So Ive been doing some heavy research into engine building since I moved to the Honda world but I like to think that this is a better place to actually get some constructive feedback and productive discussion going. Plus I know some of you engineer for a living like me (well not professionally yet. Im still interning). But I digress....
Ive started hand coding a basic engine calculator in MATLAB (for those of you familiar with it) to help me work out the subtleties of build my LSVTEC or any kind of hybrid engine for that matter (for all of you hybrid Eco Guys out there). I have it calculating CC volumes, CR, DCR based on actual intake closing time of a cam (none of that advertised at .050" sh*t), rod ratio, piston speed etc etc. It also animates the rotating assembly for fun. But Ive gotten to the point of calculating intake manifolds and I've hit a bit of a logical snag that Im hoping somebody can explain to me.
So the intake runner length is driven by the resonant freq of the intake pulses at a certain rpm youre targeting as the peak of you powerband. This relies on the stacking of air against the intake valve once it closes. So by picking a good length you let the pressure wave bounce against the back of the plenum and stack against the intake valve more and more until you can actually push the engine over 100% VE with a good design. Here's where Im having issues. When the engine reaches BDC the intake valve isnt actually closed due to the cams duration holding it open until a certain point above BDC. So wouldnt this flow of air back into the intake manifold sort of cancel out the effect of the air stacking against the valve? I havent taken Fluids in school yet so I may be missing something.
Forgive me for rambling.
CLIFFS: How does a Hemholtz resonator work in an intake manifold if air/fuel is being pushed out of the CC before the cam actually closes the intake valve?
HELP?
the air is going to carry velocity. just because the valve isn't closed, the piston doesn't automatically push it back out because it's rising back up.
overlap scavenging works in a similar fashion.
exhaust valve and intake valve open, but the piston is moving down.. the intake air charge gets pulled in by the piston mostly, but the exhaust gases continuing out the exhaust also help out.
I don't believe reversion can happen that quickly because air is technically a liquid, and 'flows' as such.
I just wrote a @!#$ bible and when I hit post it was @!#$ gone. GOOD DAMN IT TO HELL!
If I get time later I'll retype it all the @!#$ out again.
Point of my long ass detained post, you want the valve to remain open past BDC to maintain velocity in the runners and cram the highest CFM into the runners you can. All NA cams are designed this way to keep velocity and CMF as high as possible.
My other MAJOR point: Generally speaking, the RPM at peak VE coincides with the RPM at the torque peak.
I'm really pissed that it just ate about one hour of @!#$ typing explaining why it is good to be open a little after BDC. @!#$
Edited 3 time(s). Last edited Thursday, November 15, 2012 6:11 AM
PRND321 Till I DIE
Old Motor: 160whp & 152ft/lbs, 1/4 Mile 15.4 @88.2
M45 + LD9 + 4T40-E, GO GO GO
Just the two people I was hoping would show up. Haha.
So basically since the pressure wave from the piston can only move so fast youre relying on the valve to close before the wave can reach the intake valve and revert? I just have a hard time visualizing the physics of things happening 100 times a second.
Im guessing that you like the intake valve to remain open after BDC so that the scavenging from the exhaust can help to pull more air into the cylinder the piston starts to move back up? Which increases your VE?
Mike Id still love to see what you had typed. I learned awhile ago to type my novels into notepad ahead of time so the JBO cant @!#$ me. Haha
Brian (TheSundownFire) wrote:So basically since the pressure wave from the piston can only move so fast you're relying on the valve to close before the wave can reach the intake valve and revert?
For the most part yes. The waves can only move so fast. Think of it as a pond and the wind is pushing small waves across it. When you throw a rock into the lake it take some time for the faster waves to hit and disturb the larger slower moving waves. I hope that helps.
Brian (TheSundownFire) wrote:Im guessing that you like the intake valve to remain open after BDC so that the scavenging from the exhaust can help to pull more air into the cylinder the piston starts to move back up? Which increases your VE?
Yes, you'll doing everything you can to have the most air in the cylinder as possible. The exhaust valve and intake valve are open when the piston is starting to travel upward, so the air flows to the lower pressure region. Out the exhaust. This also helps pull fresh air into the cylinder.
Brian (TheSundownFire) wrote:Mike Id still love to see what you had typed. I learned awhile ago to type my novels into notepad ahead of time so the JBO cant @!#$ me. Haha
I was on a roll, and honestly do not feel like typing that much again. If you have a specific question just ask.
-MD- Enforcer wrote:My other MAJOR point: Generally speaking, the RPM at peak VE coincides with the RPM at the torque peak.
I explained this in detail, and explained why you should backwards calculate from this, then try and calculate the resonant freq.
You have all the variables, and you know you want 100% or more VE. So design to that, and back calculate you resonant frequencies to check if your assumptions are correct, and make designing your intake manifold easier. I'm all about doing things as easy as possible, and then check to see if I come to the same conclusion the hard way.
Remembered, engineering is about safety factors, and what works "good enough." Nothing you will ever design will be perfect, and correct. You just need to design to be the best it can be, since you can not control all the variables.
PRND321 Till I DIE
Old Motor: 160whp & 152ft/lbs, 1/4 Mile 15.4 @88.2
M45 + LD9 + 4T40-E, GO GO GO
Thanks a lot Mike. Im sure Ill more questions as I keep working.
Brian (TheSundownFire) wrote:
So the intake runner length is driven by the resonant freq of the intake pulses at a certain rpm youre targeting as the peak of you powerband. This relies on the stacking of air against the intake valve once it closes. So by picking a good length you let the pressure wave bounce against the back of the plenum and stack against the intake valve more and more until you can actually push the engine over 100% VE with a good design.
The air doesn't "bounce" off the plenum. If that were the case, the ITB's would never work. The air "bounces" of the mass of air at the end of the runner/stack. If you've taken any transmission line theory, acoustics work the same way. At the boundary condition of where the valve closes, there is 100% reflection (i.e. the air has nowhere to go but back), but then there is only partial reflection at the end of the runner. So, every time the air travels back to the stack, energy is lost in this wave to the plenum air. This is why you try to design the length of the runners as closely to the 3rd harmonic as you can. Sometimes runner lengths have to be astronomically long (for lower engine speeds), so instead you try for the 5th harmonic.
DaFlyinSkwirl (Pj) v2.0 wrote:
exhaust valve and intake valve open, but the piston is moving down.. the intake air charge gets pulled in by the piston mostly, but the exhaust gases continuing out the exhaust also help out.
I don't believe reversion can happen that quickly because air is technically a liquid, and 'flows' as such.
Air is a
FLUID, PJ...not a liquid. Reversion into the intake does happen at low engine speeds with long duration cams that close well after BDC.
I have no signiture
Mike that was a good read. I added it to my collection. Its a good place to start on my next section with CCs.
Whalesac (My fellow Brian, if I remember right):
That makes sense. Ive worked a little bit with transmission lines in stereo building but nothing too scientific. With a transmission line subwoofer box you would target the free air resonance of the woofer and place it in a port the size of the woofer and tune the length of the port to the 4th harmonic (I think? Its been awhile) of the the free air resonance. I may be completely off base but thats what what youre saying reminds me of.
I can definitely see how T-line theory applies to ITBs but how does the plenum affect the resonance of the air with a T-line? The reason Im saying the wave bounces (oscillates is probably a better word) is due to the inflow of air from the plenum to the runner after the intake valve closes and a vacuum exists in the runner. The air comes in and over-fills the volume of the runner due to its momentum. The extra air in the runner wants to naturally return to the plenum since its at a higher pressure. When it does, the momentum of the air overfills the plenum. The air oscillates between the plenum and runner until the valve opens again or the pressure equalizes between the two. So you hope to time the opening of the intake valve with the return of the air into the runner so the air is forced into the cylinder rather than just being sucked into it.
So I guess as I typed I proved my previous post wrong but Ive done some more reading since I posted last.
Thanks guys. I appreciate the help.
Brian (TheSundownFire) wrote:
I can definitely see how T-line theory applies to ITBs but how does the plenum affect the resonance of the air with a T-line? The reason Im saying the wave bounces (oscillates is probably a better word) is due to the inflow of air from the plenum to the runner after the intake valve closes and a vacuum exists in the runner. The air comes in and over-fills the volume of the runner due to its momentum. The extra air in the runner wants to naturally return to the plenum since its at a higher pressure. When it does, the momentum of the air overfills the plenum. The air oscillates between the plenum and runner until the valve opens again or the pressure equalizes between the two. So you hope to time the opening of the intake valve with the return of the air into the runner so the air is forced into the cylinder rather than just being sucked into it.
This is only kind of true. There is an acoustic impedance mismatch between the runner and the plenum which causes reflection at the stack. When air travels down a tube, and there is a sudden change from one pipe diameter to another there will be reflection at that junction (e.g. @ stack/plenum or @ stack/open air for ITB's). Therefore, not all of that extra air escapes into the plenum...but you've got the basic idea which is the important part.
I have no signiture
Oh and just a little FYI
You should look into Scilab as an Open-Source alternative to Matlab. Matlab definitely has
MANY more libraries than it which makes it unarguably more powerful, but for anything you will probably ever use Matlab for, Scilab can do just as well...and it's free. I used it in college and I've used it at my job for the past couple years. I was given a Matlab license about a year ago, and for anything that doesn't involve plotting or designing a GUI, I still prefer Scilab. Just some friendly advice from one engineer to another
I have no signiture
Not much to contribute but wanted to say I enjoy reading about discussions like this on here. Keep up the good work.
PSN ID: Phatchance249
Whalesac wrote:Oh and just a little FYI
You should look into Scilab as an Open-Source alternative to Matlab. Matlab definitely has MANY more libraries than it which makes it unarguably more powerful, but for anything you will probably ever use Matlab for, Scilab can do just as well...and it's free. I used it in college and I've used it at my job for the past couple years. I was given a Matlab license about a year ago, and for anything that doesn't involve plotting or designing a GUI, I still prefer Scilab. Just some friendly advice from one engineer to another
Ill have to look into Scilab when my MATLAB license expires. I was buying my MATLAB license quarter by quarter when I needed it until last quarter when I was told the bookstore now only sold licenses that last until graduation.
I really appreciate the help. Im sure this will all make a lot more sense once I take Fluids in the summer. Ive always been more skilled at Thermo and conceptual classes than ones involving mechanisms so Ive been foaming at the mouth to start working with flowing fluids.
Ill be back for more help when I keep back to working on this thing. Im kind of hitting the wall of my off hand knowledge. I need to dig my Thermodynamics book out to get some good power calculations going.
I love thermo...but it is pretty straight forward. Also scilab is great
PRND321 Till I DIE
Old Motor: 160whp & 152ft/lbs, 1/4 Mile 15.4 @88.2
M45 + LD9 + 4T40-E, GO GO GO
Five whole months later and I've finally gotten my interest back into engine building and this damned program. I started my Fluid Mechanics class at the beginning of May, and I'm currently in the process of learning to model flow and what not in FLUENT and GAMBIT. <== Kind of OT I guess. I guess the odd thing is how much my Vibrations class has helped me wrap my head around this whole modelling air flow in and out as a series of waves.
Now that I've got my head wrapped around the majority of the intake half of the engine, its time to start playing with the exhaust side.
I'm just going to ramble for the sake of people who are interested. The best way to learn is to teach. Feel free to pick out anything I'm wrong about.
I've been doing a lot of reading between classes, homework and trying to stance my Accord. From what I can gather exhaust manifold and header design is incredibly similar to intake design (who'd have thunk it haha). The main difference being how you use Helmholtz resonance to enhance exhaust scavenging and pull the hot gases out of the cylinder. The harder you can pull coupled with the ramming effect of resonance in the intact tract can push a motor well over 100% VE at certain RPMs.
Essentially you use the same target RPM as you do when you design the intake manifold runners but you choose lengths in a way that places the resonant wave in the exhaust runner out of phase (180* probably) with the one in the intake tract. Rather than using the incoming wave to ram the air in, you're using the outgoing wave to pull exhaust out of the cylinder.
The next interesting thing: Finding a happy medium between volumetric flow and exhaust velocity. Since air (just like an fluid) has mass it also has momentum. Momentum is a function of the mass (m) and the velocity (v). The higher the momentum in the flow the more it will want to pull the rest of the flow with it (more exhaust scavenging) and the more it will resist being slowed down by the no-slip condition at the wall. Since we really cant change the mass of the air in a fixed volume the best way to increase momentum is by increasing velocity.
Velocity is increased in one of two ways: Either by decreasing the viscosity of the fluid flowing in the pipe or by decreasing the diameter of the pipe while maintaining the CFM of air flowing through the pipe.
Decreasing the diameter is a tricky thing because its not like you can say X.XX" diameter pipe is best. It all depends on the amount of air the engine can put out. (Its like my favorite analogy of blowing into a straw 1" in diameter with the same force as you blow into a straw 1/16" in diameter. The air comes out faster but at a lower CFM in the smaller straw.) This is why can actually HURT your performance by choosing too big of an exhaust for you engine. (That means no 3" on your stock LN2s boys.) Velocity is too low and scavenging suffers. There's plenty of fancy math for all of this and ways to optimize the values. <== I need to play with some simulations to make sure I'm right. Keep in mind that general rules of thumb are great here too. Minimize bends and diameter changes and obstructions.
Decreasing the viscosity of the fluid is a much easier way to go about it (albeit maybe not as effective). Viscosity is the thickness of the fluid. Think maple syrup (more viscous) vs hot cooking oil (less viscous). Any Newtonian Fluid will decrease in viscosity as you heat it. (This is completely different from your engine oil for another whole set of interesting reasons.) The best way to heat your exhaust gas? Keep the heat in your exhaust! This can be done through a variety of ways: material choice, header wrap, and thermal barrier coatings or the big ones.
The major problem with all of this is the fact that fluid flow and vibrations are so dynamic with engine speeds, pressure, humidity etc. That you'd never be able to get anything PERFECT.
I just love learning about this stuff. I think thats it for tonight though. Its time for bed. Later guys.
i think with this engine....you need to figure out what your doing with it. Since your this heavy into the design. I think you need to focus on if it is going to be a HP motor, a Torque motor, a fun street / auto cross motor, or a drag racing motor.
Since you seem to be designing it past 90% of what people design the motors to, you going to have to sacrifice something, or end up being epic at some RPM, and fall on its face at others.
I remember reading that 180 degrees out of phase isn't actually correct. If I recall correctly it is like 150 degrees, but it isn't that simple. Since the waves do not flow at a constant speed. I have heard some of the best headers designed, vary the degrees that they are out of phase, but honestly that is for 100% race headers and motors on mainly V8s. "For example 150, 210, 165, &195....<--- determined by firing order, and just an example...not specific for our motors." In all honestly if you designed it for 180 degrees out of sync....you'll be ahead of pretty much 99% of the headers available.
I'm a huge fan of ceramic coatings.
PRND321 Till I DIE
Old Motor: 160whp & 152ft/lbs, 1/4 Mile 15.4 @88.2
M45 + LD9 + 4T40-E, GO GO GO
I completely agree with you. Its never going to be perfect. Its actually probably a waste of time to go this nuts but attention to detail and a little math makes the difference between a mediocre N/A build and a great one in my mind.
I really don't have a specific engine build in mind. (It would be cool to go after the N/A J-body record though.) Im just trying to understand why certain engine combinations make more power than others. In the J-body world you dont have a lot to compare to but if you start looking around in other makes and models you can find an infinite mixture of parts (custom and off-the-shelf) that make vastly different amounts of HP and TQ even with the same displacement.
I can go online all day and have people tell me that a Victor X manifold, some Skunk Pro 2 cams, USDM ITR pistons and an ITR headers makes the best budget N/A LSVTEC but nobody can tell me why besides "My homeboy did that and made XXX.X hp and XXX.X tq." I have found out through all of this that it probably comes from the fact that it ends up with almost optimum SCR, DCR, quench height and deck clearance for a pump gas motor.
I have a situation like most J owners where big power means custom parts and fabrication and not just swapping parts until I like it due to limited aftermarket. I also have a thing for nasty N/A motors.
I have to look into that exhaust phasing stuff. Its funny to think of how little research and effort goes into a lot of performance parts out there. I guess you get what you pay for.
Thanks Mike!
That is why it takes me so damn long to get parts done. If they are not 95% or better...I do not sell them.
Most companies only revise a product if there is a problem. My original LD9 crank pulley is on its 9th or 10th revision.
PRND321 Till I DIE
Old Motor: 160whp & 152ft/lbs, 1/4 Mile 15.4 @88.2
M45 + LD9 + 4T40-E, GO GO GO
Brian,
If your thinking about building a big HP N/A Ecotec car, or want some other cam shaft profiles to play with, I've found a few European companies that make cams for the 2.2 that are FAR more aggressive than anything available from comp.
If you want the links PM me.
The euro cams are also offered with some big ass lift numbers.
PRND321 Till I DIE
Old Motor: 160whp & 152ft/lbs, 1/4 Mile 15.4 @88.2
M45 + LD9 + 4T40-E, GO GO GO
the eco head is never going to flow great numbers on the intake side for really high all motor numbers
the only things left to do that haven't been done yet:
-lighter jbody
-shorter final drive gearing
-bigger cams
-solid lifters
-methanol instead of gasoline (read: NOT meth injection)
honestly, there's a lot of money to be dumped into the valvetrain that just isn't really what many would consider disposable income.
a jesel valvetrain upgrade (rockers and solid lifters) is around $170 per valve. so multiply that by 16 for the complete setup...
the reason a lot of all motor guys call it quits is because they realize just how much the next step will cost.
the competition here is gone.
if you want to prove something to the 5 or so people who will actually care, then go for it, but honestly I think it's a waste of time.
I hate to be a detractor, but there's too much design wise to overcome on the eco to make serious power n/a
Challenge accepted PJ.
I can definitely understand where you're coming from though. I've just never been much of a forced induction kind of guy.
Another big thing I've noticed is a lack of high compression pistons in any of the N/A builds. Running big cams doesn't do much for you if you're running 10:1 factoryish compression. The duration bleeds off too much compression and you end up with a really low DCR. I'm sure you're aware of this though. I've honestly had a hard time finding any high C/R pistons for any Ecos. Mostly just the 8.9:1 forged slugs.
Whats the issue with the intake side of the L61 head? Not enough meat to port out?
Brian wrote:Challenge accepted PJ. I can definitely understand where you're coming from though. I've just never been much of a forced induction kind of guy.
Another big thing I've noticed is a lack of high compression pistons in any of the N/A builds. Running big cams doesn't do much for you if you're running 10:1 factoryish compression. The duration bleeds off too much compression and you end up with a really low DCR. I'm sure you're aware of this though. I've honestly had a hard time finding any high C/R pistons for any Ecos. Mostly just the 8.9:1 forged slugs.
Whats the issue with the intake side of the L61 head? Not enough meat to port out?
sorry i'm so late getting back to you on this
it's not the size or volume of the ports, it's the shape.
our runners are sorta flat and have a sharp turn near the end. totally kills velocity and robs energy
also, because of this, it's hard to port the runner to massively increase what can make it into the cylinder.
honda heads usually have a steeper runner shape with a less sharp angle at the end... the straighter the path, the more the engine can pull in on the intake stroke.
the other problem with building an all motor ecotec, almost all your parts will be custom.
custom cams
custom pistons
custom header
custom intake
that's a lot of custom crap to fab up/ farm out someplace.
but the bottom line is airflow. you need airflow. honda k series heads flow almost if not over 300cfm on the intake side if i remember correctly.
Even a heavily ported ecotec head is only touching on 300s
the other options I listed out tho, will be your best bet to laying down some serious numbers with an n/a setup.
Not a problem sir. We all get busy.
That makes sense on the runners. I've never had an Ecotec apart or really looked too closely at the head design on them. Honda K's tend to live life in the higher RPM ranges where CFM begins to matter more than intake velocity.
I ran some numbers with my little calculator dealy. Soooo glad I made this thing haha. All of these are at 100% VE.
--2.0L at 8000 RPM - 282.4 CFM
--2.0L at 10000 RPM - 352.8 CFM
--2.2L at 8000 RPM - 310.4 CFM
--2.2L at 10000 RPM - 388 CFM
--2.4L at 8000 RPM - 338.8 CFM
--2.4L at 10000 RPM - 423.6 CFM
I can definitely see where youre coming from if you can only net 280 CFM out of an Eco head. It might feed an LSJ on a stock rev limiter but never an LE5.
Guess I'll need to start looking at a S/C for my project.
I've been cooking something up for some interchangeable 86mm pistons for boost and N/A. But schools been nuts and I havent been able to double check my facts and math.
Thanks for the answers btw.
I was doing some pencil math for head flow in class today and realized it wasnt jiving with what the calc was coming up with. I forgot to convert units in one place in the calc and the numbers that I posted above are skewed high.
It doesnt seem like anybody cares really but I'm correcting myself for anyone who might look at this down the road.
--2.0L at 8000 RPM - 221 CFM
--2.0L at 10000 RPM - 277 CFM
--2.2L at 8000 RPM - 243 CFM
--2.2L at 10000 RPM - 305 CFM
--2.4L at 8000 RPM - 266 CFM
--2.4L at 10000 RPM - 332 CFM
M2 Race rates their CNC Ported LSJ heads at 303 CFM.
