Big Bore Exhaust Manifold for 2V Engines | FerrariChat

Big Bore Exhaust Manifold for 2V Engines

Discussion in '308/328' started by bill308, Oct 1, 2019.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, Skimlinks, and others.

  1. bill308

    bill308 Formula 3
    Silver Subscribed

    May 13, 2001
    1,160
    Windsor, CT
    Full Name:
    Bill Sebestyen
    Does anybody offer a big bore exhaust manifold set, for up-rated 2V engines? OE primaries are 1.5 inch OD. I'm thinking 1.75 inch primaries, 4 into 1 style.
    Bill
     
  2. sp1der

    sp1der F1 Rookie

    Jan 10, 2009
    3,005
    UK
    Full Name:
    Simon Ashley
    you could try GTO racing in the UK
     
  3. mike32

    mike32 F1 Veteran

    May 13, 2016
    5,828
    Isle of man- uk
    Try P D Gough in Nottingham, uk
    They do all sorts of exotics in stainless
     
  4. Jonny Law

    Jonny Law F1 Rookie
    Silver Subscribed

    May 6, 2008
    3,173
    Over yonder.
    Full Name:
    IT Guy
  5. sp1der

    sp1der F1 Rookie

    Jan 10, 2009
    3,005
    UK
    Full Name:
    Simon Ashley
  6. mike32

    mike32 F1 Veteran

    May 13, 2016
    5,828
    Isle of man- uk
  7. bill308

    bill308 Formula 3
    Silver Subscribed

    May 13, 2001
    1,160
    Windsor, CT
    Full Name:
    Bill Sebestyen
    Thanks guys. I sent out several emails to the recommended folks and asked Burns Stainless to run an analysis and recommend a system specification, header to tail pipe.
    I'm having an ongoing discussion with Richard at pdgough. He offered to do a system based upon the euro model, but one size up on tubes. For me that means, 1.625 od primaries. Flow area goes up about 16%, not much more. 1.75 od primaries gives a 36% increase in flow area. My engine went from 220-->350 hp, about a 60% increase in power, if I can sort out the intake flow restriction.

    Bill
     
  8. mk e

    mk e F1 World Champ

    Oct 31, 2003
    12,899
    The twilight zone
    Full Name:
    The Butcher
    They did the design for my V12 system....I suspect they will tell you 1.375 is correct for a 3l v8 making under 400 , maybe 1.5" but that seems quite large 400-500hp , 1.625" more like 500+hp.Mine are 1.625", 458cc cylinders that should make about 70hp/cylinder.
     
  9. Jonny Law

    Jonny Law F1 Rookie
    Silver Subscribed

    May 6, 2008
    3,173
    Over yonder.
    Full Name:
    IT Guy
    I would like to see a complete 2v GTBi system with the oversized pipes. What sort of back pressure does this motor need?
     
  10. bill308

    bill308 Formula 3
    Silver Subscribed

    May 13, 2001
    1,160
    Windsor, CT
    Full Name:
    Bill Sebestyen
    Hi mk e,

    Thanks for your input. My exhaust ports are less than 1.5 inches ID. OE header primaries are 1.5 inches OD.
    The folks at Caribou Engineering, didn't seem to think OE headers were a bad choice. They may be right. The 26R header on my Lotus is a tri-Y with 1.625 inch primaries. We got 190 corrected hp out of 1.93 liters. I'm looking for 340 hp, from 3.4 liters, for the street. To get there, I need to address the inlet air system, in addition to the exhaust.

    Other features I need are OE style gas sampling ports and euro style radiation shielding. It may be possible to find high temp cloth/insulation to envelope the headers and attenuate exhaust noise and engine bay heat. Ceramic coating could also help.

    I think the only difference between the GT(i) systems and the carb systems, are the gas sampling ports, which make no sense on OE (i) injection systems.

    I wonder if there are any flexible, Aerogel composites, from which a envelope could be made?: Kind of like a big, high tech oven mitt. I wonder how it would stand up to vibration?

    Bill
     
  11. Tubi Sales

    Tubi Sales Karting

    Jul 5, 2019
    172
    PHX
    We offer a great manifold for the 308. The runners are 38mm (1.5 inches) to our main piping which is 54mm (2.13 inches).
     
  12. mk e

    mk e F1 World Champ

    Oct 31, 2003
    12,899
    The twilight zone
    Full Name:
    The Butcher
    100hp/liter is a pretty easy number to hit. See what burns says. when they "design" header you get somehting like this:
    Image Unavailable, Please Login


    It took me a while to figureout how exactly their design worked and get it modeled in the engine simulation (dynomation6) correctly...its very clever
    Image Unavailable, Please Login
    Image Unavailable, Please Login

    When it has to actually be built.....I start with foam "noodles" with an 1/8 alum welding rod stuffed inside to hold the shape
    Image Unavailable, Please Login

    Then bought a butt load of tubing
    Image Unavailable, Please Login

    made flanges
    Image Unavailable, Please Login

    then headers.

    Image Unavailable, Please Login

    but thats just the primaries, and these are tri-y and stuffed in a small place so stuff gets weird looking
    Image Unavailable, Please Login
    Image Unavailable, Please Login

    Image Unavailable, Please Login

    Image Unavailable, Please Login
    Image Unavailable, Please Login


    Its a lot of work, 100+ hours I think in the end I think....making the OEM or tubi option very very tempting I'd think :)
     
  13. bill308

    bill308 Formula 3
    Silver Subscribed

    May 13, 2001
    1,160
    Windsor, CT
    Full Name:
    Bill Sebestyen
    Hi mk e,

    You do some really nice work. I agree, 100 bhp/liter is reasonably easy to attain.I was going for torque, economy, and drive ability.

    Here is Burn's solution:
    Image Unavailable, Please Login
    The interesting thing is the stepped primary that is recommended. I've not see this feature before. Presumably, hot gases are gradually slowed down in the primaries, by both the increase in flow area and cooling of the the gas.

    This design is not all that different than the OE header, which uses 1.5 inch primaries and suggests a 1.625 inch version would also be appropriate.

    Can you remind me about the relationship between length and acoustic resonance? In the vibration world, we don't worry about harmonics beyond the 3rd mode or so, because effects (displacements) are so small at the higher frequencies.A similar result is likely true in the acoustics world, so that only the the first 3 or so harmonics are important and higher ones ,not so much.

    So what to do? I have one, OE NOS rear header and a reasonably good front bank header. Should I go with what I have, or commission a custom system with 1.625 inch or stepped primaries?

    Bill
     
  14. derekw

    derekw Formula 3
    Silver Subscribed

    Sep 7, 2010
    1,520
    London, UK
    Full Name:
    Derek W
    The way I understand it (probably a simplification for me to make sense of it) is that these small steps in the primaries reduce reversion by creating a negative pressure pulse that heads back towards the exhaust valve while the negative pulse heading downstream reduces the magnitude of the positive pressure pulse.

    It does sound like you're trying to have your cake and eat it too Bill-- good torque/drivability and 100hp/litre from a 2V seems a stretch but I'm a lazy driver so your "drivability" may be different to mine :)

    Maybe cut the OEM headers at the step length and build the rest from there?
     
  15. bill308

    bill308 Formula 3
    Silver Subscribed

    May 13, 2001
    1,160
    Windsor, CT
    Full Name:
    Bill Sebestyen
    Really good input Derek. Larger primaries result in a larger reflective, or reversion area, at the head interface. Does one get two reflective pulses from the two reversion surfaces in stepped inlet tract? I could see two surfaces generating two smaller reversion waves for a single exhaust event. A single reversion surface would return a single, larger in magnitude, reversion pulse. Compared, they probably offer near equivalent energy benefits, but the timing is different, in that the two pulses reduce and spread out the effects relative to a single equivalent pulse. Is that good at the exhaust valve end? My high lift/short duration cam has modest overlap.

    The engine was very well behaved on the dyno, where idle was at about 1300 rpm IIRC. To minimize costs, we spent little time at idle, so never worried too much about carb synchronization. We were concentrating on getting it to run, get some time on engine under varying loads and rpms, and make some performance pulls. We made one ignition timing change, from 35 BTDC to 32 BTDC for more poor fuel tolerance, and only one "jetting" change on the dyno, just changed the air correctors to smaller 180's, to richen the high end. I'll get it a lot better for the road where I'll set synchronization with the aid of bank of 8 mercury manometers measuring MAP through the re-purposed charcoal canister purge ports. Idle mixtures are set using a CO meter sampling gas from each cylinder and over the road mixtures are monitored with a wide band meter.

    I guess I am trying to have it all in a carb car. Can I get 25 mpg at an 80 mph cruise? Next issue will be the inlet/filtration system which drops me from 350 to 320 bhp.

    Bill
     
  16. mk e

    mk e F1 World Champ

    Oct 31, 2003
    12,899
    The twilight zone
    Full Name:
    The Butcher
    That is a peak hp type design....what rpm did you tell burns you were interested in?

    The reason for the 2 pipe sizes is flow. For best flow on the intake you want a taper getting smallest at the valve seat, on exhaust you want a tapper starting small at the seat and expanding out. A stepped pipe acts a lot like a tapered pipe.

    Wave wise anyplace there is an area change there is a reflected pulse but if the change is small enough that reflection will be tiny and the assembly will act like a single tapered tube so the main reflection happens at the collector entrance. So they are suggesting a length of 27+1 for the nipples in the collector + the port length which is about 3....so 32" which is tuned to about 6000 I guess.

    Collector is basically the same deal....the "choke" at the end causes a reflection and terminated the tuned length. Without the choke the entire exhaust system will count as tuned length....which is why so many race cars used to love open headers, the tuning is all buggered and wrong with the pipes on.

    Tri-y headers like mine work the same way....if you look at the notes I added the primary+1st collector forms a stepped primary on the 2nd tuning pulse (3rd pulse is most common). The long by tri-y standard first primary is tuned to about 1500 rpm lower on the 4th pulse and 2 different collector tunings. The goal is to widen the power curve as much as possible on what would want to be a peaky setup.

    This is a simple calculator to help you play around:

    http://www.wallaceracing.com/header_length.php

    Better programs are pipemax or I just use dynomation6 for everything but it requires a bit more detailed inputs to get outputs. If you want to give me all the required specs I can plug it into DM, but I need quite a bit of data like full cam and flow specs.

    Understand that there is no such thing as "best" or "right" there are bad choices and better choices and it can be hard to tell the better choices apart on the dyno.
     
  17. mk e

    mk e F1 World Champ

    Oct 31, 2003
    12,899
    The twilight zone
    Full Name:
    The Butcher
    Here's a couple screenshots from DM6 at peak torque 7000rpm and peak hp 9000rpm...there is a lot going on in the intake and exhaust manifolds and at much higher pressures than you might guess. Notice the intake manifold is over 1.8bar positive pressure at intake valve close and the heades making giving about 05.bar vacuum sduring the valve overlap period.....which is how a naturally aspirated engine can get to 160+hp/liter and 125+% volumetric efficiency.

    Image Unavailable, Please Login

    Image Unavailable, Please Login
     
  18. mk e

    mk e F1 World Champ

    Oct 31, 2003
    12,899
    The twilight zone
    Full Name:
    The Butcher
    here is what reversion looks like. 2600 the exhaust is quite positive during most of the overlap and in particular near exhaust close. then the bottom graph is the same pressure traces but now the green line is trapped mass in the cylinder, and you can see its dropping at IVC, this shows about 12% charge loss to reversion
    Image Unavailable, Please Login

    Image Unavailable, Please Login
     
  19. bill308

    bill308 Formula 3
    Silver Subscribed

    May 13, 2001
    1,160
    Windsor, CT
    Full Name:
    Bill Sebestyen
    Hi Mark,

    What are the horizontal-axis units on the pressure plot?

    I didn't get a chance to print my input form to Burns so I asked Vince to send me a copy of my inputs. As I recall, I may have asked for max torque at too high an rpm. There were a couple of Items I was not sure of like intake valve size. We went bigger, but not sure the new size. Vince will run a system analysis once I address tail pipe size and torque range to tune to. I believe the it's best to tune the exhaust to enhance mid range torque and tune intake to enhance high rpm torque. This is another strategy for spreading the torque curve.

    So where to tune the exhaust for my driving style? I'm thinking 3500-5500 rpm might be good a good range to tune to. Any thoughts?

    This is a comparison of corrected hp for 10 dyno runs using an available set of race car (inline 308 IMSA car) headers, as the OE could not be easily adapted to the inline needs of the dyno.

    Image Unavailable, Please Login

    These are the corresponding torque curves:

    Image Unavailable, Please Login

    Run 16, below, uses the full up air filtration system with the closure flap physically wired open. This is about what I expect when I first get it back on the road. I'll collect some data with the OE intake filtration system, including some pressure drops, and see what can be done. Note the power/torque dip in 4800-5700 rpm range. This is where the dyno is switched to computer control, which begins to vary the load to enforce a 300 rpm/s sweep rate.

    Image Unavailable, Please Login

    Bill
     
  20. mk e

    mk e F1 World Champ

    Oct 31, 2003
    12,899
    The twilight zone
    Full Name:
    The Butcher
    On the graphs I posted horizontal is crank degrees and 1 720 cycle is shown. That is what the waves look like through each and ever engine cycle.

    Your graphs. 1st, well done. when you were kicking out numbers I assumed it was estimated not measured so you are way ahead of the game. That's a nice engine.

    That dip....its pretty bad and I'm not buying dyno control method, its most like carburation....its pretty high pm to be anything else and it looks like stack changes didn't effect it so it can't be intake tuning. It could be exhaust but......I had a 1990 ZX6 motorcycle with a dip just like that at 9500rpm, redline was 14000, hp peak 12500. I used to roadrace it and it was pretty dangerous if I missed a downshift and ended up below 9500....around 9000 you start rolling in the throttle but as rpm climbs there is less power o you roll in more throttle then you cross 950 and the power goes up quickly with no throttle change and the rear tire comes loose and it tries to high-side....awful on the track. A dynojet kit and an hour on the dyno completely fixed it. I'm pretty sure its carburation.

    Do you know or can you ask the specs on the headers you tested with? Knowing the specs it would be a lot easier to understand what direction you might want to try form there.

    As for tuning once the porting is done and cam chosen the rpm shift has pretty much sailed. Then its about a bit of given and take at the margins. You can pull the hp peak to maybe 500rpm lower. You can move the torque peak a bit as that has a lot to do with collector design, port size too of course but the collector choice can probably shift it +/-500rpm, so yes the midrange is a lot about exhaust and the collector in particular.

    I guess remember that at peak hp the primaries will generally be on the 3 pulse, the collector on the 5th so between 0 and redline together they are in and out of tune 8 times but there is ever increasing energy in the pulses. Look at the 2600 vs 7k or 9k examples I posted.....the waves get much more powerful. The stacks are doing little to nothing below about 1/2 redline, only the collector is able to make meaningful pulses down there which is why its used to to help down there but its only gets more powerful as revs go up so don't discount it.

    On my engine the exhaust is tuned to about 8300, with peak hp around 8800-9000 by design.....but you're ahead of me with real data, I still have only simulations. Originally I had told Vince 8500, he tuned a touch below (well, his software did) as I said autoX use 2500-8500 and the widest curve I could get..so tuned a little low and tri-y which gives a very long collector when you combine the 2 collectors and tunes low.....but I'm expecting issues around 3000 +/- where everything is out of tune.
     
  21. bill308

    bill308 Formula 3
    Silver Subscribed

    May 13, 2001
    1,160
    Windsor, CT
    Full Name:
    Bill Sebestyen
    Hi Mark,

    That's some interesting pressure data. I'll do something similar in looking at the inlet system, from the carb air, box to the the scoop entrance. The OE system costs me about 30 bhp at mid to high rpm, so I need a better answer there. I'm thinking a plenum in place of the OE air filter box, with a remote (near the scoop) air filter box. I'll look at what is psia at various points along the inlet path?

    Thanks for the compliment on the data graphs. I've presented a lot of data to aerospace folks. Clarity, accuracy, and conclusions are important. Plots were done in Excel. Data was delivered in tabular form. The best way to see what is going on is to plot the results. Features, like the data dip in the plots, need to be questioned and hopefully explained. That makes it interesting.

    Given the apparent performance dip, especially in torque, I asked Vince to run a muffled case with a 2 inch inlet/collector, for tuning the exhaust to augment performance in the 4500-5500 rpm range, where the dip is located. Looking at the data, I would guess the test headers are tuned for the 6000+ rpm range. I think the donor vehicle was an IMSA 308 with an inline engine and maybe a Hewland.. My engine builder did both 2V and 4V cars in the day. They currently do the vintage engines for Miller Motor Cars, in Greenwich, CT. They recently delivered a 288 GTO engine.

    Image Unavailable, Please Login

    Inlet tract tuning is fixed, except for air horn height, on my engine. I think there needs to be more clearance between the face of the air horn and its ceiling. Inlet air flow, or more specifically pressure drops, at other points along the inlet tracts, need to be looked at. I wonder how many MAP sensors an LM2 can accomodate?

    I will find out details of the exhaust manifold used for testing. This is what the test setup looked like:

    Image Unavailable, Please Login
    Image Unavailable, Please Login
    I think porting on this engine was pretty limited. I suspect most of the attention was paid to integrating the larger intake valves and paying attention to the valve seat details.

    Bill
     
  22. Dal308

    Dal308 Karting
    Silver Subscribed

    Nov 4, 2014
    121
    Coppell, TX
    Full Name:
    Dude
    I never worked with Vince on Ferrari engines, but his designs for the 4 cylinder engines always had a pinch point at the collector.....in fact the drawings here show that. In the four into one systems the size of the ring or merge collector point seemed to be critical. Maybe be firing order of the Ferrari engine makes it not necessary. It was usually quite a bit smaller than the rest of the exhaust system.
     
  23. derekw

    derekw Formula 3
    Silver Subscribed

    Sep 7, 2010
    1,520
    London, UK
    Full Name:
    Derek W
    The flat plane crank means it behaves just like a 4-cyl.
     
  24. mk e

    mk e F1 World Champ

    Oct 31, 2003
    12,899
    The twilight zone
    Full Name:
    The Butcher
    On the headers in the dyno picture the entire length from where the 4 pipes enter to where the pipe ends IS the collector. Its very long and now that I see it the it could also be responsible for the dip...a race team wouldn't much care about a 5k dip on am 8-9k engine but not ideal on a street engine.

    The choke shown on the Burns headers effectively ends the collector length at the small diameter point, then you can go up to any size you like that is large enough to support the flow for the remained of the system without impacting system tuning This became pretty standard about 2000 iirc, older setups don't have it as it wasn't invented yet. The smaller the diameter choke the more effective it is at wave termination, but the more restrictive it is to flow so there is a balance to be struck. When you have burns do the design they want you to buy the collectors from them so exact dimensions for choke and overall collector length are missing until you pay I guess....most of my collectors ended up in very unhelpful locations and all kinds of bent so I back calculated everything and built them myself, but did buy all the tubing and anything else Burns sold that I needed from Burns.

    The 2.5" exit shown is good for something like 200-250/bank so about perfect, a 3" is maybe 400/bank, 3.5 is maybe 500....depending on system length so I'm 3.5 on the longer front bank pipe and 3" on the almost nonexistent rear bank pipe.

    I'm certain the Burns headers will be very good....its just a question of getting them made, but nothing 308 is cheap so custom might no be much worse than the off the shelf options if you can find a shop to do the fab work. I think I spent about $3k on materials for the whole system, but I have 50% more pipes and the tri-y so 3x the number of collectors. I used 18g 304 for almost everything except a couple bend radii that only came in 16g...its pretty light for how much pipe there is.
     

Share This Page