New/Modified Header Design?

[TR]

I see the similarity between a venturi effect and what is happening in a header, but the two are not related.

Venturi effect - An increase in the speed of a flow causes a decrease in pressure. Add a small hole at this location and the pressure differential pulls fuel, sand, etc. into the primary flow path.

Exhaust Header - The exhaust is slowing down as it travels through the exhaust, so it is actually the opposite of a venturi...Another way to think of it - Cut one of the primaries off of the header. The exhaust from the remaining cylinders comes out the cut primary, it doesn't suck air in the cut primary and continue down the exhaust system...

So, after a quick look...Here is a link that explains header function and makes no mention of any "theories" on how a header can violate any number of the laws of physics to make HP:

http://auto.howstuffworks.com/question172.htm

Keep the ideas coming, my day job doesn't make me think, this is fun! TR

[dbrick]

I agree, this whole forum makes me think, which is the fun part.

Also remember, according to aerodynamics and physics, a bumble bee can't fly. Luckily the bumble bee doesn't know this.

[Daryl Smith]

Actually, I think the venturi effect is very relevant.

Consider the crankcase scavenging systems (mostly used by racers as it would probably adversely effect emissions, but I've seen a few street roadsters use it) which are tied into the exhaust systems (usually at the collector). They have a one way valve so at lower RPM's the exhaust doesn't go back into the crankcase, but at medium through high rpm's the blowby is sucked into the exhaust flow and out the tailpipe.

If it has this effect on the added pipe then surely each exhaust pipe must affect the other three in the same manner. In a sealed system. Bets are off though once past the collector. (that is where ALL the action is.)

Yes I am thinking too.....Without a physics background I have to find relevant examples. Fortunately you are pointing the way as we go.
Daryl

[Mike Unger]

I assure you scavenging in VERY real when it comes to normally aspirated engine design. Why do you think camshafts have overlap? This is to increase flow into the cylinder on intake. The exiting exhaust gas helps to create low pressure to increase mixture flow into the cylinder. This is taken to the extreme in race engines and is one of the reasons they make more power. Likewise, header design becomes more critical with high overlap, long duration cams. This is why header design becomes specific when maximizing performance for given engine configuration. For street cars with moderate cams it is probably less a factor than race cars. Probably the best thing you could do for a street car is to match the intakes and mismatch the exhaust (header bigger than the port) to discourage reversion. Forced induction engines do best with as little back pressure as possible so you see very abreviated exhausts on supercharged drag cars. They have other things helping to get the mixture in. Turbo race cars still put a lot of effort into header design to maximize flow and discourage reversion.

[TR]

I think intake and exhaust valves have overlap because the optimal setting would be somewhere around intake 100% open during the entire intake pull and completely closed the rest of the time. Likewise, the exhaust valve 100% open once the charge is spent/piston reaches BDC till the intake starts when it switches closed instantly. Since the valves cannot be turned on and off like a light without impossibly steep valve open and close profiles, they tend to overlap.

If the exiting gas caused a low pressure, it would stay there. What is its motivation to leave? Ignoring that, if the exiting gas was creating this huge vacuum, it would suck the fresh charge right out of the chamber, not into it. The shortest path is valve to valve, if the vac was there, the charge will follow. There is no way to get decent fuel mileage if most of your gas is going unburned down the exhaust. On a race setup, longer durations are used in attempt to completely exhaust spent charge and draw in as much raw charge as possible without concern of polluting or wasting gas...

I do agree that "header design to maximize flow and discourage reversion" is important, but I am not sure that the port to header tube sizing plays a big role...

dbrick - I think I have heard, "Given aero-engineering, an engineer wouldn't design a bee as they now exist", but I totally agree with you - there are many very cool things in nature that we do not have the chance of simulating. The human spine, spider silk, etc...

[dbrick]

Overlap also has the important function of cooling the valves, especially the exhaust.
One thing people overlook is that a moving column of air has mass and therefore, inertia. This is one reason for the long intake runners on injected engines, tall tunnel ram intakes, or the wild example the Chrysler 413 cross ram. The incoming air charge "piles up" behind the intake valve, creates slight pressure and enters the chamber faster. Rpm range depends on length. I would guess the exhaust pulse should act in a similar manner.

For me, the only thing that really matters is the little burble sounds, almost a backfire, but not quite, when you shift or engine brake. I always felt half the personality of a car is the exhaust tone. If you have it too quiet, then the car isn't having any fun. The classic Italian Ripping Canvas exhaust tone from the 50's is great, hard to duplicate.

I put header/exhaust design into the category of black arts, sometimes the math don't matter.

[TR]

One vehicle that had a really cool sound was the old Honda inline 6 motorcycle. Six tiny cylinders with hardly any exhaust system wide open really sounded cool. I have also been a long time fan of two-strokes...

As for flow, the intake is being pulled in and keeps having a door slammed in its face, then opened...

The exhaust has the same momentum leaving, but it is magnatudes smaller than the expansion that is occuring. The exhaust wants to go anywhere, just give it a place. Including right back into the cylinder if it is allowed to...Again, the two flows heading toward each other in the same pipe...

[Mike Unger]

I think intake and exhaust valves have overlap because the optimal setting would be somewhere around intake 100% open during the entire intake pull and completely closed the rest of the time. Likewise, the exhaust valve 100% open once the charge is spent/piston reaches BDC till the intake starts when it switches closed instantly. Since the valves cannot be turned on and off like a light without impossibly steep valve open and close profiles, they tend to overlap.

I wish this was correct. It would make camshaft design much easier. All performance camshafts would be essentially the same. The only limiting factor would be getting the valves open or closed as quickly as possible around the optimum event. In reality, things like changing the centerline with the same valve duration events can alter performance greatly.

If the exiting gas caused a low pressure, it would stay there.

Only if it has no mass! The gases have mass and velocity therefore they have inertia. That is why velocity stacks, scavenging and headers all contribute to engine performance.

What is its motivation to leave? Ignoring that, if the exiting gas was creating this huge vacuum, it would suck the fresh charge right out of the chamber, not into it.

Inertia. It does to some small degree, cause intake to follow. It is worth it to get more pure new mixture in the cylinder. Those intake gases also have interia. Gases want to travel the straightest path possible. That is why a long duration camshaft doesn't work very well at low rpm. As the rpms rise the gas velocity rises and you get less mixture out the exhaust valve.


The shortest path is valve to valve, if the vac was there, the charge will follow. There is no way to get decent fuel mileage if most of your gas is going unburned down the exhaust. On a race setup, longer durations are used in attempt to completely exhaust spent charge and draw in as much raw charge as possible without concern of polluting or wasting gas...

I do agree that "header design to maximize flow and discourage reversion" is important, but I am not sure that the port to header tube sizing plays a big role...

Well, your wrong. Oldest trick in the book, proven on the dyno.

dbrick - I think I have heard, "Given aero-engineering, an engineer wouldn't design a bee as they now exist", but I totally agree with you - there are many very cool things in nature that we do not have the chance of simulating. The human spine, spider silk, etc...

[TR]

A bit difficult to follow with your comments inserted into what I wrote, but I'll try to comment...

I know the flow has mass, velocity and therefore inertia. What cannot be discounted is that for every action, there is an equal and opposite reaction. If this flow is causing a vacuum, it is pulled toward the cylinder just as much as it pulls charge from the cylinder. Also, the exhaust gas wants to expand with far greater potential than the inertia of that tiny mass...

Thanks for stating that I am wrong when I wrote, "I am not sure that the port to header tube sizing plays a big role". Maybe it does, maybe it doesn't, I am not sure. Sounds like you have Dyno results where that is the only factor changed between the two runs?

[Dave]

Wow Travis, for someone running a LOG STYLE exhaust manifold, you sure are into this discussion!

Sorry, I couldn't resist!

[TR]

Exactly! You gotta know just from looking at a log manifold that it creates more back pressure than an equal length, etc. etc. And violates everything discussed previously in this thread...But people also claim that the pressure gets to the turbo, log or otherwise...

People will say anything to sell their product...

[Dave]

Well, in reality it's heat and not pressure that spins the turbine wheel. As long as the heat gets there, what the heck. You are introducing so much backpressure by stuffing a turbo in the middle of the exhaust flow, who cares!

I've seen a number of log vs equal length dyno comparisons. The equal length manifold usually wins out by a bit. Not as much as you would think. On the other hand, the equal length manifolds all seem to crack out and fail where the log manifolds are pretty much bulletproof. I'll take a 10 Hp loss if it means I don't have to replace my manifold everytime I change my oil!

[TR]

Yep, that is what I have heard before in the log vs equal...

But it is the flow that turns a turbine, not the heat. Heat makes things hot, the flow of the exhaust makes things move.

Moving air with a fan produces heating of the air, but not vice-versa...TR

[dbrick]

This is way more fun than work. Warning, run on sentences ahead!! I'm asking questions, not stating facts.

Wouldn't the heat cause expansion which causes pressure which, when given a place to go, such as an exhaust port, causes flow which by having mass and therefore inertia, can spin a turbo impeller? If there was no inertia or "timing" involved, wouldn't every set of headers with the same diameter tubes work the same, regardless of the length of the tubes?

Also, since the column of fast moving pressurised exhaust will try to go straight rather than make a 180 degree turn in the collector to go up another primary tube which also has or just had a moving column of exhaust go down it towards the collector, it continues toward the tailpipe. Passing the other three primarys on it's way out at the collector, this pulse could then create a small, pressure drop in the three other pipes, but I doubt it would go backwards.
When viewed as a single isolated event, one exhaust pulse into a header would likley come out the other primarys, but engines aren't static.

I was running my daughters car with the #2 spark plug removed for a short time at idle to diagnose a problem(dead cylinder, turned out to be a bad injector) There were no loud exhaust type noise from the plug hole, just air being pushed out. she has a NX1600 with factory iron manifolds, maybe 10 inch long primary tubes

I always wondered why some turbos have exhaust manifolds which don't look to be efficient (log). I guess there is enough power gained by supercharging the engine that the exhaust flow wouldn't make a big difference.
Compressing air causes heat, moving air with a fan causes wind.

Anyway just my thoughts.

[jbush]

TR - I agree the Honda CBX (inline six) sounds great. In the early 80s some of the Japanese bike companies made big 2 strokes for street bikes. I have only ever heard one - a Kawasaki H2 (750 cc 3 cylinder). If you ever have a chance check one out.