I've seen some race cars use a step header design and am curious to how they help. I know that a smaller diameter tube helps low end torque by increasing backpressure. Or maybe its the valve events are timed wrong and the smaller tube helps acheive better efficiency. With a bigger pipe say 1 7/8 is it possible to over scavenge the exhaust gases? It seems by combining two different diameter pipes such as a step header design you can achieve the best of both worlds. Can anyone explain how this "tunning" works and if it can be applied to a street configuration?

Thanks

 

Let me be the first to jump on the word "backpressure".

The exhaust primaries make torque at a given RPM due to the
length and diameter of the tube which makes the tube resonant
at a certain RPM and also creates high gas velocities in a given
range of RPM.

THe backpressure term is a myth. Slowing down or impeding exhaust
flow does not improve torque.

EDIT - Correction about stepped header design

As the exhaust gas cools, it contracts. Since the tube has a lesser
volume down the length, it helps maintain or possibly increase
the gas velocity once again to create low pressure.

If the exhaust runners measure low pressures, it allows gas in
high pressure areas (such as the chamber, or intake runner) to
flow down.

The larger primary exhaust runners will become efficient at higher
engine RPM where the charge volume is greater. The chamber will
be scavenged of exhaust gas to allow fresh charge to enter.

The valve timing also matches the length of the exhaust system.
As pressure pulses bounce back, the valves are timed to recieve
the pulse to help pull in, or pull out (intake & exhaust respectively)
the charges.

You should do a search on the thread called, "The Myth: Backpressure vs. Resonant Tuning".

 

I thought the step design increases pipe diameter as it "steps up" therefore providing more volume?

So by using a smaller diameter pipe it increases the gas velocity and eliminates low pressure?

Agreed!

It just seems that most sbc use 1 3/4" header as the performance standard. Others claim a smaller 1 5/8" will produce more torque, but at the expense of high rpm power. I've also seen a larger header produce more power right across the board vs smaller.
My question is when should a step header be used? What application?

Thanks

 

I've seen some custom headers that taper as the length increases.

I don't know the effects, or principals behind a stepped design
that increases diameter as length increases. I'm sure someone
here will have the story on that.

From what I've seen, the stepped header doesn't improve
much for engine performance or power. That's not conclusive,
it's just something I've read.

I often wonder why pro-mod engines, or top fuel motors don't
use a stepped design if such a system worked well?

The exhaust port at the head, and flange of the header is probably
the most important area to consider as the exhaust gas is
hottest and moving at high velocities.

I have no idea why increasing volume down the system would
be of any benefit? Someone please jump in here and save me!

Increasing gas velocity actually lowers pressure. Fluids that move
quickly through a tube create low pressure zones. If you remember
Bernoulli's principal from Physic's class, you can relate the exhaust
system to his discoveries.

Have you studied carburetor design at all? The venturi is a tapered
tube
which increases air speed as the air travels through the tapered
section of the venturi.

The throat of the carburetor venturi is shaped like an hour glass.
At the point where the air speed is highest, the pressure is lowest.

This is the section where the boosters are situated so that the
fuel (which is sitting in a high pressure fuel bowl) is sucked out
of the bowl into the low pressure air stream flowing through
the venturi.

Relate the same concepts to the exhaust and you'll see that
low pressure in the exhaust system helps to suck intake charge
down the intake runner, into the chamber for the intake stroke
(during valve overlap period), and finally out to the exhaust
during the exhaust stroke (also during the valve overlap period).

 

DO they actually work?

There seems to be two different schools of thought on the subject. Some builders claiming they do and others saying that they're a waste of time.
I don't doubt that a pulse is reflected but I wander about the magnitude of the pulse... does it really help?
Or would you be better served by a same-size pipe with more average gas velocity?

I've wondered about this for a while now. Published dyno results by one manufacturer or another are likely always skewed to their advantage, so who do you believe?

Burns claims to have a parametric software that can put you on the right path design-wise. If I were going to build a step header I'd definitely spend time working out the lengths.

-Mindgame

 

Stepped headers can be made to work better than constant size headers, just as 4>2>1 headers can be made to work better than 4>1s, BUT the whole system has to work together.

I've seen Cup stepped headers, F1 stepped headers (in pics and engineering drawings), so I know much testing has been done to get them correct.

WHY they work isn't always obvious to me. I believe they cause reflections at each step for tuning, just like any header. Perhaps you can tune for different rpms with the steps, or perhaps different harmonics.

There is some software being used to design stepped headers, but the good stuff isn't very available and it it still "design, cut, try, redesign, recut, retry, etc.". In all these high-end applications, very specific results are desired, and small gains are important.

I would not be surprised to see Cup engines with different (stepped) headers for different engine rpm operating bands. A short track engine may need a 3500+ rpm torque/power band, while a 2 mile oval might need half that. Obviously "plate' tracks have a very narrow rpm band (300 rpm or so).

I could see "detuning" midrange torque on a short track Cup engine where traction (aka "forward bite") was a limiting factor and the TC wasn't working well. . This might be done with steps which favor higher rpm tuning.

If you can believe the "acoustic tachs" shown on F1 racing telecasts, they seem to have power bands as wide as 5000 rpm, but of course that's about 14000 to 19000. Could F1 engines have header sizes/steps specific to each track? It would not surprise me in the least. Not when Ferrari reportedly spends $150,000 per racing mile for a season!

As to generic stepped headers being better for a street/strip car, I'm skeptical that many folks ever could or would do enough dyno and track testing to either optomize the engine or the header.




Amen to that, MG!

 

I believe this is the general consesus. There might be some scenarios where power can be had, but who's to say there wasn't an handicap elsewhere in the combo. For most of us here building LT1/4 engines up to 500 hp 1 3/4" long tubes should be fine. For others with 383-400ci producing over 500 hp maybe a 1 7/8 header will be a benefit.
I quess more importance should be paid to the length of primaries, collector size, equall length etc. to maximize your combo. Also, some type of coating will help keep the heat in the pipes and move the exhaust gases faster.

 

Thanks for the perspective OS. Good points as always.

Star,

I'm on the same boat. The idea of a stepped header seems to be a good one when you're looking to wring every last ounce of power from an engine. In that, the design has to be in synch with everything else. A generic header, while perhaps working ok, isn't really exploring any advantage IMO.

Your time is better spent looking at primary lengths, collector designs, etc..

I wouldn't worry about equal lengths either. There are guys out there claiming that their headers are within 1/2" primary length to primary length. But that's like saying the header builder knew what you need ±1/2" in the first place. Besides, I think it's more important to have a design that has large smooth flowing radii than one that looks like a basket of snakes, just for the sake of "equal length".

My 2¢

-Mindgame