Header Tube and Collector Specs "Trends"

[ptuomov]

So, what’s the answer?

just need to start doing some testing to find out for sure if it will help you .

Fortunately or unfortunately, my car has very little room for exhaust headers or manifolds. So practical testing is not really possible, and what little has been done has focused on reducing the high-rpm 90-degree exhaust blowdown interference that hurts two cylinders.

I’ve tried some experiments with Vannik’s software, but no clear trends have emerged.

[frnkeore]

How does merge collectors effect tuning the wave and velocity and can PipeMax be configured to do merge collectors?

[maxracesoftware]

How does merge collectors effect tuning the wave and velocity and can PipeMax be configured to do merge collectors?

i noticed your other Post :

by frnkeore » Sun Sep 06, 2020 1:15 pm
I have Pro 4.50. Is there any way to add merge collectors to the exhaust system
and how does that effect the collector tuning?

current 4.50 version , there's no way to model Merge Collectors with rest of full exhaust system
that Combo is disabled
most probably be enabled in PipeMax v5.0

[frnkeore]

When will 5.0 be available?

[maxracesoftware]

When will 5.0 be available?

i don't exactly know yet
half of PipeMax v5.0 is finished , some parts are same as a new Dragstrip Simulation i'm working on ,
so around a month after that Dragstrip Simulation Program is finshed ??

[Erland Cox]

the EGT is the System's Speed ( Speed of Sound FPS ) you have to work with,
and the "Timing" of the returning negative pressure Wave @ TDC-Overlap is where huge amount of TQ and HP gains are ,
so the earlier EVO event is ... the longer the Primary Tubes need to be .

a "Hunting analogy" :
the faster the Duck flies ... the more you lead the shot with your Shot Gun

so basically ...
if you shoot too soon , you don't hit the Duck ( TDC-Overlap )
if you shoot too late, you don't hit the Duck either !
if you shoot at the correct time , you hit the Duck , and win the TQ and HP prize !
 

Makes sense that if you can measure the exhaust gas temperature for the given camshaft and if the exact header design doesn’t really chance that EGT much, the return pulse from the collector or even open single pipe can be timed that way.

What do you do with turbocharged engines? Those see a constraint at the end of the pipe (the turbine) and send a different kind of pulse back for the overlap. Not the good kind!

The factory solution swells to be primaries about half the usual length for the total exhaust tract and minimum (apparently ideally zero) collector length with primaries pairing right at the turbine entry. But I’ve got no science behind that, just observing some anecdotes.

here's what i've seen on my Dyno tests,
especially on 700 to 800+ cid Engines with 4-1 Headers ( thats a whole bunch of Thrust , because these were heavy Engines )
and even on Craig's A/ND and B/ND SBC 406 to 410cid engines with Zoomies
If you make the Collector length at the worst Harmonic length ... the exhaust thrust will push the Engine forward
during a Dyno test , so that i have to abort the Run, and "Chock" the wheels on the Dyno Engine Stand
to prevent the Engine from moving towards me .

Same with Craig,
if we put on 4-1 Headers in place of Zoomies , and the Collector happens to be at worst Harmonic length, it will move his Engine forward .
this Test i did was new 3-Step Dragster upswept over rear tires ... so the Collector was pointing almost straight back
so we popped-off those Collectors and ran the Primaries like they were Zoomies
and they happened to be at the best Harmonic length ... so that they made very close to his best Zoomies
and did not move the Engine stand forward anymore

So my Theory is :
for Turbo app with a lot of Boost Psi , you would want the "Worst Harmonic length" to be at the Turbo
so you get the highest Boost Psi
.... just my Theory , never tested it yet

Until the Turbo Boost Psi relatively surpasses Atmospheric pressure
it should like the Length to be at around the Best Harmonic lengths

There is a big difference between an N/A header and a turbo header.
On the N/A engine the collector ends at the atmosphere or in a muffler.
Both give a negative wave reflection.
On a turbo engine the collector ends in the turbine which will give a positive reflection.
By using shorter pipes the positive reflection arrives at the exhaust valve before overlap and will reflect back and drive the turbine again.

Erland

[ptuomov]

Erland —

I believe this is what happens with a single cylinder turbo engine.

The high pressure wave from the blowdown gets reflected back from the turbine as a high pressure wave. This wave then makes it to the open exhaust valve (assuming typical turbo pipe lengths) as the “first coming”. The first coming reflects out a low pressure wave, which does nothing good at the turbine, after which it bounces back again as a low pressure wave towards the exhaust valve. Call this the “second coming”.

If the valve is open, the second coming will send back another high pressure wave. If the valve is closed, the second coming low pressure pulse is reflected as a low pressure pulse.

My understanding is that the most important thing in a single cylinder turbo exhaust is that the first coming happens with the exhaust valve at high lifts so the wave sign is maximally reversed and that the second coming happens during the overlap to evacuate the combustion chamber and get the intake flowing. This is not a fact, just my theory.

My theory is that most of the waves driving the turbine happens with the first time the blowdown hits the turbine and the subsequent bounces are not that important or even beneficial to the turbine. The subsequent bounces are however very important to the valve overlap period and gas exchange.

All this turns out to be less important in a cross-plane V8 where the blowdown interference between cylinders, and how to mitigate it, is the most important thing. There the death is one cylinder’s blowdown making it to another cylinder during its valve overlap and that is a potentially worse disaster scenario than anything that one could create in a single cylinder turbo engine.

[Erland Cox]

A wave changes sign against an increase in size but reflects the same against a smaller size.
So both the turbine and the exhaust valve will reflect a positive wave back without changing sign.

Erland

[ptuomov]

A wave changes sign against an increase in size but reflects the same against a smaller size.
So both the turbine and the exhaust valve will reflect a positive wave back without changing sign.

Erland

I don’t believe that is the case if the exhaust valve is open. If the exhaust valve is open about 25% of the valve diameter, the cross section does not change because of the valve but it massively increases because the cylinder is large compared to port. Therefore, if the first reflection hits the exhaust port when the exhaust valve is reasonably close to the usual max lift, the wave does change sign. If the exhaust valve is closed, it doesn’t change sign. I’m fairly certain if this, but then again I’ve been fairly certain and wrong in the past...

[Erland Cox]

The area in the exhaust port is usually smaller than the area in the header.
And a valve in the middle if the port blocking the port and forcing the wave to turn around it in all directions will make
the wave see the the valve as a restriction and reflect with the same sign.
See it like this, if you flow the header backwards without the the head it will flow more than if the head is mounted with the exhaust at full lift.

Erland

[ptuomov]

The area in the exhaust port is usually smaller than the area in the header.
And a valve in the middle if the port blocking the port and forcing the wave to turn around it in all directions will make
the wave see the the valve as a restriction and reflect with the same sign.
See it like this, if you flow the header backwards without the the head it will flow more than if the head is mounted with the exhaust at full lift.

Erland

I basically disagree with you.

Yes, the exhaust port may be smaller than the header primary in many cases, but especially in a single cylinder turbo engine I see no reason why it should be. For example, in my car, I have two 34mm exhaust valves and 40mm head exhaust port matched to a 40mm cast exhaust manifold. Working it out with throat sizes, stem diameters, etc, the cross sectional area with the valve open is about constant.

Contrast this with the 100mm bore and about 500cc cylinder and combustion chamber volume at the relevant crank angle degree. That’s a big increase in the cross-sectional area and a geometry that works pretty close to an ideal helmholtz resonator. I think that the open exhaust valve is going to send out, at least initially, the opposite wave.

I’m more interested in pressures and less interested in the flows that these pressures cause at this point. I think the wave is going to mostly just propagate around the open poppet valve.

Now, there are people more experienced with modeling these things and they might be willing to chime in and tell us who is right.

[Erland Cox]

Disagreeing makes an discussion that both will learn from so that is good.
Vannik should be able to answer or I will have to find the tests in Blairs book.

Erland

[vannik]

You are both partially right.

If the valve stem area forms a restriction there will be a reflection of the same sign. But it will only be a partial reflection. Its size will depend on the size of the restriction which is a function of both the cross sectional area and the flow coefficient past the valve stem.

The valve head area will reflect the same sign when it is closed and inverse sign when opened past the point where the effective flow area is the same as the cross sectional area. In between it reflects a part of each so halfway the same sign reflected part cancels the opposite sign part and nothing gets reflected.

[Erland Cox]

From what I have seen on dyno testing with carbs and tri-Y headers I get a flat spot around 3500 RPM.
Around this RPM the lambda reads rich from double carburating.
From my understanding this is because of the second reflection from the first junction in the tri-Y header.
A positive pressure wave goes out and reflects at the junction as a rarefaction wave.
Back at the cylinder head it reflects back as a rarefaction wave and changes sign at the first junction to a compression wave.
This compression wave goes up through the intake port during overlap and causes double carburetion.
One of the reasons of tri-Y headers should be that the negative wave from the second junction arrives together with
the positive second wave from the first junction at the valve and cancel each other out.
I can not measure pressure, I can only see what happens at the carburetor.

Erland

[ptuomov]

If the valve stem area forms a restriction there will be a reflection of the same sign. But it will only be a partial reflection. Its size will depend on the size of the restriction which is a function of both the cross sectional area and the flow coefficient past the valve stem.

In my car, the cross-sectional area of the exhaust port is almost exactly constant, as cast by the factory, when you look at it normal to my assumed flow direction. Furthermore, the exhaust port is pretty steep and the angle between the valve stem and the port centerline is pretty shallow. It’s a Porsche 4V head. Although intuition is an unreliable tool for guesstimating valve reflections, my intuition says that it’s not going to reflect back much more than a pipe bend with the valve more than 0.2D open (with its throat percentage and seat geometry).