ain't broke, fix it until it is. LOL.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 6:04 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
rarson wrote:

Quote:
What exactly are you looking for? I saw the picture the first two times you posted it. It's great that someone claims to have this working, but it's not first-hand information coming from you.


So no one is allowed to post information unless they have personally developed that information?


Quote:
Honestly, I feel you aren't paying attention. No one ever said anything contrary to this at all.


People keep talking about flow volume, I'm talking about flow velocity. Flow volume is great for maximum horsepower, flow velocity is what counts for maximum efficiency. I pointed out why this might be so in my last post.

Quote:
Wrong. An engine optimized for horsepower will run somewhat rich to prevent detonation. Excess fuel is added for a cooling effect, which is why you can simply add water (with water injection) instead and remove some fuel for the same effect. Anything less than stoic is doing squat for "utilizing oxygen" because there's already more fuel available than ideal. There is absolutely no way a cylinder can suck in air and not "use" it (unless you're talking about a Miller cycle or Atkinson cycle engine), perhaps I'm misunderstanding you. If anything, it's the fuel that is going unused.


How are you going to get an engine running open loop to run exactly stoic?

I didn't mention the cooling effect since it wasn't relevant to the discussion.

I had a water injection unit made by either Holley or Edelbrock on a 327/350 hp Corvette engine at least thirty five years ago.

WWII fighters often had water/alcohol injection to be run at War Emergency Power, it's far from a new technology.


Quote:
An engine optimized for efficiency runs leaner because it can. It's not pushing a setup to the ragged edge, so detonation, even if it occurs, is less of a problem. Performance is all about making the engine consume more air, while efficiency is about making it consume less fuel. The difference in tuning comes about because of the difference in the amount of air the engine is consuming.


Leaner mixture = less fuel consumed.

Quote:
I might also play devil's advocate and mention that detonation is less likely to occur at high rpms, which is one of the reasons many people tend to lean their fuel out as rpms increase, in a power-optimized setup. But the fact that the engine is also consuming more air at those rpms means it's still using more fuel (which serves to illustrate the point that the goal of tuning for efficiency isn't to *tune the car leaner*, but to get the car to *use less fuel*).


But high rpm detonation is far more damaging than detonation at low rpms.

Once again, when the engine is running open loop how do you determine the exact stoic mixture? Running lean at high rpm/high load will destroy an engine quite rapidly. One of the reasons that ensuring adequate fuel flow is so important with a nitrous setup.


Quote:
Again, I'm not really sure what your point is. We know this. You are the one that brought up performance in the first place; the rest of us are talking about the efficiency merits of the particular intake port mod you posted.


It is the constant harping on flow volume, which is what is required for high rpm torque and horsepower.



Quote:
I'm not going to go into the whole closed system vs. open system discussion, because I think t3 is a lot more qualified to comment on that. However, from the Wikipedia article you posted about the Coanda effect:



I already mention that about convex surfaces and pointed out where there was indeed a convex surface in the port.

Quote:
I'm having a hard time understanding how increasing headloss could increase efficiency. The only thing I can think is that better mixing of the air and fuel would create a more efficient burn and might decrease the amount of fuel required to create the same power. But that's more to do with air/fuel mixing than some kind of fluid dynamics effect. Increasing headloss is also going to add heat to the fluid due to the increased friction. If the head is hot, the textured surface will create more surface area to transfer heat into the fluid, etc. etc. etc. I question whether actually threading the ports might be taking the "rough wall" concept a bit too far. If you create (speaking relatively, of course) huge chasms in the port, I would think that would likely also cause a hindrance to the flow of fuel.


The overall efficiency of the port is *raised* not lowered at low flow rates by texturing. By creating a thicker turbulent boundary layer, one increases port velocity and reduces fluid friction at the same time. At low Reynolds numbers the boundary effect becomes more important, the slower the fluid flow, the lower the Reynolds number.

Increased port velocity leads to a better ram effect at low rpms, which leads to better cylinder filling.

Engines specifically designed for low rpm torque have long, small diameter intake ports. What you are trying to do in the case we are discussing here is to try and simulate that smaller port diameter in an engine already designed for higher rpm power.

Quote:
I'm no expert by any means and I'm probably a bit more than rusty on the little fluid dynamics that I used to know, so feel free to correct me, but I have to agree with t3 that it's just not making sense to me. And I still maintain that a mirror-like polish is a bad idea for the runners of a carbureted setup due to the increased likelihood of fuel sticking and forming large drops on the runner walls, in either performance or economy setups (by the way, I'm a fuel injection guy so I'm definitely no carb expert).


Good carbs, such as a Weber, do quite a good job of atomizing fuel at high flow rates. At high flow rates the fuel just doesn't have time to stick to much of anything, it is swept along by the air flow.

My brother even has a Holley carb on an LT1 engine that has special metering blocks which use Weber jets and emulsion tubes. You can actually change the main jets with the engine running.

I'm really not sure that a mirrored port finish is the best setup, but I know that it has been done even since the time of side valve engines in autos.

Quote:
It's all well and good to have theoretical discussions about this stuff, but when the argument is indecisive, it's nice to have some actual real-world results to look upon. That's why I suggested that maybe talking about spoons and golf balls was getting a little off the topic of intake porting. Furthermore, I would have to assume from a simplistic standpoint that sucking air through a pipe seems vastly different than pushing a wing through the air.


Of course there are differences but the physics is a constant, it is the application of those physical rules which changes.

In the one case, the air moves past an aerodynamic surface and in the other the surface moves past the air.

Quote:
Oh, and I hope I wasn't coming off as arrogant or anything with my comment, I was just trying to make a point with a little humor in the mix. Tongue-in-cheek, that's all. :)


No offense taken, I have an awfully dry sense of humor myself. I'm refraining from that here because I'm a n00b and y'all don't know my ways.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 6:28 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
From the article posted by crazymanz:

Quote:
Normal street trim involved an electronic ignition conversion with an MSD box, OD 4-speed out of a ’75 Dart Sport, and a screen under the L-6 2-bbl carburetor. At first, I thought the screen should just be flat. Then somebody showed me a trick. Take a piece of the screen and something aerosol like Brakleen. Spray the Brakl.een directly at the screen. As you do, you will see the spray come out of the can and go past the screen. Slowly angle the screen away from the can. At just the precise angle the spray will disappear from behind the screen. At this precise angle, the screen will mostly vaporize the Brakleen as it passes through.


From the Wikipedia article on the Coanda effect:

Quote:
An important practical use of the Coandă effect is for inclined hydropower screens, which separate debris, fish, etc., otherwise in the input flow to the turbines. Due to the slope, the debris falls from the screens without mechanical clearing, and due to the wires of the screen optimising the Coandă effect, the water flows though the screen to the penstock leading the water to the turbine.


As you can see, the Coanda effect has something to do with the increased atomization of the fuel.

Great article by the way, thanks for posting.

It's interesting to note that the slant six was already designed with low rpm torque in mind, long, small diameter intake ports, small valves, etc.

BTW, my brother has a good metal lathe and could make those special taps quite easily. The steel could then be case hardened (carburized) to make it resistant to wear. He also has a kiln in which to do the carburizing, he uses the kiln to slump glass for custom furniture.

Peace,

Jon


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 6:38 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
Oh, I just thought I'd mention that I ran my first tank of gas through my 93 Swift GT. 36.4 mpg with my depleted uranium toenails. :D

A moderate amount of backroads driving where I managed to largely take it easy and stay in fifth gear most of the time.

A lot more interstate driving between 70 to 80 mph with occasional bursts to 85 depending on traffic.

I did chase a pack of crotch rockets for a couple of miles at 90 plus too. :D

With my aggressive interstate driving style (I merge in third gear and let it wind) that is truly excellent gas mileage.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 9:59 am 
Offline
Suzuki Elder
User avatar

Joined: Mon Dec 01, 2003 12:47 pm
Posts: 11674
Location: columbus, ohio
Soliton wrote:
I note that no one has responded to the point regarding the roughened intake port of which I reposted the photo from the earlier part of the thread.

And people still don't seem to understand that an engine optimized for low rpm torque and efficiency will not be tuned the same as for high rpm torque and horsepower.


the g13 twincam produces most of it's useable torque in the higher rpm range. at low rpm it's useable grunt is pretty small.

optimizing low rpm torque on a twincam might not be the way to go, eh? when it comes to the 3 banger, if you don't work with the bigger end you may as well just walk to work. :lol:

3 play and rarson, i appreciate your participation. i already conceded the places where bernoulli and coanda were engineered into induction systems and port runner designs in a post up the thread. i took exception to the misuse of principles in positing synergism in port flow as illustrated by the use of flow cohesion to accelerate flow. flow in that situation is faster in a region only in relationship to the slowing of flow in another region.

c'mon guys, jump into the discussion.

_________________
1991 Blue Geo Metro Convertible highly modified 1.0L Turbo3 5 spd. - 1991 Red Geo Metro Convertible customized with a Twincam 5 spd.

My Turbo3 Project
My Cardomain Page -Ol' Blue
My YouTube Channel
My Photo Gallery
SAAB Sonett II


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 10:46 am 
Offline

Joined: Fri Jun 03, 2005 4:36 pm
Posts: 493
Location: Roscommon, MI
t3 ragtop wrote:
the g13 twincam produces most of it's useable torque in the higher rpm range. at low rpm it's useable grunt is pretty small.


No kidding! I know it's been forever since I've driven a G10, but honestly, I think the G10 has more grunt down low than the G13B. The twin cam feels a lot faster because it pulls pretty well past 5000 rpm (which is where my stock G10 would pretty much fall on its face). Both were automatics, the G10 being my old vert and the twin cam being *your* old GTi. :D The vert just felt a lot "zippier," like a scooter almost, where the GTi built up to its power powerband more steadily. Probably a bit like a Honda D-series versus a Honda B-series.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 11:21 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
My GT will pull away smoothly at 1500 rpm in fifth, which is somewhere around 35 mph. I've been trying to drive it that way on backstreets in order to compensate for my aggressive driving style on the interstate.

I don't know a damn thing about the 1.0 3 cyl, I was really talking in terms of basic principles that could be applied to just about any engine. Of course those principles would have to be tweaked for the individual case.

It is my opinion that texturing the intake port would improve the low end while not hurting the high end very much. I suspect that the other mods I mentioned would offset any losses on the high end that might be due to a textured intake port.

The cold air induction system I mentioned would help the high end also, as would porting and insulating the exhaust system. The ignition also would help the high end as well.

The more I think about it the more I like the idea of applying a thin layer of pliable formagasket and texturing it by dabbing at it with an old toothbrush. It sure would be a lot easier than cutting grooves in the metal of the port wall.

It's been my experience that small engines and automatic transmissions do not go well together. Manual transmissions seem to be far more suited for the normal power characteristics of the small engine.

With a manual tranny you can give the engine a lot of throttle for good acceleration while at the same time short shifting to keep the rpms down, you can't really do that with an automatic, if you give it plenty of throttle the automatic is going to let the engine rev..


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 1:04 pm 
Offline

Joined: Fri Jun 03, 2005 4:36 pm
Posts: 493
Location: Roscommon, MI
Soliton wrote:
So no one is allowed to post information unless they have personally developed that information?


No, but that information, naturally, should be taken with a grain of salt.

Soliton wrote:
People keep talking about flow volume, I'm talking about flow velocity. Flow volume is great for maximum horsepower, flow velocity is what counts for maximum efficiency. I pointed out why this might be so in my last post.


Port volume is directly related to velocity. The smaller the port, the greater the velocity. However, it doesn't matter how much velocity you have, if the port is huge, you're going to be guzzling fuel. I agree that velocity is beneficial in both performance and economy applications (and I actually tend to prefer relatively smaller ports... for instance, if you're comparing a first-generation 4g63 head to a second-generation head, I prefer the 2g head because it's got smaller ports which help fill the cylinders better due to the higher velocity... and this is a turbocharged engine, where velocity is already going to be relatively high. Of course, it helps that you can always enlarge the runners to the larger size if you really wanted to, but I've never felt it necessary because I've gotten great results without touching the head. Cams help!).

Soliton wrote:
How are you going to get an engine running open loop to run exactly stoic?


You are seriously frustrating me with these logic leaps. Where did I ever say you would run the engine stoic? Or even attempt to? Even in an economy setup, you're not going to run stoich. In a power setup, if you COULD run stoich, you wouldn't, because you'd almost certainly make more power at a lower AFR.

Soliton wrote:
I didn't mention the cooling effect since it wasn't relevant to the discussion.


You were talking about tuning fuel. Preventing detonation is a major aspect of engine tuning.

Soliton wrote:
WWII fighters often had water/alcohol injection to be run at War Emergency Power, it's far from a new technology.


I know, but I wasn't commenting on the age of the technology. Internal combustion is pretty damn old, too! :)

Soliton wrote:
Leaner mixture = less fuel consumed.


I know that, you're missing my point. You can lean the mixture out all you want, but if the engine consumes more air, then it's consuming more fuel. AFR is a RATIO, a proportionate amount. Amount of fuel consumed is directly proportional to amount of air consumed. A G10 at 11:1 AFR will still be consuming less fuel than a G13 at 12.5:1.

Soliton wrote:
But high rpm detonation is far more damaging than detonation at low rpms.

Once again, when the engine is running open loop how do you determine the exact stoic mixture? Running lean at high rpm/high load will destroy an engine quite rapidly. One of the reasons that ensuring adequate fuel flow is so important with a nitrous setup.


Excessive detonation at any RPM is a bad thing. The point is, the ability to tune leaner is there because the engine is less likely to detonate at high rpms because there simply isn't enough time. If you're operating the engine at 100% load, it's a lot more taxing on the engine at low rpms than it is at high rpms.

And "tuning leaner" isn't "running stoich," okay? "Tuning leaner at high rpms" is taking an AFR that is 11.5:1 at 4000 rpm and leaning it out to 12.5:1 at 8000 rpm. Well below stoich.

I tend to think that catastrophic engine failure caused by detonation is a myth. I'd say the real culprit is pre-ignition. And on the subject of nitrous, most nitrous catastrophes are caused by one or more of the following:

1. Improper install
2. Spraying at too low an rpm
3. Spaying at the rev limiter

All of which can cause dangerously lean conditions. Naturally, if you tune TOO lean you will run into problems, but this isn't really specific to nitrous only (it's kind of like tuning a boosted rotary... you have to be extra careful!). I think the problem is that you're misunderstanding my idea of "lean" when tuning under full load. I'm talking in relative terms, like the example I mentioned above.

Soliton wrote:
It is the constant harping on flow volume, which is what is required for high rpm torque and horsepower.


Velocity is JUST AS important for making power as volume, if not moreso, and thus, velocity is equally important in a power setup as it is in an economy setup. I'm not arguing for volume. I'm questioning whether you're correct in assuming that what you are doing is actually going to increase the velocity of the air.

Soliton wrote:
I already mention that about convex surfaces and pointed out where there was indeed a convex surface in the port.


"A" surface. That's exactly my point. 90% of the port isn't a convex surface. I just don't see why it's significant to bring it up when it doesn't apply to most of the port.

Soliton wrote:
The overall efficiency of the port is *raised* not lowered at low flow rates by texturing. By creating a thicker turbulent boundary layer, one increases port velocity and reduces fluid friction at the same time.


There's your problem right there. Let me be exceedingly clear: "By creating a thicker turbulent boundary layer, one increases port velocity and reduces fluid friction at the same time." This is what I'm questioning. I don't think you're going to significantly affect velocity by texturing the port, and I think threading isn't the best idea.

Furthermore, increasing turbulence is only going to impede flow, and reduce velocity. t3 already stated this. Increased turbulence is also going to increase heat transfer to the air/fuel mixture.

The entire nature of turbulence makes it inefficient, because you're constantly ripping fluid molecules off the wall of the "pipe" and replacing them with new ones, and throwing the molecules around in a chaotic frenzy. We don't want the molecules going in every possible direction, we want them all going towards the cylinder. Either we're increasing velocity, or we're increasing turbulence. Increasing velocity can increase turbulence, but I just cannot see how the opposite can be true.

Soliton wrote:
Increased port velocity leads to a better ram effect at low rpms, which leads to better cylinder filling.

Engines specifically designed for low rpm torque have long, small diameter intake ports. What you are trying to do in the case we are discussing here is to try and simulate that smaller port diameter in an engine already designed for higher rpm power.


The properties of fluid flow in a closed system aren't going to change at all until the fluid becomes supersonic. What works at low flows will work exactly the same at high flows (unless somehow you manage to get the intake air to go supersonic... that would be pretty impressive).

What is different is the nature of the design that determines when the flow peaks; a two-valve head will naturally create higher velocity at lower rpms than a 4-valve head due to it's smaller port cross-section. Valve shrouding is a fairly large impediment to flow on 2-valve heads, which help the engines create more velocity at lower rpms. The "geometry" of the lower half of the engine is also a factor, with regards to the stroke, rod angle, and even squareness of the engine determining how quickly the cylinder will attempt to fill itself. This is another respect in which the G10 and G13B differ, and it plays a role in where each engine makes its power.

Intake manifold runner design is as much a function of utilizing Helmholtz resonance as it is fluid dynamics (maybe more?). Long runners are used to time pulses created by the intake valves. If you want a particularly interesting example. look at the Mazda's rotary engine. Even back in 1984, the 13b was using what they call a "Dynamic Effect Intake," which snakes runners around and directs the pulses from one rotor to another. This intake actually produces *positive pressure* inside the manifold. The Renesis uses the latest and greatest version of this, along with auxiliary intake ports and a valve which, above 6500 rpm (I believe, I could be wrong on that number), switches to bypass the extended-length runner path for a more direct shot into the engine (the resonance tuning takes somewhat of a backseat to keeping the air velocity high and getting it into the chamber as quickly as possible).

t3 ragtop wrote:
i wish i had just a penny for every "increase your mileage with my special 'vortex' adapter" scam that was run on somebody using an out of context scientific principle.


I think there is a misconception of what swirl is and how it's created for most non-technical people. If you look at the 160-hp engine that came in the 2002-2005 Honda Civic Si, the K20A3, you'll see some interesting things. Honda positioned it as a sporty car and touted its use of the new i-VTEC technology, but what it really had was a lame, fuel-economy version of it.

For one thing, most VTEC engines have the VTEC function on both the intake and exhaust cams, but this one didn't (intake only). Secondly, "hot" cam profile actually wasn't "hot," it was the normal cam profile! The "normal" cam profile leaves half the valves almost completely shut. The wall dividing the runner between the two valves had a hole in the middle of it to allow air to flow from the almost-shut valve's side to the open valve's side. I seem to recall that the idea of using one valve at low engine speeds was not only to improve velocity, but also introduce swirl to the chamber by filling it from the one side.

Anyway, my question is, why is Honda bothering with all this VTEC stuff when they could be threading their intake ports from the factory?


Top
 Profile  
 
 Post subject:
PostPosted: Mon Oct 08, 2007 1:14 pm 
Offline

Joined: Fri Jun 03, 2005 4:36 pm
Posts: 493
Location: Roscommon, MI
I don't like autos in general, but having only 3 gears, I especially dislike the Swift/Metro tranny and I don't feel it is very well matched to either engine (but it's definitely more livable attached to the GTi engine).

Also, I want to go out on a limb here and say that I generally like the idea of coatings. I look at them as similar to something like nitriding a crank. The issue I have with coatings though is that they *must* be professionally applied. I've heard more than a few stories of people having parts coated that came back too thick because there was too much coating applied (seems to be more common, and more of a problem, with rotary engines).

That said though, I think they're about as much a band aid as water injection is. That is, I don't look at it so much as a band aid, but an alternative way of doing things. But like water injection, it has to be done right to be most effective. And also, they're not necessities either.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Oct 09, 2007 1:32 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
I honestly don't know if texturing will improve intake port velocity.

After going through a bunch of technical material I have concluded that I'm probably wrong on a couple of things though. The cylinder head port should be slightly *larger* than the intake manifold port, it helps keep mixture reversion down. Also, heating the intake air will probably increase efficiency more by aiding fuel vaporization than keeping it cool will increase efficiency by aiding the Carnot cycle.

Some means of switching from cold to hot air intake at different throttle settings would probably be the ideal compromise for both power and efficiency. Perhaps a vacuum actuator could be used to induct cold air at very low vacuums and hot air at higher vacuums.

It also looks like a multi angle valve job will help fuel economy as will porting right behind the valve seats in order to increase flow at low valve lifts. Unshrouding the valves in the combustion chamber will help this also. Increased flow at low valve lifts gives some of the benefits of a higher lift, longer duration cam profile without the attendant problems with valve springs, cam lobe wear, increased valve timing overlap, etc, etc.

Done properly, you could probably slightly decrease valve timing and overlap for better low rpm torque while keeping the same or even more top end horsepower.

Slightly decreasing the port size with formagasket would also help raise low rpm torque and efficiency by increasing flow velocity.

It's late and I'll get back to this later.

Peace,

Jon[/quote]


Top
 Profile  
 
 Post subject:
PostPosted: Tue Oct 09, 2007 3:44 am 
Offline

Joined: Sun Aug 20, 2006 11:31 pm
Posts: 57
Location: Lummi Island
>Also, heating the intake air will probably increase efficiency more by aiding >fuel vaporization


This is sort of what I was trying to allude to, inducing a turbulence in the intake will increase the fuels "contact time" with the air and create a better fuel air mixture. I believe this is far more important to efficiency than velocity.
I am actually heating my fuel line before the injector and have seen a 3mpg. increase from this alone. Hot fuel has a lower vapor pressure and will evaporate faster. I would contend that the ultimate goal for efficiency is a completely evaporated fuel air gas, not an atomized liquid.
Take a good look at the Pogue carb design, it is a complex system for evaporating fuel before it enters the cylinder.
I have come up with an updated (much more simple) design for achieving the same effect. I have all the parts to put the carb together (except a spare manifold), but haven't had the time to work on the car. I also need to figure out an easy way to hook up the fuel injection/ignition system to a switch, so I can switch it all off when I experiment with putting on the vac. advance dist. and the special carb/manifold.
When I do finally prove this concept, I will post it all.
If anyone has a good plan for switching the ignition/injection on and off,
please let me know!

MM
1994 Geo Metro w/96 suspension (and heated fuel line)


Top
 Profile  
 
 Post subject:
PostPosted: Tue Oct 09, 2007 6:37 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
3play,

A minor nit: Hot fuel has a *higher* vapor pressure, not a lower one. Frozen gasoline has a vapor pressure approaching zero.

How are you heating your fuel line?

Do you have the lines between the heater and the injectors insulated?

As for switching from carb to fuel injection, you will have to get a vacuum advance distributor and the electronics for it. Then you will need an electrically actuated air valve to shut off the vacuum advance and a relay to switch ignition systems from the computer controlled one to the vacuum advance one.

For the fuel injection it should be fairly easy just to put a four pole relay in series with the injector control wires so the injectors can be turned on and off.

Shutting off fuel flow to the carb is going to be the trickiest thing, just blocking the fuel line will still leave the gas in the float bowl, which will play heck with the fuel injection when it is turned back on.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Oct 09, 2007 7:59 am 
Offline

Joined: Fri Jun 03, 2005 4:36 pm
Posts: 493
Location: Roscommon, MI
Soliton wrote:
The cylinder head port should be slightly *larger* than the intake manifold port, it helps keep mixture reversion down.


I agree with you there. Larry Widmer at ENDYN is big on that, but I ended up accidentally doing the same thing with the exhaust manifold on my Eclipse. I gasket-matched the runners before realizing the runners are way too large for that. Ended up bolting it back up to the stock head.

Now I don't have back-to-back numbers for that alone, but my car has been pretty damn fast for the turbo I'm running. I feel as though I might have stumbled upon a good idea. Like I said, Larry Widmer is all about using those type of "steps" on both the intake and exhaust. My question is, with regards to the intake, how much is that going to affect Helmholtz resonance? I would think if it's cutting back on reversion, it's going to cut back on that as well. It seems to me that it might be best to do this in conjunction with a short-runner intake manifold, but then this all is for performance-oriented stuff.

Getting into some good stuff here, but I don't have the time at the moment to reply to it all!


Top
 Profile  
 
 Post subject:
PostPosted: Tue Oct 09, 2007 9:07 am 
Offline

Joined: Mon Sep 24, 2007 8:03 pm
Posts: 43
OK, here's an idea that might have an application for the Coanda effect.

http://www.allpar.com/fix/holler/performance-and-economy.html

Quote:
Then somebody showed me a trick. Take a piece of the screen and something aerosol like Brakleen. Spray the Brakl.een directly at the screen. As you do, you will see the spray come out of the can and go past the screen. Slowly angle the screen away from the can. At just the precise angle the spray will disappear from behind the screen. At this precise angle, the screen will mostly vaporize the Brakleen as it passes through.


Find out the critical angle of the screen by performing this experiment and then put a piece of screen just downstream from your injectors and angled at the correct slant.

Glue it in place with formagasket and away you go.

As far as the Helmholtz resonance goes, since the ports are more optimized for high rpm performance than low rpm torque then the step will not make much difference in low rpm torque although it might have a slight effect on high rpm performance.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Oct 09, 2007 7:56 pm 
Offline
Suzuki Elder
User avatar

Joined: Mon Dec 01, 2003 12:47 pm
Posts: 11674
Location: columbus, ohio
the twincam 1.3 goes through several resonance points and has a distinctive buzz at "oh, crap" rpm. :lol: that engine has a heated throttle body (as does the turbo3) that really aids fueling. the manifold plenum that i recontoured for my 60mm throttle body and the tb itself don't have any coolant lines and that engine is quite bitchy and cold blooded.

the 1.0 just heats the whole intake manifold and injector body. it's pretty darned effective, too. i've driven through ice storms when almost nothing else would run in my trusty hoopty 3 banger.

i've had ice balls collect on carburetor jets before and shut the engine down. after engine heat came back up in the throat of the carb, the ice would melt, and the car would start right back up. i've iced up throttle body injected machinery like that, too. i've been known to rail at the "cold air intake on my 3 banger" crowd before. :-P it just seems counter productive in the winter.

3play, probably the slickest way to suspend pulses to the injectors without spiking the ecu would be to use an scr and switch the gate on or off. the timing issue would probably best be handled by something like an msd along with the vacuum advance dizzy. what engine are you using? if it's a g10, you're in like skin. just use a mk2 lsi distributor. it has 2 vacuum canisters and the magnetic hysteresis pickup will work fine as a trigger for an msd unit set up for 6 cylinders, lost spark.

_________________
1991 Blue Geo Metro Convertible highly modified 1.0L Turbo3 5 spd. - 1991 Red Geo Metro Convertible customized with a Twincam 5 spd.

My Turbo3 Project
My Cardomain Page -Ol' Blue
My YouTube Channel
My Photo Gallery
SAAB Sonett II


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 64 posts ]  Go to page Previous  1, 2, 3  Next

All times are UTC - 5 hours


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group