Notices

Full length tubular scavenge header for Porsche flat-6

Thread Tools
 
Search this Thread
 
Old 12-24-2016, 10:48 AM
  #1  
959fan
Advanced
Thread Starter
 
959fan's Avatar
 
Join Date: Apr 2012
Location: SF, OC, Taiwan
Posts: 62
Likes: 0
Received 2 Likes on 2 Posts
Default Full length tubular scavenge header for Porsche flat-6

My background is Honda and I deal mostly with inline 4 engines. The Honda K series engines is ultra efficient and we can generate tons of hp naturally aspirated. The big 2.7 liter race motor is now up to close to 500 whp. Pretty crazy. Anyway, when it comes to NA performance, getting the entire induction system to work together is very important. One key element of exhaust tuning is taking advantage of the exhaust pulses and letting it help extract/scavenge the spent gas out.

So there is the concept of cylinder pairing. Someone else wrote this already so I'll just copy and paste.

"Basically, if you want the cylinders to be equally scavenged, then you need an even pattern of pulses, as much as possible. Which means an even length of piping and number of turns (as much as possible), and pairing the evenly spaced cylinders, 1 & 4, 2 & 3. Pairing those leads to exhaust events every 360 degrees for both cylinders. Pairing it any other way leads to an uneven pattern of 180 degrees/540 degrees for the paired cylinders. Which probably doesn't really matter *that* much, but if you're bothering to sweat the details, you may as well sweat them correctly."

I was wondering if anyone ever produce a full length properly paired race header for the Porsche flat 6 familly? I have a 997.2 with 3.8 that I will develop full titanium exhaust as well as a race header for... we have in our possession a GMG race header setup. It is a simple 3-1, and there is no muffler it's just a race pipe center exit. I'll post some pix later. The way the header is design it causes a bit of stress on the 3-1 collector and it cracked. We were able to buy the header for a huge discount. Normally the exhaust kit is somewhere around 6k retail I think. We have the collector repaired/replaced and will be soon testing it on our 997.2. We'll tune it with COBB AccessTuner Pro and maximize the gains. Dyno results before after will be shared.
Old 12-24-2016, 04:56 PM
  #2  
959fan
Advanced
Thread Starter
 
959fan's Avatar
 
Join Date: Apr 2012
Location: SF, OC, Taiwan
Posts: 62
Likes: 0
Received 2 Likes on 2 Posts
Default

could anyone confirm or deny the Porsche firing order being 1-6-2-4-3-5? other interesting things i've seen from random forums:

"The Porsche engine is NOT a true boxer. In a true boxer engine the opposing cylinders are on the same cycle and (in theory) cancel out each others lateral force.

The Porsche engine is in fact a V6, but with a 180 degree V. Because of 4 stroke, 6 cylinder equation, 6 cylinder engines are inherently balanced, leading to a nice smooth feel and (more importantly) very high revving ability free from significant vibration."

"1.2 Gees wanted to know why its not strictly speaking a boxer engine:

If the 911 was a true boxer engine, the timing and hence firing order would be different, since each pair of cylinders would be on the same stroke...

Its kind of hard to explain without a diagram...

But if you look at the firing order of your 911 engine, then it starts with 1, followed 6, and while No: 1 is firing, No: 6 is on the compression stroke...

The Ferrari Boxer is no more a boxer than a 911 - its a V12 with a 180 degree "V."

"I believe a true boxer engine has all cylinders hitting TDC at the same time. Firing order doesn't matter (i.e. one can be on a compression stroke and another on exhaust). Someone feel free to correct me though.

In practice, the flat engine design offers very little advantages compared to others out there. They all have advantages and disadvantages that can be overcome to the same effect with design, materials, etc. As others have mentioned Porsche uses this design because that was what Mr. Porsche was used to. It has since become a legacy for them (it also suit aircooling - great for an aircraft since air is "free" and a reliable coolant source. you don't want to be at 15k ft and loose a water pump)

Porsche has proven very good at ensuring materials, manufacturing and design ensures a great engine. But, that doesn't make the basic engine layout great. Personally, I view the american v-8 as one of the best engines out there. For its displacement, it is extremely compact. It can produce tons of power in a reliable fashion and is extremely simple. Hi-Tech is cool, but there is genius in simplicity."

"I had to bring this back...great info!!!


But it is incorrect, firing order is not relevent.

The 911 has a boxer engine largely because Hitler was driven on campaign thousand of miles in a Tatra T11. This 1921 Hans Ledwinka design had an force fan, air cooled, boxer engine with a backbone chassis and independent suspension. Hitler dictated these design specs to Porsche. Porsche knew he was stealing Ledwinka's work because he inherited Ledwinka's desk at Steyr and testified that Hitler assured him that Hitler would take care of the patent issues. VW settled out of court after Ledwinka died."

"One of the big advantages of the 911 engine layout is the simplicity of design. There are ways of doing lots of the things in the 911 motor better but there is a BUNCH to be said about refining a design over decades.

A couple examples of how a design attributes could be improved:
- "Low center of gravity". Yes the flat six has a low center of gravity compared to a straight 6 but if you really wanted to go nuts with a low center of gravity you would change it to a 120° V6. That would allow you to put the crankcase closer to the ground but leave enough room at the heads for the headers.

- Air/oil cooling of the heads makes for a simple, tough and reliable engine... But the lack of concentrated heat exchange (like you get with water cooling) limits the ultimate power output along with reliability of the heads/valves in turbo motors and keeps the air cooled motor from getting 4 valves per cylinder.


One of my favorite engine related quotes is from Smokey Yunick: "An engine doesn't know what name is on the valve covers". The Porsche engine isn't magical but it is the rare case of engineering in a road car heavily tilted towards true racing needs and then refined for decades."

"- Air/oil cooling of the heads makes for a simple, tough and reliable engine... But the lack of concentrated heat exchange (like you get with water cooling) limits the ultimate power output along with reliability of the heads/valves in turbo motors and keeps the air cooled motor from getting 4 valves per cylinder.


Water and four valves are better on paper, but only if you make use of their benefits. There is an interesting comparison of the air cooled, two valve 917 with the water cooled, 4 valve Ferrari 512S, in one of Colin Campbell's books. The Ferrari made 8% less power and the 917 won 24 out of 31 races. The Ferrari won once. For a sport car racing engine, 2 valves and air cooling were "good enough". I suspect things would have been different for GP tuned engines.

The important point about the advantages of the flat engine or air cooling is that neither were chosen for their "benefits". The choice was not made by Porsche and it was not made for technical reasons."

Last edited by 959fan; 12-24-2016 at 05:12 PM.
Old 12-30-2016, 12:46 PM
  #3  
959fan
Advanced
Thread Starter
 
959fan's Avatar
 
Join Date: Apr 2012
Location: SF, OC, Taiwan
Posts: 62
Likes: 0
Received 2 Likes on 2 Posts
Default



O_O Magnaflow factory is freaking HUGE!! Wow. Super good tech info, no fluff no BS. Awesome. Just from this you know Magnaflow products are properly engineered. Good stuff.
Old 02-14-2017, 07:07 PM
  #4  
V2Rocket
Rainman
Rennlist Member
 
V2Rocket's Avatar
 
Join Date: Jan 2007
Location: Nashville, TN
Posts: 45,497
Received 632 Likes on 490 Posts
Default

you've got a few things wrong in your essay above.
the 911 engine is technically a "flat 6" or "boxer" engine.
it doesn't count as a V6 like that Ferrari engine because the crank pins for the conrods are not shared by adjacent cylinders - 911 has 6 cylinders and 6 crank pins. if it were a 180 degree v6 there'd only be 3 pins.

boxer engine is just a nickname for "horizontally opposed" engine used by Subaru and borrowed by Porsche for the Boxster name. the "cancelled lateral movement" is because the motion of each piston/rod combination is always cancelled out by the opposite side moving the opposite direction.

in a subaru or Porsche flat-4 when a cylinder on the left is at TDC the corresponding cylinder on the right is at BDC and every point in between is perfectly opposed (unlike in an inline 4 due to crank/conrod geometry). adding 2 more cylinders just makes more piston/rods in opposition to another set.

as far as air vs water cooling, the beetle/356/911 were disadvantaged by their key design feature - rear engine placement.
hard to get enough airflow to reasonably cool the engine back there. if it were a front-mounted air cooled engine you could really turn the thing up.

however the switch over to water cooling was largely a result of emissions regulations and car buyers seeking "refinement"...air cooled engines make a lot of noise, from the fan whirring to the valves ticking to the cylinder barrels "ringing"...and the very high operating temperature possible from a high-output air cooled engine leads to very high NOx production out the exhaust which is a key component of SMOG.

switching to water cooling adds more metal and water (mass) to absorb any noises the typical driver might find offensive as well as stabilizes the combustion process for optimum EMISSIONS efficiency... a water cooled engine gets in trouble at 250 degrees, but an air cooled engine can keep humming happily.
Old 02-16-2017, 05:19 AM
  #5  
959fan
Advanced
Thread Starter
 
959fan's Avatar
 
Join Date: Apr 2012
Location: SF, OC, Taiwan
Posts: 62
Likes: 0
Received 2 Likes on 2 Posts
Default

well, i really didn't write anything above, it was all just quotes from other forum postings. but primary topic is full length headers... but it seems like nobody here is even interested in a discussion. oh well.

ps, good info on water vs. air cooled.
Old 03-17-2017, 11:12 AM
  #6  
Bill Verburg
Addict
Rennlist Member

 
Bill Verburg's Avatar
 
Join Date: May 2001
Posts: 12,246
Received 502 Likes on 345 Posts
Default

Originally Posted by 959fan
well, i really didn't write anything above, it was all just quotes from other forum postings. but primary topic is full length headers... but it seems like nobody here is even interested in a discussion. oh well.

ps, good info on water vs. air cooled.
You need to do a lot more research, here's a start
There is a fundamental difference between a Porsche flat H6(doesn't matter if it's air or water cooled) and a Honda I4

The H6 is even fire, this means that all firing is from alternate banks. Here each firing is 120° apart(this is determined by the crank layout where the rod jornals are in 3 planes 120° apart) and each firing into the same header collector is 240° apart. The result is well spaced exhaust gas pulses so no special exhaust plumbing is needed to enhance flow. Most exhaust designers feel this 240° spacing to be ideal for flow and acoustic tuning opportunities. Acoustic tuning relies on the geometry of the exhaust and dynamics of the gas flow to place a low pressure pulse at the exhaust port while it is open, this tuning is further enhanced if the there is valve timing overlap where both intake and exhaust valves are open at the same time, most modern street motors have little of this overlap, the exception being motors w/ varible cam timing.


The I4 is odd fire, this means that at least 2 cylinders on the same bank fire consecutively. Here it's the worst case where all 4 cylinders on the same bank fire consecutively. Here the cylinders fire 180° apart(again determined by the crossplane lane out where the 2 crank planes are 90° apart) and that is the same firing interval that the collector sees, the result is a more jumbled mass of gasses in the collector, w/ a simple 4 into 1 collector acoustic tuning can still work, just not as well as w/ a more spaced firing order, the result is a very narrow power band. Too alleviate this most I4s use a tri-Y setup which pairs the most widely spaced firings, here cyl 1 & 4 are paired as is cyl 2 & 3, This provides a sub collector where gas pulses are 360° apart, then the main collector where the 2 amalgamated flows are 180° apart. The effect is to smear out the low pressure acoustic signal at the exhaust port over a wider range of rpm, result is a broader torque spread w/ a bit lower max hp. Most find this to be a good trade off.
Old 03-17-2017, 01:58 PM
  #7  
959fan
Advanced
Thread Starter
 
959fan's Avatar
 
Join Date: Apr 2012
Location: SF, OC, Taiwan
Posts: 62
Likes: 0
Received 2 Likes on 2 Posts
Default

thanks for the reply!

is there a general consensus of the best NA header made for Porsche then? even firing it is but pairing different cylinders might produce slightly different results still. just wondering if anyone's seen any full length header testing on flat 6's??
Old 03-20-2017, 08:33 PM
  #8  
Bill Verburg
Addict
Rennlist Member

 
Bill Verburg's Avatar
 
Join Date: May 2001
Posts: 12,246
Received 502 Likes on 345 Posts
Default

Originally Posted by 959fan
thanks for the reply!

is there a general consensus of the best NA header made for Porsche then? even firing it is but pairing different cylinders might produce slightly different results still. just wondering if anyone's seen any full length header testing on flat 6's??
more or less standard for a 911
31" header w/ merge collectors, +/- this spec
pipe ID varies according to displacement and rpm from ~35mm for 2 liter & ~7k limit
to ~44mm for big bore high 8.5k rpm motors

Reverse cone stingers again varying in length and diameter and cone geometry depending on engine displacement and rpm are an asset, but of course not generally streetable

The thing is as soon as a muffler is added it will negate much of the benefits of any available acoustic tuning
Old 03-23-2017, 11:43 AM
  #9  
JasonAndreas
Technical Guru
Rennlist Member

 
JasonAndreas's Avatar
 
Join Date: May 2002
Location: USVI
Posts: 8,138
Received 112 Likes on 90 Posts
Default

Originally Posted by 959fan
is there a general consensus of the best NA header made for Porsche then?
If you want to see what the possibilities are for your 997.2, just look at the three different solutions that Porsche Motorsports used on the GT3-CUP, GT3-R and GT3-RSR.



Quick Reply: Full length tubular scavenge header for Porsche flat-6



All times are GMT -3. The time now is 10:27 PM.