eclou

Mark,

anything that allows more exhaust flow will cause you to increase airflow into the engine and thus increase your fuel needs to maintain the proper AFR. Cats restrict flow, and thus could cause a car to run richer than the same car without cats. The engine computer can adapt the fuel mixture to a certain point over time - adaptation, but only to the extent that the injector size and fuel pressure will allow (there is a max). Air density increases (colder air) also leads to potentially leaner conditions just due to the increased # of oxygen molecules. Also, do your tuning on a dyno with a wideband O2 sensor. Tuning by stopwatch is like unprotected sex in Haiti.

Mark in Baltimore

Thanks for the info.

I know the G Timer is not ideal, but I'm not sure if I'm up for renting a dyno.

eclou

Dyno time is about $100/hour. Most likely you won't be able to increase the airflow thru the engine enough to max out your fuel injectors on a naturally aspirated car unless you do cams and headwork, but it is far safer to measure your gains and AFR with each mod. A good fuel injector calculator is here:

http://www.rceng.com/technical.aspx

Cheers

Peter R.

ok dude, what about this...

You send me a pair of them there sports cats. I'll pay for the shipping. I'll put the car on a dyno. Then I'll replace the old cat. I'll weigh it on a digital scale, install it and put the car on the dyno.

If I loose 20 lbs and gain 8 hp I'll pay for it. If I don't, it's on you.

Can't be fairer than that, can I ?

Peter R.

Bill Verburg

my comments in blue

Bill Verburg

Acoustice or "resonance" tuning does work well when you have an open exhaust and cams w/ a reasonable amount of overlap for the acoustic signals to work w/. Unfortunately w/ mufflers and smog cams w/ little or no overlap acoustic tuning doesn't work so well.

In any event the stock cast iron 993 headers are far from ideal wrt equal length & colector design. The best they can do is keep the pulses separated as far apart as possible untill the collector and to have as few bends as possible and to be correctly sized for expected flow volume and to provide heat.

eclou

The first reflection occurs at the collector, and realistically is the only important one for resonance tuning. Everything aft may lead to some higher orders but will have trivial contribution. Even with minimal overlap there is still some benefit, but surely not 20-30hp worth on an otherwise stock car. The stock manifolds are not optimal for performance, but for longevity, noise control, heat, cost, etc as you have mentioned.

Bill Verburg

First the 993 MAF is more resposive than the 964 barndoor, second perhaps w/ an open exhaust but even then i'm doubtful, third it's always best to check anyway wideband O sensors are widely available

No, you shouldn't have any lean issues even without a cat. How are you going to adjust timing? All fuel and timing parameters are totally controlled by the chip in the DME. I haven't installed my Car chemistry disks, I bought them to control noise but they will certainly raise back pressure and slow flow rate.

As noted in post 28 & 36 A correctly sized and constructed equal length header system w/ properly sized and constructed collector used w/ cams that have a fair amount of overlap can use the acoustic echos in the pipes to put a negative pressure signal at the exhaust port at the right time to suck additional charge into the combustion chamber. The pipes are like organ pipes, width and length resonate at different frequencies This is sometimes referred to as "the 5th cycle" or "exhaust scavenging". I've dynoed 3.0, 3.6 & 3.8 w/ stock and sport cams w/ stock and various aftermarket exhausts. It works best on open exhaust b w/ cams w/ a button up exhaust and stock cams the effect is greatly muted. The resonance effect only happens at 1 frequency range, w/ 31" true headers that is ~6500 for our engines, the down side is that way down the rev range the headers put a positive signal the reduces flow. You can both hear and feel both of these effects when driving.

Bill Verburg

There is a reflection form the collector but aagain there will be no extra charge in the cuylinder unless bioth intake and exhaust valve are open at the same time(overlap) in af=ddition every back to back comparison shows a relative insensitivity to pipe lengeth variation of <~10%. The collector merge and lenth is much more important. Open exhast can be simulated by terminating the post collector pipe an inch or more into the muffler.

eclou

Here is a simple primary manifold length calculator here, as Bill mentioned tuning is really optimizing for a narrow rpm range:

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

BTW here is a pic of the mis-fit of my Fabspeed Cat deletes on my TT:



I needed to grind off about 1/2" of each heat pipe duct to allow the unit to fit. Even then, the angle of 1 of the O2 sensor bungs makes 1 sensor impinge upon the crankcase itself.

eclou

Absolutely. But every cam does have some degree of overlap so there is always some scavenging possible. Collector design is more about reducing turbulence and flow restriction. Most collectors suck because of the expense to manufacture something with a nicer transition like a Burns collector.

Bill Verburg

Let me reiterate something that I have posted before
first understand the difference between the engines being discussed and alluded to.
Even fire engines
Porsche 911/964/993/996/997 in all of it's versions
Ferrari V8s, Cart V8s all V8s w/ 90* V angle and a flat single plane(180*) crank


Odd fire engines
US V8s all V8s w/ a 90* v angle and a 90* dual plane(X)crank

What's the difference?
Even fire always fires alternate banks sequentially(LRLRLR....) so that there is a maximum separation and flow of exhaust gas through a collector. In a flat 6 each collector sees an exhaust pulse only every 120* of crankshaft rotation, in 90* V8 w/ flat crank each collector sees a pulse of gas every 90* of crankshaft rotation.

Odd fire always fires consecutively(LRRLRLLR for a Corvette) into a single collector on one side then consecutively into the other collector. Some gas pulses in a given collector are 180* apart in a given collector BUT there are 2 that are only 90* apart on one side and 2 only 90* apart on the other side. These consecutive pulses into a single collector overload any collector that is sized for all the other pulses, so it either needs to be over sized for most of the time or other accommodations need to be made.

What are the other accommodations appropriate for an odd fire engine?
There are 2
NASCAR and the US hotrod worlds have popularized the crossover. It is very effective in that application. Placement, merge style and size are not critical

Ford on their Lemans GT40s used a different solution. They used a "bundle of snakes" exhaust aka 180* exhaust where one left cylinder goes to the right collector and one right cylinder goes to the left collector. The resulting configuration eliminates the consecutive fire into a single collector. Every pulse has a full 90* separation from the others. The collector can then be sized appropriately. This solution is occasionally seen on other V8s but is rarely implemented because of cost, packaging and manufacturing concerns.

second understand that acoustic tuning by varying the length & diameter of the primary & secondary pipes and the configuration of the collectors due to reflected signals in the pipes is largely negated by hooking them up to a muffler. The effectiveness of acousic exhaust tuning is also negated by increasingly short cam duration, timing & overlap periods.

third the effect of any merged exhaust whether on one side or in a x-over situation is to broaden the torque curve at the expense of peak hp

That said a 911 flat 6 doesn't receive the significant benefit from a x -over as implemented in a US V8. It may recieve a benificial result of enhanced midrange torque from a properly designed merge x- over. The specific implementation on the 6 is far more critical than on the V8 where the x-over can be placed just about any where in either an X or H configuration.

Lastly you will note that Porsche has never used a x-over on any race engine. Prefering proper pipe length, diameter, merge and megaphones(where allowed)

Mark in Baltimore

Thanks for all of your info, Bill.

RallyJon

Porsche racing, Ninemeister, Fabspeed and you all agree that for max top end power, split flow is best. Consider it settled, then.

But how many of these bypasses, sport cats and X pipes are sold for pure track cars, vs street cars where 10hp at 4000 rpm is a much greater benefit than 10hp at 6500 rpm? So, as you say:Seems like that would be important to about 90% of the end users. What IS a properly designed merge x- over for our engines?

eclou

some x-overs or balance tubes are done purely for tonal qualities and reducing exhaust resonance. Also, putting Fabspeed in the same sentence as Porsche racing is heresy