Engine and Transmission Tech

> Wow, lots of Vizard fans here.

Yup. Vizard turned me on to Dynomation.

> I've noted that secondary length is important (seen 18" to 24" as good rule
> of thumb)... but your note is the first I've seen keying in on collector
> dimension importance for the SBF. Am I making a mistake in nomenclature - in
> this discussion are you using collectors and secondaries interchangeably?

Secondaries typically apply to tri-y headers where you have a primary length
(applies to 4 tubes) which then merge into two short collectors, followed
by two secondaries which then merge into a single collector. The collector
length I refer to is specific to V8 headers where all four tubes (per side)
go into a single collector. To further complicate matters, there are various
styles of collectors including tri-y collectors. The Flowmaster 4-2-1
collectors are of this type. I refer to pipes aft of the collector as
intermediate pipes and pipes after the muffler as tail pipes.

> If not, what guidance can you give on collector dimensions?
> The Edelbrock heads I'm using are the Performer RPM - 1.6 inch exhaust
> valve / recommended 1500 to 6500 rpm range:

See Vizard's equations above and my recommendations specific to SBF V8's.
Unported Performer RPM heads flow around 175 CFM @ 0.500" lift and 180 CFM
@ 0.550" lift. Assume a 400 HP goal.

1.66" primary diameter
primary length of between 24 and 36 inches
collector diameter of 2.91"
collector length of 20 to 24 inches
dual mufflers of 440 CFM flow

The flow bench numbers above are with a pipe stub so are optimistic for our
purposes and the exhaust port blows down early in the lift cycle so use a
slightly smaller primary diameter. As a starting point in my Dynomation
simulations, I'd use:

1 5/8" = 1.625" primary diameter
3" collector (I'd prefer 2.75" but not many commercially available)
20 to 24 inch collector length
2 1/4" intermediate pipes leading to 2.5" inlet Magnaflow mufflers

Commercially available SBF 5.0L headers typically have primary lengths between
28 and 31 inches long. Hedman makes 1 1/2 to 1 5/8 step 4-into-1 full length
for both classic and Fox body Mustangs. There are also tri-y designs which
might package more conviently.

> Having a hard time finding commercial headers for the SBF that have primaries
> of larger diameter than 1.6" and will fit my car (short / center dump /
> non-block hugger - need to clear 23" outside width of frame rails / 1958 AH
> 100-6 on a custom Jule Enterprises frame).

Most long tube SBF headers fall into early Ford/Mustang designs which turn
down sharply at the exhaust port to clear the narrow shock towers or later
Fox body Mustang headers which are designed for a more generous engine
compartment. There are also a variety of race, swap and street rod headers.
The only center dump designs I've seen were street rod block huggers.

> I'm assuming I should be going with headers with 1 5/8" ID primaries because
> of the heads I'm using... I expect the primaries should be ~18" long and lead
> to 4-into-1 collectors with conventional 2.5" ID terminations that will be
> plumbed down to ~24" long 2 1/4" ID secondaries leading to a 3" ID resonator
> (maybe a Dynomax 'X' pipe / killing two birds with one stone / expansion to 3"
> used to tune length of secondaries) before plumbing to a pair of Dynomax Super
> Turbo mufflers (or comparable) that flow ~440 cfm each. I will likely follow
> your recommendatin and not skinny down the exhaust to 2" after the mufflers as
> was previously considered.

See above for my recommendations.

> Afraid I might be going custom on yet another aspect of this project for the
> headers... In that event, you noted that bend radiuses are important - any
> specific guidance assuming 1 5/8" ID tube?

Not really. The chassis usually dictates the bends.

Dan Jones

This is an outstanding thread!

One sub-topic I'd like to see discussed in a little more depth is optimization of tri-y headers for a V8. Which of the 4-into-1 rules of thumb apply, and which ones don't? Is it true that the diameter of tri-y primaries can reasonably be smaller? Sized to match intake port size? How long should they be? I'm thinking of building a set one day for a Rover 4L motor.

Dan,

You pretty much mapped it all out there didn't you...

Thank you so very much!

I loved reading that your modeling software was written in Fortran. I learned to code in Basic, Cobol, Fortran and Pascal my freshman year of college - 28 years ago!

If you need a hand buying or selling a company (Internet media has been my specialty of late), give me a ping and I'll be happy to give you some advice!

Martin Jansen of Jule Enterprises, the party building my car for me, asked me if he could start photographing my car as it comes together and posting the pictures on his site for other 'modified-Healey' guys to see. I told him I'd love it. So if you are interested in how your guidance gets applied you should go to Martin's site and look for my car in a while (Jule Enterprises is listed as a sponsor of BritishV8 - I expect the moderator won't mind the plug).

Thanks again Dan! Your knowledge is obviously deep, and your advice worth heeding. Best of luck with your many projects.

Cheers,

Mark

Dan, not to take anything away from other replies, but you appear to be the one that excels on this subject. I feel like some poor guy showing up to a gun fight with a knife (LOL).

From the beginning of this MGB project, I proceeded with a plan based on personal experiences, member's feedbacks and wishing like heck that my approach later would not result in major adjustments and redo. You asserted some of my choices based on similar rationales, except that my choices were based more on wishful deductions.

I'm now 58 years old and semi-retired; during my career as a mechanic I've worked more as a repair technician in a large automotive fleet. Our company sent us to seminars and training through Ford, GM and Chrysler. Basics and fundamentals were learned based on street vehicle to address its preventive maintenance, repairs and components overhaul based on factory specifications and procedures.

Hot rodding modification has always been a weekend interest of mine, again, based on factory high performance vehicles. During my younger years, there were a lot of "snake oil" products purchased with great disappointment. What I'm trying to say is that unless you are a seasoned race mechanic that understands in depth fundamentals to all out performance, and, that has extensively worked to acquire a winning edge, than, the rest of us have to look for these "Guru" out there.

I'll answer some of your questions concerning my 331 FSB heads and intake choices; when I started the engine build up consideration, I decided to give a call to "Air Flow Research". I've explained to them that I wanted to build the car mostly for joy ridding to include an occasional fun street fight. First question asked, was what camshaft selection was used; since I already had purchase the Ford Racing, roller E-303, they recommended the AFR 165 heads with Scorpion 1.72 roller rockers. Only later did I decide to go with a stroker kit. I've called them back to ask if the heads and matching components were still adequate, they asserted that the "low end" torque would be pleasing with the way their heads flow and the new combo.

It was AFR that suggested using a 1-5/8" primary tube and 2-1/2" pipes. They indicated that because I was using the Ford Racing GT-40 EFI intake and the AFR 165 smaller valves (1.60 and 1.95) were going to create real nice air velocity. The rest of the engine is all 4340 forged steel. A 65 MM throttle body and 24 pounds injectors finish off the induction side. An aluminum flywheel was added since plenty torque allowed a trade off between inertia and quick revving.

All the same questions were asked concerning, horsepower, torque, and how could fuel efficiency find a spot by addressing good EFI management. It has a new Summit Racing T-5 with a tall OD, tires are P205/60VR15 and the differential ratio stands at 3:42.

Last but least, is the exhaust tuning to work in harmony with the above and that is why I started this thread. So far, because of the stellar answers given, I have a pretty good grip on how to proceed. If it wasn’t for of all you accomplished performance members, I don’t think that anyone could find answers in a factory repair manual.

I forgot to attach the link to my AFR 165 heads; it has awesome flow numbers for up to 331 cubic inches: [www.airflowresearch.com]

Jacques

> One sub-topic I'd like to see discussed in a little more depth is
> optimization of tri-y headers for a V8.

Unfortunately, my simulation program doesn't have an option for tri-y
designs. Vizard's dyno testing experience suggests they work well for
relatively mild cam timing (under 270 degrees seat duration) and are
relatively insensitive to length. He suggests the primary/secondary
split be on the order of 1/3rd to 2/3rd the overall length. There's
a header sizing program called PipeMax that supposedly does tri-y's
well. I've been meaning to look into that one of these days.

> Which of the 4-into-1 rules of thumb apply
> Is it true that the diameter of tri-y primaries can reasonably be
> smaller?

I'd still use the primary pipe sizing from above. That needs to match
the exhaust port flow capability (within the cam lift limits). After
the first merge, I'd step up a size.

> Sized to match intake port size?

They should be sized to the exhaust port size. The intake side is
larger because the exhaust is under higher cylinder pressure while
the intake side see only atmospheric pressure.

> You pretty much mapped it all out there didn't you...

Remember, those are just my starting points (and rationality checks) for
iterative Dynomation simulations but should be pretty close.

> I loved reading that your modeling software was written in Fortran.
> I learned to code in Basic, Cobol, Fortran and Pascal my freshman year of
> college - 28 years ago!

Sounds like we were in college taking programming around the same time,
though I actually was introduced to Fortran while in high school.

> If you need a hand buying or selling a company (Internet media has been my
> specialty of late), give me a ping and I'll be happy to give you some advice!

Thanks!

> So if you are interested in how your guidance gets applied you should
> go to Martin's site and look for my car in a while (Jule Enterprises is
> listed as a sponsor of BritishV8 - I expect the moderator won't mind the
> plug).

Cool, I'll watch for it.

> During my younger years, there were a lot of "snake oil" products purchased
> with great disappointment.

That's a motivating factor for me, trying to help others avoid mistakes
I and friends have made along the way. Back then, before the internet,
we had to work for every drop of information now it's forced through a
firehose and the problem is sorting out fact from fallacy.

> since I already had purchase the Ford Racing, roller E-303, they recommended
> the AFR 165 heads with Scorpion 1.72 roller rockers.

I'm using AFR 165 heads on my own '87 Mustang GT with plans to add a Novi 1000
supercharger. You'll probably want AFR's upgrade springs with the larger
rocker arm ratio.

> Only later did I decide to go with a stroker kit.

If you're using the 58cc AFR heads, you should be at around 10.4:1 compression,
closer to 10:1 with the larger chamber size. If using the SCAT kit, be aware
it is designed for a 28.2 oz-in balance factor. Your balance and flywheel need
to be designed for the same. If using a 5.0L roller block as your donor, be
aware it was 50 oz-in. Pre-'81 302's use a 28.2 oz-in balance weight, '81-up
5.0's are 50 oz-in, as are all 351W's and 351C's use 28.2 oz-in balance factors.

> I've called them back to ask if the heads and matching components were still
> adequate, they asserted that the "low end" torque would be pleasing with the
> way their heads flow and the new combo.

Yes. Those heads will work just fine with the cubes and other componenets.

> It was AFR that suggested using a 1-5/8" primary tube and 2-1/2" pipes.

Agreed.

> They indicated that because I was using the Ford Racing GT-40 EFI intake and
> the AFR 165 smaller valves (1.60 and 1.95) were going to create real nice air
> velocity.

True. You can pick up some power by porting the lower manifold of the GT40
intake set. It's cross-sectional area will be a limiting factor in your
build.

> The rest of the engine is all 4340 forged steel.

For a build like this, the weakest link are the two bolt main caps walking
around at high RPM. However, with the rev limiter of the EEC-IV and the
combination of parts, you won't be spinning it high enough where that will
be a problem. The Melling M-68HV's is a great SBF oil pump and it's worth
adding an FRPP Boss 302 windage tray. The Boss 302 windage tray works with
front sump pans and MPG Head Service has a similar part that works with
the Fox body double sump oil pans.

> 65 MM throttle body and 24 pounds injectors finish off the induction side.

I'd rather see 70mm and 30 lbs/hr injectors. You may be at the limits of
the injector flow rate, though that can be crutched with more fuel pressure.

> An aluminum flywheel was added since plenty torque allowed a trade off
> between inertia and quick revving.

Good move. You'll notice the difference, particularly in the lower gears.
I'm pretty sure I posted the analysis I did on the effect of running a
lighter flywheel.

> All the same questions were asked concerning, horsepower, torque, and how
> could fuel efficiency find a spot by addressing good EFI management. It has
> a new Summit Racing T-5 with a tall OD, tires are P205/60VR15 and the
> differential ratio stands at 3:42.

Traction will be a problem in lower gears.

> Last but least, is the exhaust tuning to work in harmony with the above
> and that is why I started this thread.

1 5/8" headers, 2 1/2" pipes and good mufflers should work just fine.
Most of the 1 5/8" primary diameter long tube headers will have a 3"
collector but a smaller 2.5" collector will usually make better power
overall.

Make sure you use a good timing chain. If your chain says Rolon (made
in India), do not use it. Be aware that most companies offering timing sets
do not make their timing sets. They repackage them from multiple vendors.
You need to know who made the sprockets and chains. One of the guys on the
FE board used to work at Speed Pro. When he worked there they had a bunch
of offshore companies trying to be suppliers, so they piggybacked some chain
and sprocket testing on an OE bearing durability dyno run. The Rolon chain
from India cost them the test motor a couple times when it failed before the
test was completed. A summary of the test results for the chains is listed
below. Speed Pro ended up using Dynagear sprockets and Morse chain but had
some quality control issues then Dynagear went out of business. After that
they sourced the high end Cloyes sets (which used high quality Renold and
Iwis chain) and the quality control complaints went away. The Ford Motorsport
timing sets I've purchased have used the Renold or Iwis chain and I've used
some of the higher end Comp Cams sets which used Cloyes. The sprockets come
from Rollmaster (Australia), SA Gear (US but poor quality) and Cloyes (US,
not pretty but good quality). Avon also makes some sprockets but sources
others. Dynagear (US) used to make sprockets but went out of business.
A bunch of the performance aftermarket companies are selling poor quality
chain from India (Rolon), along with sprockets from Australia (Rollmaster,
J.P Performance) or SA gear.

Chain durability testing summary:

Iwis (German) - looks very nice but was not tested, but has excellent
reputation as an OE supplier, used in high end Rollmaster
and some high end Cloyes sets
Cloyes (US) - tested OK
Renold (France) - tested excellent, used in most high end Cloyes sets but not
always
Morse (US and Mexico) - tested excellent
Daido (Japan) - tested excellent
Tsubaki (Japan) - tested excellent
KCM (Japan) - looks very nice but was not tested
Rolon (India) - failed test

Sprockets:

Rollmaster - Aus - pretty - never examined QA, good reputation
SA Gear - US - ugly - crappy
Cloyes - US - not pretty - good QA
Dynagear - US - out of business - so-so when they existed
Avon - some sprockets, buys everything else
Crane - buys everything
Comp - buys everything
Speed-Pro - buys everything
Ebrock - buys everything
Melling - buys everything
Elgin - buys everything

Note: Some of the test data was from Speed Pro, other from TRW.

The high end Cloyes and Ford Motorsport stuff tends to use the Renold chain
but the lower end Cloyes may use inferior chain. Rollmaster usually uses
Iwis chain which is top notch. I picked up an Australian Muscle Parts set,
thinking it was a Rollmaster but it turned out to have the inferior Rolon
chain from India.

> I forgot to attach the link to my AFR 165 heads; it has awesome flow numbers
> for up to 331 cubic inches: [www.airflowresearch.com]

I know them well and have a set sitting on my bench at the moment. The AFR's
cost a bit more due to the CNC runner and bowl porting but flow quite well
for the valve size. I ran your combo through Dynomation, assuming flat top
pistons, 58cc chambers for 10.4:1 compression ratio, long tube headers and a
GT-40 intake with ported lower and 70mm throttle body. The simulation predicts
around 385 HP. With an unported lower and 65mm throttle body, expect that to
drop into the 350 to 360 HP range.

Dan Jones

So, no further elaboration on the step down in tailpipe size I guess?

JB

> So, no further elaboration on the step down in tailpipe size I guess?

Vizard's testing has shown it to cost power to step down in pipe size
as it increases backpressure. Also most mufflers flow less than a
pipe of equal cross-sectional area so it's recommended for the muffler
to have at least the same ID and in many cases a step to a larger size
can help (e.g. 2 1/4" intermedaite tubes leading to a 2 1/2" inlet/outlet
tail pipe).

Dan Jones

Dan, thank you for your thoughts about tri-y headers. This thread is giving me the itch to fabricate something!

> Dan, thank you for your thoughts about tri-y headers. This thread
> is giving me the itch to fabricate something!

I'll be building and dynoing a 4.2L over the winter. Just got
the flow bench numbers back on the heads and designed the cam.
Should be good for 350 HP and be streetable. I've got a set
of small tube tri-y's, a set of Hedman 4-into-1 Range Rover
headers plus a box of headers tubes for something larger diameter.
If you bend up some headers, I can dyno them.

Dan Jones

Other than the "paint stripe", has anyone mentioned the "thermal factor"? I "suspect" OEM. reasoned smaller tailpies were an acceptable compromize, because of thermal losses ? Quite a bit of temp/pressure loss between point"A" and "Z" ? Production cars are a combo of thousands of compromizes. Almost always, the tailpipes must overpass the rear end.For this task, the smaller pipes are a fit. 2 cents, roverman.

Art, that was sort of my point, thank you. I suspect my post got lost at the tail of the last page, as Dan and I posted at nearly the same time. Similar to a production car, a street driven LBC has to have a lot of compromises, and where a big diameter exhaust system or a 180* header may be fine on other cars they will often be seen to create more problems than they cure here. Same as a cast manifold can be seen as part of the progression to an uncompromising set of headers, so too can the tailpipe size, and the best is not always the right choice. Equal length 1-5/8" primary, 3" collector headers in a lowered MGB are going to be quite challenging, and I'm not even convinced it can be done, at least not without a rock hard suspension, big cutouts in the fenderwells and you still might have a problem scraping the exhaust. If oversized tires have been fitted interference there is likely to also be a problem. The 2-1/2" or larger tailpipe is very likely going to have to go under the rear axle rather than over it and I suppose that's fine for all out performance but I did that once and won't do it again. You can't argue with test results and it'd be nice to know exactly what David's test setup was and what results, but common sense dictates that a cooler gas can use a smaller pipe and the temp drops hundreds of degrees from head to tailpipe. Also the empty spaces between the exhaust pulses are filled in as the slugs of exhaust gas are stretched out, making the pulses less dense which should also require less area for efficient flow. So one would think that the need for pipe area would diminish the further back you go. Again, referring to dirt bike systems, often the spark arrester is actually smaller than the head pipe. As for the 2-stroke being a totally different animal from the 4-stroke, well, not so much really. Both use the scavenging effect, both have overlap, both have pressure pulses and partial vacuum. The 2-stroke benefits from a partial pressure in the transfer ports (some do) and double the firing rate, but aside from that they are very much the same and for the most part the same principles apply. Very likely a 2-stroke style scavenger pipe would work just fine on a 4-stroke engine. I don't remember what the newer 4-stroke dirt bikes are running specifically, but I never noticed a difference in them when they first came out, and as an item of much curiosity I think a big change would have been seen and pointed out by someone at the time. Head pipes were smaller but that's all I remember.

So getting back to the cast manifold. Not to name names, but one very compact (but sleek) block hugger design which is popular can't be any more efficient than some of the factory cast manifolds we've seen. Looks much better and fills a distinct need, and it *is* a header, but in layout and tube size they aren't as good as the later model GM V8 cast iron manifolds for the LS1, and certainly no attempt at equal length tubes. So all I'm saying, it's a point in the spectrum. Where the all out race car needs to be and where the street driven LBC needs to be are worlds apart.

JB

Jim and clan, Any gain for MGB's/British cars to run headers,"forward" of engine, and then sweep back? How about through rocker panels ? 2 cents,roverman.

Jim, what I've got out of Dan’s advance explanation (and what I've also read so far) is that there's a small window of opportunity to scavenge exhaust and help draft in a new oxygen rich charge; that small opportunity reveals itself at a hard to define exhaust location/length. Most of the factors that he talks about affecting this area are before the mufflers; any source after that becomes a possible restriction to that important scavenging/filling process.

In what I've also gathered, proper muffler inlets can help setting the tone for what ever good is supposed to take place in that muffler (sound control and flow, etc.) and how dropping the seize of the tailpipes can ruin the prior well tuned process by foresee ably applying back pressure. I just don't see any reasons to reduce the tail pipe seize other than making it easier to bend and fabricate smooth turns, however, I do see a single concern, "back pressure" but having said that I like yourself wonders where the justification/danger occurs? I’m also still puzzled on the “H” or “X” pipe principles?

Auto manufacturers are faster in drawing a line to avoid manufacturing complications and cost savings than a performance purist is willing to. After all these years, they've come to realize that some SEMA member will take on costly performance challenges for profit.

I’m not sure if other than getting a strong lesson in “performance exhaust” I have a better plan to operate by. I’m also still undecided if I should use an “H” pipe, mufflers and resonators, or just mufflers. I may also find it too difficult making 2-1/2” tail pipe bends, so what is the lost? One thing that I can be sure of is the fact that I don’t want anything to lesser the money invested in the full potential of my engine.

BTW, Dan I’m considering moving up to the 70 MM throttle body as you suggested since I’ve already have a set of 30 pounds injectors. You’ve mentioned a possible 385 H.P. on the engine dyno, what would translate to at the rear wheels? There’s also a worthy modification to do on the GT-40 intake that is well justified by the flow increase.

I hope that I don’t wear my welcome out on this thread but I really enjoy reading everyone’s comments or helping anyone out if I possibly could.

Jacques

In a perfect world there'd be no bends. (BTW, if anyone's wearing out their welcome it'd be me so don't worry, be happy.)

Dirt bike riders found that removing the spark arrester from the scavenger pipe made the bike louder but actually caused a slight decrease in power. So if a smaller tailpipe is always going to increase restriction, why is that? Shouldn't power go up? I'd say it's because it changed the resonance of the system.

I don't buy for a minute the idea that beyond the muffler the tuning of the system as a whole has no effect. I'm sorry for being hard headed about it, but the entire system from end to end does have a resonance, and that resonance can effect the engine I believe. If you are tuning for maximum power, then yes, perhaps going as large as is practical there makes sense, but if you are tuning for some measure of fuel economy, and therefore looking for resonance at a lower engine speed I think it likely that a smaller tailpipe makes sense. Does a longer exhaust system add nothing but back pressure? Or did the testers overlook other potential advantages because they weren't looking for those results?

Please understand, I'm not disagreeing with Dan's theories, but he is oriented towards maximum performance and as far as I know always has been. I've gone down that road but with any engine, going for maximum output not only has no limit, it's cost runs to infinity. So my approach is to use less expensive methods and seek to achieve maximum performance at minimum cost, both in development and in operation. There is much to be learned from Dan's approach that can also be used in mine, and perhaps occasionally vice-versa. So where Dan promotes a particular approach for power, and that will probably be appropriate for many applications including many forms of sanctioned racing, I am inclined to take more of an outlaw approach, and one which not only does not conform to racing rules in any forum, but tends to go outside generally accepted norms in other ways too. Because of that I will frequently challenge the generally accepted principles, and I will often be right. It doesn't mean you should follow me, but it may give you more to consider.

Now, I don't have the resources to do some of the testing Dan does, and I'm a great believer in test results. So where the test conditions are known and the results verified, I strongly believe in making use of the results. But before I will buy into a flat statement, especially about something as complex as an exhaust system, I'll have to see results from back-to-back testing of competing theories using setups that accurately portray those theories and do it under the appropriate test conditions. For instance, you may get more power across the board with a larger tailpipe in a given car, engine and exhaust system combination, but was the test also run to determine relative fuel consumption at different road speeds? Even if it was, if you change the vehicle, the engine, the headers, or even the muffler the results could be dramatically different. So I don't just automatically accept that what works for one car transfers straight across to another. The problem of course is that formal testing can be prohibitively expensive, whereas informal empirical testing is usually very time consuming, difficult, and also not cheap.

JB

Jim, as a youngster (17) I had a snowmobile with a two cycle Rotax motor. It ran great for a long time until I removed the muffler and replaced it with a megaphone; three engines later and many carb re-jetting, I went back to a muffler but a tuned exhaust type (funny looking). As far as I know, two cycle exhaust behaves totally different than a four cycle exhaust, "brother from an other mother eh!"

Yes I have heard that said, but why would it? So far nobody has presented any fundamental reasons why that would be the case. You have the expansion of the exhaust pulse in a scavenger type pipe of course, presumably to the point of collapse of the pressure front but that is by design of the exhaust header, not the engine. Is there any reason why the same principle would not apply to a 4 cycle design? I'm just asking. Obviously the biggest difference is that the 2-stroke has a lot more overlap, and without either a very good extraction or very good transfer function it just won't run, or won't run well. Fundamentally, why would good extraction be any less beneficient for a 4 stroke engine? Again though, this is theoretical because the practical considerations make an expansion extractor system for a V8 a non-starter, as it would take as much room as the engine itself, or more. But the example was to illustrate the point that the outlet is as small or in many cases smaller than the inlet. Are we to assume that because of the dynamics of the expansion chamber the outcome is somehow different? The "black box" representing the exhaust system between the port and the tailpipe somehow makes one tailpipe a restriction but makes the other just the opposite? It needs a better analysis to make that argument palatable.

So I'll go back to what I said before. You have to pick a point in the spectrum for your own application. In a perfect world with no compromises there are no bends and the exhaust operates at maximum efficiency, whatever that is, whether via megaphone or whatever is the best, most optimized approach. In the real world though we live with all kinds of auxiliary considerations and have to balance out the concerns of one against the other. Size of the tailpipe is one such, especially if routing it over the axle. Ground clearance is another. Obviously your mileage will vary greatly depending on where you put your priorities and you may well find that with a given vehicle and engine that a 2 or 2-1/4" set of tailpipes do not hurt performance enough in your operating range of the engine to make it that much of a concern. Would I give up 20 hp to route the exhaust high in an MGB with a 400 hp engine? I probably would for a purely street driven car, because I've already come to the conclusion that at 300 hp the car is more than capable of blowing the doors off at least 99% of the cars out on the street and to do so very convincingly. (That's a 8:1 weight to HP ratio BTW, 380 is about 6.3:1 and 400 is 6:1) The benefits outweigh the disadvantages. Besides, how often am I even going to see that one guy who I can't outrun, and even then how often am I going to feel like racing him? I don't live in a small town anymore and bragging rights just aren't that important. Aside from that, I subscribe to the Carl Floyd school of competitive driving, even if I do like horsepower, and the skill of the driver can often more than make up for a difference in the engine.

JB



Edited 2 time(s). Last edit at 11/09/2010 09:19AM by BlownMGB-V8.

Jim and clan, don't the rules change for forced induction ? Do you not enhance boost by cutting back on valve overlap and exhaust duration ? 1 3/8" for 600 hp. jus' don't sound rite, lol. roverman.



Edited 1 time(s). Last edit at 11/10/2010 11:20AM by roverman.

Funny Art ;-)

I never claimed 600 HP. Last I checked I think I was saying I expected to make something upwards of 300. I think 1-3/8" should support that just fine. (and my lobe centers are at 112*, not sure what the overlap is, it was supposed to be ground explicitly for a blower engine)

This is not your typical max performance engine, and is designed with specific goals in mind, including a wide powerband and extended rpm range (up to 7000). It is not expected to continue significantly building power above about 5500, but to be capable of the higher speeds to add flexibility in gear selection, i.e. not having to shift in the middle of a curve.



JB

> I’m also still puzzled on the “H” or “X” pipe principles?

Cross-overs work by one of two different principles. On an engine without
mufflers, if you could watch the flow in the cross-over, you'd see the exhaust
move only a few inches back-and-forth. In an open exhaust, the cross-over works
by acoustic tuning, transferring shock waves. On an engine with mufflers, they
allow the cylinders on one side of the engine to share the muffler flow of
the other side. While they don't double the muffler flow, a 25% increase in
apparent muffler flow is possible. Vizard has test both H-pipe and double
cross-over systems. The double cross-over looks like an X-pipe superimposed
upon a dual exhaust but is not a simple X-pipe. Vizard's testing indicates
the H-pipe version only works with mufflers that already flow very well in
relation to the engine's power (i.e. have low operating back pressure). Placement
should be as near to the header collectors as possible. On less efficient mufflers,
the double cross-over works better.

> You’ve mentioned a possible 385 H.P. on the engine dyno, what would translate
> to at the rear wheels?

A little over 300 RWHP (305 to 310 RWHP).

> Is there any reason why the same principle would not apply to a 4 cycle design?
> Obviously the biggest difference is that the 2-stroke has a lot more overlap

The fundamentals of tuning (inertial scavenging and finite amplitude waves) may be
the same but the required plumbing and the resultant effects are different between
2 and 4 strokes and between 4 strokes of different piston layouts (single cylinder
versus inline 4 versus 90 degree V8).

> going as large as is practical there makes sense

There are optimal lengths and diameters for a particular engine in a particular
operating range, exceeding them with larger sizes will reduce performance,
as will smaller. That said, some parts of the system are sensitive to sizes,
others are not (more on that below).

> he is oriented towards maximum performance

Performance can mean many things. It may mean maximum HP but it may also mean
maximum fuel economy. I try to maximize the system within whatever set of
parameters I'm handed. For instance, I'm currently working on one SBF that is
constrained by the diameter of the exhaust pipe that runs through the convertible
bracing and by a goal of 20+ MPG on the highway.

> So I don't just automatically accept that what works for one car transfers straight
> across to another.

Nor do I. That's why I've been testing Dynomation against a variety of dyno tests
on engines ranging from a 200 HP Rover V8 with tiny inline valves to a 650 HP 407
cubic inch Cleveland Ford with canted valve high port heads. The basic physics is
the same but those two engines behave very differently because one is limited by
the port area and the other is not. The fundamental differences between a 2 stroke
single cylinder engine and a 4 stroke V8 are much larger so why would you think
tuning for the 2 stroke directly applies to a 4 stroke? My 2 stroke experience is
limited but I've never seen any 4 stroke exhaust tuning respond like a 400 Kawasaki
2 stroke as it comes on the pipe.

Back to the problem at hand. Small and large re relative terms and some parts of
an optimal exhaust system are less sensitive to change than others. A straight
pipe of certain cross-sectional area flows quite well compared to a muffler with
the same ID. If you can't fit 2 1/2" diameter intermediate pipes, 2 1/4" pipes may
work well as long as you step up the muffler ID to 2 1/2". With the SBF we've been
discussing, reducing the collector outlet diameter to as little as 2.5" may slightly
hurt peak power potential but will likely have only a small penalty (or perhaps none
at all) on average power between the shift points. In the Engine Masters Competition,
the guy I run the Cleveland Ford dyno program with (Dave McLain) and Kaase both found
the best results with a 3" diameter collector on engines that made upwards of 660 HP.
That competition was scored on a best average power between 2500 and 6500 RPM. While
they could make better peak power with larger ID collectors, they gave up more power
lower in the RPM range so the averages suffered. Avoid reducing diameters at curves
like those at the axle hop as a curved pipe flows less than a straight pipe and use
smoothly bent pipe. I've not personally tested using a smaller tail pipe size to
see the direct effect but you can test it with a pressure gauge.

> I am inclined to take more of an outlaw approach, and one which not only does not
> conform to racing rules in any forum, but tends to go outside generally accepted
> norms in other ways too

Jim, have you read The High Speed Internal Combustion Engine by Sir Harry R. Ricardo?
Funded to develop piston engines for the war effort, Ricardo did so without any
regard to class rules. His work is at the base of everything hot rodders do, whether
they realize it or not.

Dan Jones