Subaru exhaust

[RJES]

Sorry my typing in the post above is crap as usual!

[HarryMann]

True for peak power, witness US dragsters and the such-like, and also for blown engines. The cylinder configuration and firing angles obviously makes the unequal length tuning cases different from engine to engine...

and being able to change the length of pipes, or their volume, as revs change would be very useful. This is done with inlet systems (which can have an equally powerful good or bad effect on pressure wave tuning) and a relatively new idea on 2-strokes is to change the effective length of sections of the expansion chambers and tail pipes, using water injection to alter the speed of sound in the exhaust gases - as good as the same as a physical length change - now that is a bit trick!

[klatuu]

Well this has been an interesting discussion, My exhaust solution was to get the van on the road, (I was sick of using a 13 yr old leaky cavalier that ran like a bag of hammers).

I figured that Subaru could get the length of the headers correct, and everything after the join is a single pipe, with a couple of boxes between that and athmosphere, so thats what I did.

I like it, its quiet enough, the van goes well, no stutters throughout the rev range.

I'm sure that smallcar and KEP and other manufacturers have done the required research and as Diamond says, many happy customers, but I prefer something abit more even.

[HarryMann]

Even, as in even length pipes?

As Richard says, for flat fours, this usually is the solution if:-

1) You want max power and not max torque (the torque curve is shifted up the range, but maybe its max value not as high as possible with other configurations)

2) The length required suits the vehicle layout, shorter equals higher rpm, whether to the manifold pipe joins or the total length.

The camshaft timing, and any overlap, as well as inlet length and volume work directly with any exhaust tuning, so together with the engine's cylinders and layout will determine largely what scope there is for imrpoving the exhaust one way or another.
It's almost a given though, that gains at one engine speed result in losses at another, and optimising the gains for what you want is the name of the game..

One way or another, this is what you are trying to achive:-

Create a tuned system through the inlet tract, cylinder and exhaust so that:

a) there is a high differential pressure helping the fresh charge enter and fill the cylinder - pressure fluctuations can exceed atmospehric pressure and cylinder filling can exceed 100% (volumetric efficiency) in normally aspirated engines;

b) for economy and good mid-range torque - as little unburnt charge ends up in the exhaust as possible, particularly for engines with lots of overlap;

c) for maximum power - as long as the cylinder is well charged at high rpm and a big hole in the torque curve doesn't result, b) is less important;

All achieved by arranging for positive pressure waves to arrive at the exhaust port as the valve just closes and negative ones as the valve opens and for a suitable duration thereafter.
Due to the speed of sound being nominally a constant in the hot spent gas, this only occurs strongly over one rpm band, but can be reinforced to some extent at a second with clever design.

Unequal length pipes and/or connections between manifold pipes (balancers) are often used to bring the rpm range down where this happens, using one cylinder's pressure waves to reinforce or ameliorate those of another... an lcb (long centre branch) manifold (A series engine with bifurcated centre exh ports) or 4>2>1 manifolds (other in-line 4's) requiring a lot of pipe bending are usually the design of choice for a wide spread of torque from the mid-range onwards and would make that much difference that re-jetting upwards was required, an indication that something was working quite a bit better.

[starspider]

Harry I got a headache reading all that

[Horza]

Doesn't really tell you how to do it though unless you are already in possesion of a degree in automotive engineering and have access to some time on a supercomputer to run simulations.

Klatuu's and TC's solutions are ideal for your home converter, relying on Subaru having it right or the established reputation of an aftermarket supplier respectively.

Sucking it and seeing would seem to be not ideal unless you have plenty of time, money and energy to keep rehashing it until you get the desired result.

[scoobyvan]

i think you will find also that the different length pipes into the equaliser are also different diameter pipes.if subaru had kept the pipes equal length and diameter they would have had to move the gearbox,so they just went for the lazy solution and sorted the exaust

[HarryMann]

Harry I got a headache reading all that

Crikey, its just explaining what is usually the name of the game, roughly what is required to gain any tuning benefit, so if it had been aimed at telling you how that might be achieved, I suppose I'd be responsible for serious illness or death...

Having established one of the two or three ways of getting a positive benefit, someone has to do the calcs, make a prototype and see if there are any downsides to it - that's really what the testing phase is for. A bit easy to get drop-outs in the torque curve or upset clean carburation...

[mr_bricolage]

Nah, you don't need a supercomputer to do it. I wrote a program in FORTRAN back in my university days which worked it all the pipe lengths and diameters out for you . We used it for air conditioning ducts but then I also changed it so it worked with exhausts as well. You needed to understand some basic engineering maths (gaussian integration) to get it working. I changed it about a year later and it ran on an Excel spreadsheet!

[HarryMann]

Great! So lets have it up here then... can put it on Syncro website or Wikipedia...

[mr_bricolage]

Hmm that spreadsheet was a while ago. I still have the project in text so creating the appropriate programming Excel should not be a problem. I'll post it up when I've got it done.

[HarryMann]

That'd be great... if you need any help with VBA behind Excel, give us a shout.

[Hacksawbob]

VBA behind Excel

your a bit of a dark horse script kiddy Clive!

[HarryMann]

earns a penny here a penny there...
But its a full blooded language rather than simple script behind there!

[starspider]

Been getting the engine ready today putting new timing belt on etc and in a moment of madness put my headers on the wrong way round and look at the result







now I reckon if I do a bit of surgery where the tape is I might end up with the beginnings of an exhaust system what do you think