Page 1 of 3 123 LastLast
Results 1 to 30 of 71

Thread: Intakes, plenums and runners....again...

  1. #1
    Registered User 333pg333's Avatar
    Join Date
    Feb 2011
    Location
    Sydney, Australia
    Posts
    2,278

    Intakes, plenums and runners....again...

    Realise that Intake theory and development can be a black art but thought I'd throw this to the wind as there are some smart cookies in here.

    Have a new motor coming together shortly. Going to be a bit different from last year's version. Looks like we're going to have to find or knock up a new intake to suit. Builder (who is not supplying intake) says that in his experience, in a nutshell, he said that turbulence is the main cause of choke points regarding intakes. His brief recommendation was to make or buy something with offset runners.

    Trying to research at work leads me to quite a bit of variation on a theme here. Does offset refer to runners that come in at an angle (while still on the same plane) or those that are curved/angled on a different plane? Clearly a lot of intakes are affected by packaging which I'm sure has a big effect on the design.

    Another question that has cropped up. Plenum volumes. Seems to be differing opinion on this. Some go by a rough formula of 1.5 - 2 times the capacity of the motor where I've also heard that a well constructed intake only has to have enough space for 1 cylinder at a time (i4 motor).
    On the previous motor we went from the stock intake (small plenum/long runners) to large plenum (approx 5 litres) and shorter runners. (50mm Bellmouth into some taper then 41 i.d.) Shown in pics.

    The new motor will be smaller in capacity (2.5ltr vs 3.1ltr) but have much bigger inlet ports, very high lift cam (.595) and solid lifters. 2v head got flowed earlier this week and saw 271cfm@ 28" / 600 lift. Old head was 246cfm @ 28" / 550 lift.

    So we've got a motor that will keep pulling to target 8000rpm where the old one nosed off at 6000rpm. With the new ports being approx 53mm x 49mm should we be basically trying to make something with shorter runners again and similarly large id on the runners? If so, does that mean we need to source components with even larger bellmouths. eg 60mm ?
    http://www.jntperformance.com.au/ind...l?catId=269035

    Is it ok to have the taper at the top but then go into straight pipe?

    Thanks for any ideas etc...
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	WTAC_PORSCHE_0584-800x533.jpg 
Views:	48 
Size:	118.1 KB 
ID:	81573   Click image for larger version. 

Name:	2014-07-28 11.58.18.jpg 
Views:	47 
Size:	1.84 MB 
ID:	81574   Click image for larger version. 

Name:	2014-07-29134432_zps815cd5fd.jpg 
Views:	48 
Size:	90.5 KB 
ID:	81575  

  2. #2
    Registered User
    Join Date
    Apr 2008
    Location
    NSW
    Posts
    557
    plenum looks pretty big like you would see on a NA engine, why not something smaller for more response? length looks ok for that rpm.
    E30 325is M20 3.1L

  3. #3
    Registered User jimmyg's Avatar
    Join Date
    Mar 2009
    Location
    Bendigo, Vic
    Posts
    1,710
    In relation to the volume of air being pumped through the system the volume is SFA!
    Mmmmmmmmm beer.

    Quote Originally Posted by ALLMTR View Post
    You had one fucking job, Imran. One. Now look, there's Flibberty Gibbets all up in here, seeing straight through your ruse. One. Fucking. Job!

  4. #4
    Registered User
    Join Date
    Nov 2005
    Location
    Wrong.
    Posts
    7,157
    This thread has interesting results. But forced induction, still a handy read. As much as the maths says, it's cool to see someone actually physically bolt on 10 different designs, and dyno 10 different designs, instead of theorise about 500 things and try 1.

    http://highboostforum.com/forum/show...and-facts-only

    Some conclusions empirically were that a large protruding ridge like a prolapsed arsehole around the runner was largely beneficial, the velocity stack needed little to no length more than maybe 1/3 the runner ID in a force fed engine (I assume as you're valve c/s area flow limited?), the huge throttle body provided fuck all difference to the stock evo TB.

  5. #5
    Interested bystander
    Join Date
    Dec 2001
    Location
    Australia
    Posts
    849
    What is the engine stroke?

  6. #6
    Registered User 9triton's Avatar
    Join Date
    May 2009
    Location
    melbourne
    Posts
    4,958
    my recommendation is buy pipemax- appprox 60 USD and start pumping numbers into that.


    takes a bit of effort to understand what its telling you -

    but 'as used by TK' cant be bad

    will size manifold runners,port taper , ports cross section,exhausts , lift ,choke points etc etc

    edit - yours is turbo'd?
    pipemax can still be used

    TK may let on how you compensate for turbo as its its primarily for NA engines -

    but relevant for turbo as density changes - not volume when turboed ( for equivalent cfm)
    Last edited by 9triton; 13-02-15 at 05:47 PM.

  7. #7
    GTFO of my ED doctor ed's Avatar
    Join Date
    Jun 2002
    Location
    irgendwo
    Posts
    4,703
    Quote Originally Posted by cracka View Post
    it's cool to see someone actually physically bolt on 10 different designs, and dyno 10 different designs, instead of theorise about 500 things and try 1.
    whilst yes thats cool, its also re-inventing the wheel of the 100000's of actual quantitative studies done by people, like, you know, NASA

    theres some stubbornly fundamental basics of fluid dynamics and acoustics that any runner/plenum design can benefit from.

    those runners on your current (custom made) plenum are too long and have too small a dia. Boundary layer thickness will expand along the length of the runner, sonically throttling that intake's flow capability. even if the cams can breath, and the head flow is epic, that manifold will stall HP once runner velocity hits critical. runners need to be much fatter, also probably shorter. get pipemax, run calcs on the new engine. that'll get you in your ballpark.

    ideal taper is ~1.5% along the length of the runner, but fabrication/design limits often dictates otherwise.

    JNT has some great, very long, bellmouths with taper. use them as far as packaging allows
    Mit freundlichen Gre

    Quote Originally Posted by Keith Duckworth View Post
    "I think that in a racing engine, the closer it is to disintegrating, in general the better its performance will be "

  8. #8
    Now with more ebay parts. 2JZR31's Avatar
    Join Date
    Sep 2002
    Location
    Brisbane
    Posts
    11,230
    Quote Originally Posted by cracka View Post
    This thread has interesting results. But forced induction, still a handy read. As much as the math says, it's cool to see someone actually physically bolt on 10 different designs, and dyno 10 different designs, instead of theorise about 500 things and try 1.

    http://highboostforum.com/forum/show...and-facts-only

    Some conclusions empirically were that a large protruding ridge like a prolapsed arsehole around the runner was largely beneficial, the velocity stack needed little to no length more than maybe 1/3 the runner ID in a force fed engine (I assume as you're valve c/s area flow limited?), the huge throttle body provided fuck all difference to the stock evo TB.
    new sig
    10.83 @ 125

    Quickest stock exhaust manifold stud 2JZ in Aus.


    Quote Originally Posted by cracka View Post
    Some conclusions empirically were that a large protruding ridge like a prolapsed arsehole around the runner was largely beneficial.

  9. #9
    Registered User
    Join Date
    Nov 2005
    Location
    Wrong.
    Posts
    7,157
    Fucks me, you can make all the pretty you like. It's very interesting to me to see the prolapsed arse ahead of the sunken velocity stack. Yep, all also 100% calculable shit. But CFD doesn't help the backyarder layman.

  10. #10
    Now with more ebay parts. 2JZR31's Avatar
    Join Date
    Sep 2002
    Location
    Brisbane
    Posts
    11,230
    whilst yes thats cool, its also re-inventing the wheel of the 100000's of actual quantitative studies done by people, like, you know, NASA
    I am right because NASA!
    10.83 @ 125

    Quickest stock exhaust manifold stud 2JZ in Aus.


    Quote Originally Posted by cracka View Post
    Some conclusions empirically were that a large protruding ridge like a prolapsed arsehole around the runner was largely beneficial.

  11. #11
    GTFO of my ED doctor ed's Avatar
    Join Date
    Jun 2002
    Location
    irgendwo
    Posts
    4,703
    Quote Originally Posted by cracka View Post
    It's very interesting to me to see the prolapsed arse ahead of the sunken velocity stack. Yep, all also 100% calculable shit. But CFD doesn't help the backyarder layman.
    you dont do the CFD yourself, just use the results of other smart people's CFD.
    it should be common knowledge for eg, that a raised radius flows better. didnt need to do the CFD myself. read it in a book.

    books, the crazy things hide heaps of info
    Last edited by doctor ed; 13-02-15 at 07:54 PM.
    Mit freundlichen Gre

    Quote Originally Posted by Keith Duckworth View Post
    "I think that in a racing engine, the closer it is to disintegrating, in general the better its performance will be "

  12. #12
    Registered User
    Join Date
    Nov 2005
    Location
    Wrong.
    Posts
    7,157
    Quote Originally Posted by 2JZR31 View Post
    I am right because NASA!
    NASA wrote the book on writing books on thing. That shit is final.

    Ed, design me a 1uz manifold pls.

  13. #13
    GTFO of my ED doctor ed's Avatar
    Join Date
    Jun 2002
    Location
    irgendwo
    Posts
    4,703
    why the trolling antagonism?

    shit is complex, but studied by lots of smart people, and obeys reliable rules of physics. you wont get a lemans winning manifold, but youll get a good one that works, just apply a few principles that have been worked out already for you.
    Mit freundlichen Gre

    Quote Originally Posted by Keith Duckworth View Post
    "I think that in a racing engine, the closer it is to disintegrating, in general the better its performance will be "

  14. #14
    Registered User 333pg333's Avatar
    Join Date
    Feb 2011
    Location
    Sydney, Australia
    Posts
    2,278
    Thanks for replies. A bit more info. The intake we made up was a last minute decision and not a lot of science went into it. Our idea was to increase the plenum volume, shorten the runners, add bellmouth and see what happened. The results were pretty good, especially if you go by some of the results in that Evo link in post #4. Essentially we made 30whp over the stock setup at 20psi and 50whp at 25psi. I'd say that's not a bad improvement for what we made.
    Old motor was 105.5mm bore x 88mm stroke. New one is 102mm x 78mm stroke, so a very different motor. Both still old school 2v / single OHC. Not sure if the GTX3582 might not be a bit too big for the new motor too.

    Another thing I'm not 100% about is matching the runners/flange to the port shape/size? From what I know very simplistically, if you have a smaller aperture on the flange leading into a bigger port opening, the mix will slow down. Of course this is an incredibly basic viewpoint from me as it reflects my knowledge. But is it advisable for the flange opening to mirror the port or isn't it absolutely crucial? As you can see on the old head vs new one, quite a difference. Would it be catastrophic to run with our current intake for now until we come up with something better?

    Again, referring to that EVO thread, seems like most of them have a similarity about their general shape plenum and short straight runners. So I'm still not quite sure what an 'offset' runner is. I suspect it's those with a bend in them on the same plane vs the curved ones that are on an LS motor for example?
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	1st 8v head.JPG 
Views:	29 
Size:	1.36 MB 
ID:	81609   Click image for larger version. 

Name:	Neils big 8v port.jpg 
Views:	30 
Size:	2.16 MB 
ID:	81610   Click image for larger version. 

Name:	new and old intakes.jpg 
Views:	24 
Size:	101.4 KB 
ID:	81607   Click image for larger version. 

Name:	20psi comparison new and old intake.jpg 
Views:	23 
Size:	66.0 KB 
ID:	81608  

  15. #15
    Registered User
    Join Date
    Apr 2008
    Location
    NSW
    Posts
    557
    Quote Originally Posted by doctor ed View Post
    you dont do the CFD yourself, just use the results of other smart people's CFD.
    it should be common knowledge for eg, that a raised radius flows better. didnt need to do the CFD myself. read it in a book.

    books, the crazy things hide heaps of info
    it depends on the angle of attack as to whether a raised bellmouth is better than a sunken radius for flow. its not an absolute thing. for example if the plenum was small and the entry was at 90degrees to the runner axis then it would be bad as the air would need flow up and over the mouth making a sharp turn.
    E30 325is M20 3.1L

  16. #16
    Registered User
    Join Date
    Nov 2005
    Location
    Wrong.
    Posts
    7,157
    Quote Originally Posted by doctor ed View Post
    why the trolling antagonism?
    Ladies and gentlemen of troll-court, I present these initial antagonistic pearls of trolling such as:
    its also re-inventing the wheel of the 100000's of actual quantitative studies done by people, like, you know, NASA
    and
    books, the crazy things hide heaps of info
    The defense rests.

    Kidding aside, that's real world testing, with visual clues as to what to look for. Not all of us are so smart as to have a read through a Colin Chapman book and be able to picture the exact orientation/shape/function/design/finish of an item, and that thread is a really good overview of how a dozen different items all not varying very much at all in design can perform very differently.

    If you described those evo manifolds to me and asked me to sketch them based on your description, I think I would draw the same thing at least 8 times. Being able to visualise the prolapsed arsehole, the tapered end, the angled intake, the port shape and changes from circle-square-rectangle etc... I never said it was a manual to manifold design, just that it's got interesting results, and that the velocity profile and runner length was not nearly as long as I would have imagined for the more optimum designs.

    Edit:
    Quote Originally Posted by 333pg333 View Post
    referring to that EVO thread, seems like most of them have a similarity about their general shape plenum and short straight runners.
    This is exactly why I found it interesting. There's 3 general shape/designs there, but an 80hp disparity between them all. That's huge, and relevant to me as I'm going to get RZ to knock me up an amazing intake manifold for the 1UZ and someone like Dr Ed would be very helpful in some tips to get it right for my application. I could very easily fuck it up if not for asking the questions and looking at what other people have tested and concluded.
    Last edited by cracka; 14-02-15 at 07:01 AM. Reason: niggers

  17. #17
    Registered User TK's Avatar
    Join Date
    Sep 2006
    Location
    Adelaide
    Posts
    6,984
    To design an intake the very best 1st step is to buy a copy of Larry Meux's Pipemax.
    Then learn how to use it & follow it's recommendations as close as you possibly can - Pipemax never fails.

  18. #18
    Registered User 9triton's Avatar
    Join Date
    May 2009
    Location
    melbourne
    Posts
    4,958
    Quote Originally Posted by TK View Post
    To design an intake the very best 1st step is to buy a copy of Larry Meux's Pipemax.
    Then learn how to use it & follow it's recommendations as close as you possibly can - Pipemax never fails.
    ive got a copy and slowly trying to understand what it all means - BUT how do you model for turbo manifold pressures?

    VE ?
    Last edited by 9triton; 14-02-15 at 07:24 AM.

  19. #19
    Registered User
    Join Date
    Dec 2007
    Location
    melbourne
    Posts
    5,201
    Quote Originally Posted by 9triton View Post
    ive got a copy and slowly trying to understand what it all means - BUT how do you model for turbo manifold pressures?

    VE ?
    If there's a setting for atmospheric pressure, fiddle it until you're at -11ty thousand feet DA (the equivalent air density of your boosted airflow)

  20. #20
    GTFO of my ED doctor ed's Avatar
    Join Date
    Jun 2002
    Location
    irgendwo
    Posts
    4,703
    Sory for adding a sense of smart-assedness, meant to only make a dry bit of info slightly more entertaining to read. The other (NASA) stuff is real world testing too. Just because it's not 'not tested on a car' does not mean it has zero relevance. Air doesn't know what kind of device it's flowing through. The most prolific use of the theories you'll find implemented in mundane stuff like industrial duct work and big ventilation systems. Those guys care about losses!

    NASA is just one example of smart guys with lots of money doing the hard yards that the backyarder can benefit from. I wasn't saying I'm right because NASA, im saying NASA is right because NASA. I've got nothing personal to contribute. The sheer volume and complexity of info on this topic is slightly mind bending. But it is all written down, SAE papers and other shit a plenty. It's really interesting reading if you're that way inclined. I tried to summarize it on my site before it got hacked. Unfortunately it's all gone.

    Re port matching... Yes yes yes. One of the problems of forced induction is, depending on intake charge pressure, the flow has an increasing tendency to form a thicker boundary layer (stationary air stuck to the runner walls) which reduces the effective diameter of your runner, and increases intake velocity. Sonic choke because of a narrow runner is bad (Google sonic choke). Bellmouths and tapers reduce boundary layers, normalizing the internal diameter of your pipe. Steps, whacky bends etc cause separation of flow, and fuck your boundary layer conditions. Always, ALWAYS better to have smooth transitions, even if it means your cross sectional area wanders from ideal. Best case, weld/fill the negative side of the steps.

    Ps - Proud bellmouth always wins. Even if the magnitude of benefit is marginalized by the orientation of incoming air, it still has less loss than any other duct interface profile. And the 90deg entry eg is a great example of idea vs reality. Mind's eye says flow perpendicular to the proud bellmouth orientation would be bad, it isn't.

    Pps, harmonic length for forced induction remains the same, just you need to keep that fatter boundary layer in mind when looking at the calculated runner dia. It's hard to say how much fatter the runners need to be, as its dependent on how the boundary layer conditions develop because of bellmouths, bends, taper, surface finish etc etc. best I can chip in is that forced induction simply requires a slightly bigger dia than the same call for NA, and that will increase the more pressure you add.

    Ppps, boundary layer increases with increasing flow. So if you have a shitty intact tract, it'll act 'fat' at low flows, and thin right up at higher flow rates. It's a dynamic thing that's really hard to nail down, and since cars operate through a a broad range of flow conditions, it's very very hard to make an optimal 'suit all' solution. Hence OEMs using variable intake geometries etc. race cars need to pic which rpm/load condition is where the money is at.

    Straight shot intakes with a nice bellmouth and a gradual (1.5%) taper to the port throat minimizes the flow fuckups you'll otherwise encounter.

    Pppps, I have no idea what an offset runner is
    Last edited by doctor ed; 14-02-15 at 08:13 AM.
    Mit freundlichen Gre

    Quote Originally Posted by Keith Duckworth View Post
    "I think that in a racing engine, the closer it is to disintegrating, in general the better its performance will be "

  21. #21
    Registered User
    Join Date
    Nov 2005
    Location
    Wrong.
    Posts
    7,157
    Where is your site? do you not have a backup? I really enjoyed reading it.

    Speedtalk all that's left for casual reading technical stuff I find interesting.

  22. #22
    Registered User
    Join Date
    Apr 2008
    Location
    NSW
    Posts
    557
    Quote Originally Posted by doctor ed View Post

    Ps - Proud bellmouth always wins. Even if the magnitude of benefit is marginalized by the orientation of incoming air, it still has less loss than any other duct interface profile. And the 90deg entry eg is a great example of idea vs reality. Mind's eye says flow perpendicular to the proud bellmouth orientation would be bad, it isn't.
    Always is a strong term, this radiused design was about 1% better for flow over the bellmouth design…..the devil is in the details and an engine is about making the best compromises


    Last edited by E30is; 14-02-15 at 09:40 AM.
    E30 325is M20 3.1L

  23. #23
    Registered User Dudeman's Avatar
    Join Date
    Oct 2011
    Location
    Perth
    Posts
    365
    That simulation doesn't mean much, runner ends moved towards centre axis of plenum entry would make a fair difference.

  24. #24
    Registered User
    Join Date
    Apr 2008
    Location
    NSW
    Posts
    557
    Quote Originally Posted by Dudeman View Post
    That simulation doesn't mean much, runner ends moved towards centre axis of plenum entry would make a fair difference.
    thats kind of the point.........i said erlier that dont assume a bellmouth is always the best option as it depends on the inlet position etc this simulation is just illustrating this.....if there were no packaging limitations you could easily fix it but sometimes there is shit in the way
    Last edited by E30is; 14-02-15 at 11:05 AM.
    E30 325is M20 3.1L

  25. #25
    Registered User 333pg333's Avatar
    Join Date
    Feb 2011
    Location
    Sydney, Australia
    Posts
    2,278
    Not sure I have the brain space and time to learn how to fly that Pipemax program and spit out a worthy result inside a month...might try it in the end though.

    Failing that we are still faced with some actual constraints. As you can see in the 1st pic of this thread we have to deal with the cooling core ducting plus the turbo itself. So by shortening the runners further we will be bringing the plenum closer and therefore have to figure out where to locate the throttle body. Made me think of doing one with the centrally located t/b. Any feelings on these shaped versions? Problem with this could be heat from turbo into t/b and tps. Not sure how much shielding would help.

    Is there a general or even approx rule on plenum volume? Seems to be varying opinion on plenum shape too. Some favour the tapered where others don't. Is the single shaped log style really just about simplicity, affordability and packaging?

    Packaging aside, if we just bought a pre fab plenum similar to both shown here, stuck some shorter, bigger id and tapered runners with bellmouths how bad could it be? Realising that this is the 'hit and hope' syndrome.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	Reverie.jpg 
Views:	24 
Size:	57.6 KB 
ID:	81614   Click image for larger version. 

Name:	2014-08-02 09.49.13.jpg 
Views:	25 
Size:	2.54 MB 
ID:	81617  
    Attached Images Attached Images  
    Last edited by 333pg333; 14-02-15 at 11:34 AM.

  26. #26
    Registered User
    Join Date
    Apr 2008
    Location
    NSW
    Posts
    557
    in pipemax according to Larry to do FI set it to 100% VE, use the same length as for NA and the CSA for 240-260fps.
    E30 325is M20 3.1L

  27. #27
    GTFO of my ED doctor ed's Avatar
    Join Date
    Jun 2002
    Location
    irgendwo
    Posts
    4,703
    E30is - those aren't proud bellmouths, they're protruding. Basically want the rolled lip of the bell turning back and touching the floor. Ie 10mm proud. Always is a very unambiguous word design constraints of course dictate limits but I n a 1:1 comparison, regardless of flow orientation, rolled edge wins.
    Last edited by doctor ed; 14-02-15 at 06:36 PM.
    Mit freundlichen Gre

    Quote Originally Posted by Keith Duckworth View Post
    "I think that in a racing engine, the closer it is to disintegrating, in general the better its performance will be "

  28. #28
    GTFO of my ED doctor ed's Avatar
    Join Date
    Jun 2002
    Location
    irgendwo
    Posts
    4,703
    Plenum volume is debatable.

    On a closed pressure system like a turbo, I'd be going to the smaller side, trying to reduce your turbo-valve volume, increasing responsiveness. Something like 1.5-2x engine volume. Shape is really guesswork in the absence of CFD, it's all about cylinder distribution. Hit up Google image search for CFD runs of similar shapes you're considering, you'll see a trend at least, and get a gut feel for what will work best with your layout.

    Ps, in the absence of pipemax, nor knowing too much about your engine, I'd be guessing runner length needs to be around 220-240mm from the valve, and a 45mm ID runner on a long taper running to a 90mm bell. Total guesstimate
    Mit freundlichen Gre

    Quote Originally Posted by Keith Duckworth View Post
    "I think that in a racing engine, the closer it is to disintegrating, in general the better its performance will be "

  29. #29
    Registered User
    Join Date
    Apr 2008
    Location
    NSW
    Posts
    557
    Quote Originally Posted by doctor ed View Post
    E30is - those aren't proud bellmouths, they're protruding. Basically want the rolled lip of the bell turning back and touching the floor. Ie 10mm proud. Always is a very unambiguous word design constraints of course dictate limits but I n a 1:1 comparison, regardless of flow orientation, rolled edge wins.
    still falls short, ...

    Last edited by E30is; 14-02-15 at 07:06 PM.
    E30 325is M20 3.1L

  30. #30
    GTFO of my ED doctor ed's Avatar
    Join Date
    Jun 2002
    Location
    irgendwo
    Posts
    4,703
    You're comparing different sized bellmouths with differing bell and throat radii

    It's a packaging / fabrication constraint of your design.
    Last edited by doctor ed; 14-02-15 at 11:01 PM.
    Mit freundlichen Gre

    Quote Originally Posted by Keith Duckworth View Post
    "I think that in a racing engine, the closer it is to disintegrating, in general the better its performance will be "

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •