Step okay from exhaust port to header tube?

[900HP]

To take it a step further, here is the Chevrolet R07 *RACING* cylinder head. I don't see a "D" port or square port here

R07 exhaust port.jpg

[900HP]

Here is the Rousch/Yates D3 head................ notice anything familiar

Roush D3.jpg

[In-Tech]

yeah, all those heads suck Oh wait, the motors do

[900HP]

Here's the AJPE billet head exhaust port and while not perfectly round easy enough to match a round header tube to.

AJPE exhaust port.jpg

[900HP]

I'll do one more, the Ultra-Pro Machining billet 9º head, a work of art.........................

Ultra pro machining 9 exhaust port.jpg

[Sir Yun]

I'm not contesting the empirical knowledge of anyone. I'm not an expert in thermodynamics nor exhausts for that matter ( working on it ). ( end of disclaimer).

Not to sound all Yoda but : There is no head only pipe. The port in the head is not one Iota different than the header for the gas. If you make a large step up (or down) you have now made a stepped outflow tract, the step being wherever the head happened to run out of metal. I very much doubt this step was a design parameter for an after market head. If not, it is pretty much down to serendipity if it is good or bad for power.

It you paid $2500 for a custom header and you asked the ''guru'' who build it why the fancy steps are they way they are, I don't imagine you'd want to hear: I ran out of pipe/that was what was in my scrap bin/couldndbearsed/looks purrty like this..

The effect of the step-up is a like a open end of a pipe but a bit less effective in it's reflection. That means that the pressure ratios for both the compression and expansion waves will change as well. Same for the step down, like a less effective closed end.

btw for an non-isoentropic expansion ( step up), the temperature (and entropy) will go up.



V8 F1 exhaust, there is nothing here what is not intentional.

[900HP]

^^^excellent post^^^

I should add, look at the effort put forth to smoothly transition from the oval/oblong exhaust port to the round tube. The reason, of course, for the oddly shaped exhaust port is because it has 2 exhaust valves and 2 exhaust paths that blend to one and it's nearly impossible to make an effective round port with this configuration. Those headers and the transition from oval to round are pure art.

[900HP]

here are some exhaust port flanges that I matched to existing exhaust ports. The first is for a 409 Chev, the second fits a Dart 355 cnc exh port.

409 Chev headers 7.jpg

headerflange.jpg

[900HP]

The headers from photos #1 and #2 above....................

409 Chev headers 13.jpg

taylor1.jpg

[gordonr]

What I notice about all those pics is that ports are raised in comparison to the standard. Ports that aren't "raised" will have a D. Trying to get 160+cfm out of a non raised round port is very difficult unless you raise the floor.

[900HP]

What I notice about all those pics is that ports are raised in comparison to the standard. Ports that aren't "raised" will have a D. Trying to get 160+cfm out of a non raised round port is very difficult unless you raise the floor.

I was trying to demonstrate that it's important to have a good port match and in a lot of race-only applications there is a lot of effort that goes into that.

Sometimes we have to deal with the exhaust port the manufacture has given us and it's less than ideal. Putting a step right at the head creates turbulence and that is a very bad spot to have turbulence. Calvin Elston has said many times that he believes the most important part of a header is how the primary tube matches the head and the first 8" or so of the primary tube. If you go to Elston Headers and look at his list of customers you will have to agree he must be doing something right.

[groberts101]

What I notice about all those pics is that ports are raised in comparison to the standard. Ports that aren't "raised" will have a D. Trying to get 160+cfm out of a non raised round port is very difficult unless you raise the floor.

And that's surely no accident. You can't or shouldn't look at, say.. a Cleveland 4V head and think "wow, that's awesome.. those engineers really had it all figured out". When in reality they screwed the pooch big time on that one. Which is of course why the old Pro-stockers cut the hell out of them and plated them up to fix that crappy flow-bench howling bastard. They were so bad that even the old stingers made them sound better and changed the exhaust note to some audible degree. And also why that design characteristic has been carried forward in one iteration or another ever since. But.. ask a Pro-stocker to design a head that needs to be squashed down between the frame-rails of an early Mustang and you'll likely get a similarly poor result as there's simply not enough room to make a really nice high-exit port and still be able get it to bend down all at once in the manifold itself while avoiding the shock towers. The harsh reality is that the turn needs to start in the port just to make it all fit and still be servicable.

I believe that many posters here are right to one degree or another but things need to be taken within the context of the engine design and its intended usage parameters(temperature, velocity, and flow capacity). While others who make total blanket statements and only believe it's got to be entirely one way or the other.. maybe not quite so much. Life isn't like a box of chocolates.. it's more like a header design and compromise after compromise is going to be needed to get through it.

I myself will always shoot for optimum flow paths and having the port nearly perfectly matching the pipe is such a design. I shoot for about .020 mismatch to allow some margin for error during fit-up. If I was racing at the top level?.. I'd probably shoot for about .005 mismatch or just run it perfectly matched with locating dowels. Anything else is purely a compromise or sometimes even a band-aid to one degree or another. If you need a mis-match to create a reversion dam?.. then your cam and header design is wrong to begin with and you're apllying a band-aid. D-ports work well.. or even squashed ports.. on space limited applications that must be routed around these stupid things called.. frame-rails. Otherwise, a taller upswept and typically also longer port can get it all done and hit the primary as it should. In round form and relatively straight on fashion in continuance to the ports exit angle. Good luck doing that on much more than an open engine with zoomies. Beautiful as they surely are, even the F1 headers pictured above are constrained to some degree and could be made better if they didn't need to be mashed into a bundle of snakes and routed around and away from surrounding components. Always some level of compromise.

Personally speaking, I'd like to see more hydro-formed header designs being implemented. Then you would see some wilder "round isn't always best" designs. Round is great when things go straight.. but when we need to turn corners?.. it seems to have been proven long ago that it's often better to scrub some energy off in trade to make it go round a corner better to reduce the tendency for creating other more undesirable characteristics that rob even greater amounts of energy further downstream. Exhaust systems are a complete exercise in the art of compromise.

[900HP]

[user-23911]

I was trying to demonstrate that it's important to have a good port match and in a lot of race-only applications there is a lot of effort that goes into that.

The key words there are "race only application".

That's because you're raising the power band by making more torque at high RPMs and losing it at low RPMs.

So really the question /answer should have some reference to the application that it's going to be used for.

[900HP]

I was trying to demonstrate that it's important to have a good port match and in a lot of race-only applications there is a lot of effort that goes into that.

The key words there are "race only application".

That's because you're raising the power band by making more torque at high RPMs and losing it at low RPMs.

So really the question /answer should have some reference to the application that it's going to be used for.

True but I will contend that having a mis-match at the port that close to the valve is not an ideal situation. Can you make plenty of power that way? sure. Could you make more torque and have a broader power curve if the header was designed correctly? absolutely. Put your anti-reversion after the collector, that works really well.