who has done this??

[goonie]

The plenum gets smaller to simulate a rough 90* bend. There may be a wave tuning affect but that I know nothing about. I've seen an intake manifold on the 1995-1996 FORD/Cosworth RS200 rally car that had the throttle body in the middle, then a diffuser shaped plenum of average volume (like a good intercooler end tank) which encapsulated the 4 short manifold runners. The plenum was bolt on too. By no means did Buschur or Magnus invent the short manifold runner, they are just the first to adapt it to a DSM.

I give them a lot of respect though, no doubt about it. The only problem with the FORD manifold layout is the throttle body gets put high or low dead center in the engine compartment, not an easy install but this thing has a straight shot into whichever cylinder is aspirating at that moment. My Desktop Dyno text book suggests a 2*-5* runner taper for best results, I'll track down the exact number if anybodies interested.

I'm right now looking at a chart plotting volumetric efficiency vs 4 different manifold runner lengths on a 2.3L spark ignition engine. Assume a 6000 rpm rev limit, so torque values are going to be at lower rpms than our 2.0 liters. Lets see, a 26" long runner gives a strong hit at 3000 rpm (95% efficiency) but dies off to 59% at 5800 rpm. A 13.3" runner gives a double peak, one at 3000 rpm and another at 5000 rpm (90% efficiency in both cases). This one dies off to 75% volumetric efficiency @ 5800 rpm. This is most likely our stock manifold runner length. The 6.69" runner length peaks once at 3800 rpm @ 91% efficiency and holds 82% efficiency to 6000 rpm. The 3.3" runner length is a tiny bit worse than the 6.69" runner but very close in shape. Everybody thinks they invented this stuff first but I'm looking at charts that were published back in 1988 at the very latest.

I've got all the data I need to build my own, don't think I'll spend $600-$700 for a thin piece of aluminum sheetmetal. Using a round plenum is good because you don't need to get access to a sheetmetal pressbrake to form it up. I would slice it in half to mount the radiused inlet horns and runners. Then weld it back together. Cheers.

do you care to share that info i want to make my own mannifold from where i am there are no venders for any performence parts what so ever and ordering from the us
is too expensive with aal the shipping and taxes(add 50% of the price)
so a complete cylindrical wont outflow a square one

 

[pboglio]

do you care to share that info i want to make my own mannifold from where i am there are no venders for any performence parts what so ever and ordering from the us
is too expensive with aal the shipping and taxes(add 50% of the price)
so a complete cylindrical wont outflow a square one

I've got a VERY good Idea what design is ideal as far as pressure drop reduction is concerned, got an A+ in Fluid dynamics II afterall. Pulse tuning I'm relying on the chart I'm about to post now, that is one subject that is much harder to grasp.

Yeah, I'm looking at 5 charts that would blow your mind. Intake manifold pressure drop results every inch along the manifold in 500 rpm increments. Exhaust manifold pressure vs. rpm graph. The volumetric efficiencies graphs for 4 different runner length manifolds. A graph showing the total effect of ram tuning, pulse tuning, viscous drag effects all plotted together. I'll take 4-5 pics and post. With this info, you guys don't need to copy somebody elses design. Just follow the couple of good design rules and use your imagination in fabricating it. None of these guys building these manifolds are rocket scientists. I'd love to buy the software to simulate pulse tuning but I'm not that rich

 

[goonie]

great thanks man will wait for the pics

 

[silverbulletAWD]

I every spare moment I had last winter researching and designing the ultimate DSM manifold. I designed it with production in mind and figured at the least I would have a sweet mani. for myself if nothing else. I even had some respectable engineers at Ford helping me w/ my software. Summer came around and summer things took over. This winter I'm back at it and we'll see what happens. Long story short though, I can probably answer about any question you have. To address a few things I've already seen in this thread my opinion is this:

- Yes a round plenum is better than a square plenum. Not necessarily because it will flow more but by removing static turbulence causing corners you can improve efficiency and in turn improve flow.

- Tapered runners can help tremendously. However, there is no reason to go w/ more than a 2° taper. Anything else will only hurt flow. Optimal taper will be around 1.5°.

- The 4g63's TB position is about the worse place you could get but due to packaging constraints we are dealing. In a full custom situation one could benefit from getting the TB farther away from #4 and building in some sort of radius transition as opposed to the others abrupt 90° turn. I would also suggest pointing the TB more towards the pass. seat to improve entry angle but not to far as to come short of #1.

- This is totally my opinion but the only real reason why magnus' manifold has been reviewed as having some of the best low end power of aftermarkets is because the tapered plenum is substantially smaller than the others. The difference is more in plenum size than some scientifically designed plenum shape.

- Ram tuning has been around since the 60's there's not really a rocket science to that either. Although it can get pretty in depth for those who aren't familiar with it.

Here is a fun formula you can play around with and see what you get:
*This is by no means a proper explanation for a beginner but more of a summary for the already involved.

((ECD x .25 x SOS x 2) / (RPM x RV) - .5D = Runner length (Including port)

ECD = Effective Cam Duration (HKS 272 = Apprx. 478)
SOS= Speed of Sound (I use 80°F for reference that speed would be 1082ft/sec
RPM= Peak HP RPM (Peak should go 500-1000 RPM below shift point, I use -600)
RV= Reflective Value (this is which quarter wave you utilize, standard is 3rd)
.5D= Half the runners equivalent circular diameter (In a 1g's heads case it's 1.02)

Now this formula is actually pretty close to what you will want. But it doesn't consider engine size into the equation. When I was working with the Ford guys the spreadsheet we designed equated this but not ECD. After time and reconsideration I believe that ECD will have more impact than displacement and piston speed. For what it's worth, it seems that if anything our smaller engines call for a bit longer runners. In that spread sheet I get lengths of 1-2.5" longer typically.

Of course there are a ton of variables here but wave speed is the most obvious. So remember the size of the runner will play a large role but even more so will the air temp. which directly influences the speed of sound. Example: If your running a crappy intercooler play around with speeds in the 1200-1400ft/sec range.

I hope some of this can help you guys and it's not just the ramblings of a mad man. I don't have a full week to type all of my thoughts of this subject in detail.

Disclaimer- The above information is what I live by and use. It is believed by me to be correct. Some statements are based on my assumptions but rest assured they are all well backed by tons of research and discussion with experts. If anyone has any better thoughts or ideas I'd like to hear them.

Andy

 

[Black95TSIawd]

My magnus manifold took 5/10ths off my time (half second) and added 4mph to my overall 1/4 with no other modifications.

Honestly I'm a skeptic on many things but not when it comes to an intake manifold on a turbocharged car. They make big power. When the JG Edelbrock one come out at first for the Honda's we put one on a Turbo GSR making 330whp and it picked up 37whp with no other changes over the GSR intake manifold which is a pretty good design to start with. I couldn't believe it honestly it was huge and we all thought something else changed for the longest time. AFter seeing the results again on my own car I believe how much difference it makes now.

jim, i kno you put a 1g TB when u put your manifold on. People have seen gains of .2 or .3 by upgrading 2g vehicle to a 1g TB. Which mod do you think helped you more on pump, the TB or mani?

 

[pboglio]

This is the manifold I'd copy, if it were practical space wise. Here are some graphs:

 

[myblack98gst]

I have 2 also though of making one, and one day when I know how to tig weld good I may take on the task... But heres what I have brained so far..



for the runners.. use some with a diameter Equal in surface volume to the 2"x4" opening on the head.. and form each end to match that perfectly. allow the rest of the run to be round.. Inside the manifold I have seen small air horns for sale somewhere on the net.. these would help generate more velocity entering the runner. make the runner side of the manfold flat. for the rest of the housing, take a large sheet of aluminum and form a 180@ bend around something the same diameter as the width of the base.. then sit it down on the base with the 2 ends having a 2-3" reduction in height and mark where it needs to be cut to be 90* at the throttle body and say 40-50* at the capped end.. cut-fit the capped end and the TB end.. and weld it all together..


I dont like the idea of just cutting a piece of 4" pipe apart because it doesn't give the volume the plenum needs to be.. with forming the mass of the plenum out of one piece it reduces angles which can cause unwanted flow impedance..

now the trick for all of this... if you want to be real hardcore, make the first one and cut the runners in half and use silcone couplers to hold it together and a brace system so it doesn't colapse.. then take it to the dyno and adjust the runner length to get a power curve you like... then A) build a new manifold better with your learned experience or B) weld the runners back together at the length you want..

There is no way to 100% calculate what you want on paper then it actually happen the first try there are too many variables that can happen

 

[pboglio]

Sounds like you got a good plan for fabbing. I worked in a prototype shop and worked with sheetmetal extensively, along with designing sheetmetal weldments. I've done complicated weldments and the fixtures to weld them. I could do the sheetmetal manifold with the minimal weld lines, keeping them symmetrical to minimize warping. This would all have curve walls too. With .060 thk or so material, you could dead blow hammer form over a die to close in the seams, maybe use a little heat if need be.

Everybody needs to fabricate and design based on the equipment available and their skill sets. This is what causes people to compromise their design. I'd for sure farm out the welding, I've got my local welder thats done all of my aluminum welding, real nice work. Design and fabrication I'd tackle myself. The intake manifold flange I'd purchase cut already, too much work to cut a flat aluminum plate & cut the runner openings and then radius everything. I'd shell out the $85 for the manifold flange.

There's a lot of good ideas in everybody's posts here.

 

[NDgsx]

That's an interesting cosworth manifold, did it say what chasis that came out of? To fit that on a escort cosworth you would have to move the header tank completly(for those that don't know european cars that's kind of like a overflow bottle for coolant, but better). I can't remember if the sierra has more room around there.
On the escort cosworth I worked on the throtle body was mounted under the plenum, the air comes in at a the center of the plenum, parellel with the motor, then travels up into the plenum. The plenum was modular so different sizes can be used and the runner length was ~4".
There is a very good way to get equal flow distribution while mounting the TB on the end of the plenum. Look at the HKS 4G63 or dahlback audi 1.8T manifolds to get the idea. It would be hard to make with sheetmetal, you would cast it ideally.
Jenvey has velocity stacks in various sizes. For a 1G manifold they are a little bit small (largest being 52mm), but you can always cut them down.

 

[pboglio]

That's an interesting cosworth manifold, did it say what chasis that came out of? To fit that on a escort cosworth you would have to move the header tank completly(for those that don't know european cars that's kind of like a overflow bottle for coolant, but better). I can't remember if the sierra has more room around there.
On the escort cosworth I worked on the throtle body was mounted under the plenum, the air comes in at a the center of the plenum, parellel with the motor, then travels up into the plenum. The plenum was modular so different sizes can be used and the runner length was ~4".
There is a very good way to get equal flow distribution while mounting the TB on the end of the plenum. Look at the HKS 4G63 or dahlback audi 1.8T manifolds to get the idea. It would be hard to make with sheetmetal, you would cast it ideally.
Jenvey has velocity stacks in various sizes. For a 1G manifold they are a little bit small (largest being 52mm), but you can always cut them down.

Good eye. It's actually off of the 1995-1996 Ford RS200 Rally car. If you see a photo of it you'll notice a roof mounted large intercooler. The throttle body sweeps upwards to meet with the intercooler in the roof. I'd do the cosworth style manifold but turn the plenum upside down, and cut off the throttle body connection and weld in a 90* bend facing the passenger side. My guess is its much more important for the 90* bend to come before the throttlebody, rather trying to do it inside of the plenum. Anything Cosworth I like, these guys can definitely engineer a race car.

 

[DSMJim]

jim, i kno you put a 1g TB when u put your manifold on. People have seen gains of .2 or .3 by upgrading 2g vehicle to a 1g TB. Which mod do you think helped you more on pump, the TB or mani?

I have never heard of seen anybody take .2 or .3 off a Mid 12's second car with a throttle body upgrade from 52 to 60mm. That is saying that it added 20-30hp with a throttlebody. You might take .2 or .3 off a 14 or high 13 second car which is a lot easier and takes much less HP. Not gonna happen on a mid 12 second car pulling 111mph then going to low 12 second with 115mph. I'll bet the throttle body made up for many one 1/10. Maybe...

 

[gixrman]

To bad no one can come up with variable lenght runners. I know they did this on GP bikes, servo controled by throttle and rpm, and it was banned after the first yr. Nice flat torque curve. Or how about equal lenght intake runners like the old GM Quad 4's?

 

[awd94talon]

I have been thinking about making my own intake. I have talked to a couple machinist. I had to weld up my magnus after it cracked the #4 runner. They said it was a nice piece but there would be too much swirling air. Waist of air space. I was thinking of making one that is similar to an tubular exhast manifold. It would need to have the throttle body placed where most people put there gm mas on a blow-thru setup. After the tb the there should be a straight area that straightens the air. then seperate equally before the elbow into 4 equal length pipes that end up at the head. Basicall the oppisite of a tubular exhaust.
What would be wrong with this setup? And why has no one done it yet?

Robert

 

[myblack98gst]

bad part is I was gonna suggest this first last night but it was late and I couldn;t make sense of myself...

the old GM Quad 4 used a manifold that looked like a tubular exhaust manifold with a throttle body on the collector end of it..

Interesting to see how it would work.. but I am unsure of the adverse effects of no plenum

 

[92redman]

To bad no one can come up with variable lenght runners. I know they did this on GP bikes, servo controled by throttle and rpm, and it was banned after the first yr. Nice flat torque curve. Or how about equal lenght intake runners like the old GM Quad 4's?

You could probable come up with a decent dual runner manifold similar to the cyclone intake or perhaps Chrysler's 3.5 HO. You could use the VTEC SAFC and use the VTEC output to control the butterflys.

 

[pboglio]

I have been thinking about making my own intake. I have talked to a couple machinist. I had to weld up my magnus after it cracked the #4 runner. They said it was a nice piece but there would be too much swirling air. Waist of air space. I was thinking of making one that is similar to an tubular exhast manifold. It would need to have the throttle body placed where most people put there gm mas on a blow-thru setup. After the tb the there should be a straight area that straightens the air. then seperate equally before the elbow into 4 equal length pipes that end up at the head. Basicall the oppisite of a tubular exhaust.
What would be wrong with this setup? And why has no one done it yet?

Robert

Pure keyboard bench racing here but. If you run a tubular style intake manifold like an exhaust manifold with no plenum, you effectively lengthen the intake runners to the length of your entire intake tract up to the intercooler exit plenum. Opposite of what a short runner manifold achieves. We already have proof that a particular short runner length produces a significant power increase. PART of the reason for the plenum is to have a terminating box that at least acoustically, tricks the engine into thinking the intake plenum is the outside of the engine, allowing the engine to SEE only the short intake manifold runner and ignore everything upstream of the throttlebody. Only way to know for sure is to build it and find out. I mean the plenum has to be a certain minimum volume. You need 3 times the runner diameter for the intake runner velocity bells at a minimum. You also need at least the diameter of the throttlebody and the length from cylinder 4 thru 1. This really fixes your minimum plenum diameter and volume, of course you can go bigger.

 

[goonie]

here is another drawing ,will it help flow when i put the mannifold and the runners in an angle like this ? or just the mannifold in an angle and the runners straight
and should i keep an equal air traveling distance for each runner(messured from the tr-body to the cilinder head?)
for so far thanks for all the replies
good info

 

[thegreatms]

Got the book because I took the class in Mechanical Engineering Graduate School.

pboglio,

Do you mind if I ask what school you went to, and the name of the class and professor? My grad school had very few courses of that nature. It was very theoretical, not much experimental research going on there. Hearing about a course like that makes me want to go back to school.

thanks
mike

 

[thegreatms]

You could probable come up with a decent dual runner manifold similar to the cyclone intake

I have though about this, instead of have the two sets of runners being very similar in size and shape (like in a cyclone), you could have a almost stock set for low end torque and a second set tuned for higher rpm's. If done correctly you could have an efficient IM over a much higher range of rpms

 

[pboglio]

With that design, you're going to have different tuning peaks for each cylinder. That might be a cool way to average out the power increase across the rpm band. One cylinder would run at a tuned peak, carrying the other slightly less tuned cylinders, and vice versa throughout the rpm band. On average you could have a higher volumetric efficiency. It would also run each particular cylinder lean because of the increase in volumetric efficiency over the other cylinders. If you had individual cylinder 02 feedback and individual cylinder fuel injection control, it would be good. I'd stick with all equal length runners to be safe, this is what I've seen on every manifold I've looked at. Equal airflow to each cylinder would be first priority on my engine, just to keep it from blowing up. Seems like plenum shape, size, runner length, runner diameter are the variables; but all the runners should be the same length.

Looking at the Helmholtz resonator model equation I have, it appears that the taper angle allows a varying cross sectional runner AREA to help "broaden" the tuning peak, but also reducing slightly the amplitude of it. This is similar to the use of a "MEGS" pipe on a bike, the tuning peak is broadened over a wider rpm band, at the expense of reducing the maximum peak a bit. The recommended taper angle is so minimal I doubt the increased velocity from the start of the pipe to the end would have much effect on in cylinder mixing.

 

[pboglio]

pboglio,

Do you mind if I ask what school you went to, and the name of the class and professor? My grad school had very few courses of that nature. It was very theoretical, not much experimental research going on there. Hearing about a course like that makes me want to go back to school.

thanks
mike

University of Illinois at Chicago, Mechanical Engineering Graduate Program. I did research in Biomechanics but it was mostly a joke. My school had some of the best Internal Combustion research in the country (Dr. Minkowitz (sp) and Dr. Puri). I took the "Internal Combustion Engine" class to fill out my electives, against my professor's/advisors/employers wishes. I thought it would be fun since I was into cars but the class only focused on the first 5 chapters which was boring as hell. Most of that was figureing out a/f ratio's and chemistry believe it or not. Chapters 6-14 is where all the good stuff is, for all you gear heads I had my undergraduate degree in Mechanical Engineering, but most of the guys in my class were Automotive Engineering majors. There are better schools for Automotive Engineering, University of Michigan in Ann Arbor comes to mind as one of the best that the average human being can get into.

 

[Black95TSIawd]

I have never heard of seen anybody take .2 or .3 off a Mid 12's second car with a throttle body upgrade from 52 to 60mm. That is saying that it added 20-30hp with a throttlebody. You might take .2 or .3 off a 14 or high 13 second car which is a lot easier and takes much less HP. Not gonna happen on a mid 12 second car pulling 111mph then going to low 12 second with 115mph. I'll bet the throttle body made up for many one 1/10. Maybe...

good point, thanks jim.

 

[92tsi123]

you guys have alot to talk about with intake manifold.

you can change shape and size all you want but, what alot of you are forgetting is it has to be tuned, like a header system, each runner like exhaust pipe has to flow even amounts of air. and them you have to time the air waves with the intake valves, the waves bounce around in the manifold and must be syncronized to bounce into thehead exactly when the valve open. thats is very important. most aftermarket intake manifolds do not have the money and equipment to to make a manifold do this , there is alot of testing. the stock manifold however is, the manufacture has the money to make a manifold do that.

 

[2.4 turbo Hyped]

you guys have alot to talk about with intake manifold.

you can change shape and size all you want but, what alot of you are forgetting is it has to be tuned, like a header system, each runner like exhaust pipe has to flow even amounts of air. and them you have to time the air waves with the intake valves, the waves bounce around in the manifold and must be syncronized to bounce into thehead exactly when the valve open. thats is very important. most aftermarket intake manifolds do not have the money and equipment to to make a manifold do this , there is alot of testing. the stock manifold however is, the manufacture has the money to make a manifold do that.

All your base belong to us

Seriously though, make me one and I will dyno it before and after. Sean

 

[pboglio]

you guys have alot to talk about with intake manifold.

you can change shape and size all you want but, what alot of you are forgetting is it has to be tuned, like a header system, each runner like exhaust pipe has to flow even amounts of air. and them you have to time the air waves with the intake valves, the waves bounce around in the manifold and must be syncronized to bounce into thehead exactly when the valve open. thats is very important. most aftermarket intake manifolds do not have the money and equipment to to make a manifold do this , there is alot of testing. the stock manifold however is, the manufacture has the money to make a manifold do that.

Nobody is forgetting about synchronizing tuning. You could tune in an already built aftermarket manifold with adjustable cam gears. I'm looking at a field test chart with 3 different valve timing settings run on a 1.6L spark ignition engine, and there is almost a 10% difference in volumetric efficiency between them. You can wave tune by adjusting the runner lengths OR adjusting the valve timing to a certain extent. You could also dial in your manifold with the cam gears for a perfect match.

Correct, the Helmholtz equations don't take into account valve timing. It does give you a ball park to start with though. In Magnus's defense, if you recall, he used to sell his manifolds with silicone hose runners. He switched to all aluminum once he field tested a few different runner lengths and perfected it. I have no doubt his manifolds are dialed in close for somebody running your average HKS cams, way too many people are having a blast running that manifold. Building your own manifold will alter the manifold volume as well as runner length and diameter, which IS accounted for in the Helmholtz equation for multibranch runners.

For next to no extra effort, a crafty guy could build his manifold with mid runner flanged inserts. They would look just like mini test pipes, flange on both ends, threaded studs for clearance. Say build 4 different insert lengths in .5" increments. Say one insert thats 1" long, second at 1.5" long, etc. Still would be able to support the weight of the manifold and throttlebody, looks good, clean flow with a little internal clean up. Could also be used to adjust for changes in camshaft lift and timing when changing to different camshafts. Use the Helmholtz equations as a baseline and work from there. Nobody said they would build a manifold that would be perfect out of the box.