Pelican Parts
Parts Catalog Accessories Catalog How To Articles Tech Forums
Call Pelican Parts at 888-280-7799
Shopping Cart Cart | Project List | Order Status | Help



Go Back   Pelican Parts Forums > Porsche Forums > Porsche 911 Technical Forum


Reply
 
LinkBack Thread Tools Rate Thread
Author
Thread Post New Thread    Reply
Moderator
 
Bill Verburg's Avatar
 
Join Date: Dec 2000
Posts: 23,230
Garage
Quote:
Originally Posted by 67dubcab View Post
Thanks Steve for chiming in, it was one of your posts that I read during my pre-posting research stage that mentioned the benefits of merged banks.
I understand that the details can make huge differences in performance and the whole system must be addressed- I wish I had the budget to dyno test multiple set-ups, I will eventually chassis dyno to get info to Steve Wong for a chip.

In the context of my build-3.2 993 h.e. mid -engine/ daily driver shop truck I still need to figure this out.

.....
Steve has been very free w/ the comments but very light on any supporting detail
__________________
Bill Verburg
'76 Carrera 3.6RS(nee C3/hotrod), '95 993RS/CS(clone)
| Pelican Home |Rennlist Wheels |Rennlist Brakes |
Old 02-19-2011, 01:05 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #21 (permalink)
Bland
 
RWebb's Avatar
 
Join Date: Jul 2007
Location: Greater Metropolitan Nimrod, Orygun
Posts: 42,390
Garage
my guess is they like to keep some secrets for their race customers
Old 02-19-2011, 03:10 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #22 (permalink)
Registered User
 
Join Date: Jul 2001
Location: Portland Oregon
Posts: 6,992
Quote:
Originally Posted by RWebb View Post
my guess is they like to keep some secrets for their race customers
Thats correct,....

Keeping all our clients competitive is our main priority and requires a great deal of dyno time and product development, none of which is inexpensive to do.
__________________
Steve Weiner
Rennsport Systems
Portland Oregon
(503) 244-0990
porsche@rennsportsystems.com
www.rennsportsystems.com
Old 02-19-2011, 03:28 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #23 (permalink)
Registered User
 
Join Date: Jul 2001
Location: Portland Oregon
Posts: 6,992
Quote:
Originally Posted by Bill Verburg View Post
Steve and I have disagreed about this for years, I have articulated my understanding of what is going on, the factory appears to agree w/ me based on what I've seen, Steve?
Perhaps not as much as you may think,... My only points have been to offer some context and real-world experiences that other may find as useful as the principles and theory.

Indeed, the factory used separate branch systems on all of their race engines since 1965 to provide peak HP at the expense of "area under the curve" as that has been their priority in professional competition. They have employed very close-ratio gearing and professional drivers to make their overall philosophy work and under those specific circumstances, its hard to argue with results.

Like many things about engines; we have all learned a LOT in the past 10-15 years about camshaft profiles, intake systems, cylinder head design and configuration as well as optimal exhaust systems (headers as well as mufflers) to achieve the desired results. In short, this is a very dynamic environment with a lot of ongoing changes and developments.

We are not always looking for absolute peak HP: instead we want a usable spread of power as well as competitive HP and to achieve that goal, we will employ a whole host of things that depend on each engine's displacement vehicle weight, gearing, and type of use.

Porsche's air-cooled and water-cooled race engines operate in very different RPM ranges and shouldn't be compared to each other. To get competitive power for professional racing requires a lot of RPM and the GT-3RSR engines have been optimized to operate to 9800 RPM. Naturally, they use very close-ratio sequential gearboxes to keep the engine in its best RPM range for this express objective.

Many of those principles do not translate cleanly to the air-cooled cars unless one is willing to live with a very narrow power range and frankly, thats not competitive for anyone who is not a very skilled driver with professional-level experience.

Most people do not understand how sensitive the air-cooled 911 engine is to exhaust configuration (I sure didn't until I began doing a lot of dyno development and testing many years ago). Its MUCH easier to make a good sounding system that looks right compared to one that makes more power than a factory system.

Further, each and every exhaust variable: header length, header diameter, collector design, secondary length, secondary diameter, muffler design, cross-over positioning (if used), and even tailpipe length all play a big role in the engine's power and torque curves. Crossovers (or common mufflers) also make a big difference in the air-fuel delivery curve and thats nothing to be ignored.

As one example, several years ago we were doing a 2.0 race engine looking for more power (naturally). Like all 2.0 race motors, these are little fussy and cantankerous when off-cam and this one was no different. By substituting our normal single large Flowmaster for a similar Magnaflow, we picked up over 18 BHP
and the air-fuel curve lost most of its peaks & valleys which translated into a much better driving machine. You cannot learn these things without hundreds of dyno hours and trying a LOT of different components and configurations.

My deepest apologies for running on, however I hope I've shed a little light on the process and to maybe illustrate that we are mostly on the same page. Only the context may vary.

I would also apologize for the lack of specific application data. I hope you understand how many resources ($$$$$) have been poured into this and to be fair to my customer base, some of this is proprietary. My main objective is to keep these folks as competitive as possible in a very competitive industry.

I'm always available to discuss this on the phone, too.
__________________
Steve Weiner
Rennsport Systems
Portland Oregon
(503) 244-0990
porsche@rennsportsystems.com
www.rennsportsystems.com

Last edited by Steve@Rennsport; 02-20-2011 at 04:35 PM..
Old 02-19-2011, 04:14 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #24 (permalink)
Moderator
 
Bill Verburg's Avatar
 
Join Date: Dec 2000
Posts: 23,230
Garage
So mysterious
Show a dyno chart, what better way to get customers calling?

Here everyone is wasting money on ssi and BB and S-Car-Go etc
__________________
Bill Verburg
'76 Carrera 3.6RS(nee C3/hotrod), '95 993RS/CS(clone)
| Pelican Home |Rennlist Wheels |Rennlist Brakes |
Old 02-19-2011, 05:00 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #25 (permalink)
Registered User
 
feelyx's Avatar
 
Join Date: Oct 2000
Location: sac, ca. usa
Posts: 1,137
Thanks 67. I went to the Samba and checked it out, That is going to be a fantastic ride.
Tim
__________________
----------------------------------------------------------------
98 Boxster
87 560sl
04 S2000
Old 02-19-2011, 05:09 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #26 (permalink)
Registered User
 
Join Date: Jul 2001
Location: Portland Oregon
Posts: 6,992
Quote:
Originally Posted by Bill Verburg View Post
Show a dyno chart, what better way to get customers calling?
I certainly share that data with our customers,....
__________________
Steve Weiner
Rennsport Systems
Portland Oregon
(503) 244-0990
porsche@rennsportsystems.com
www.rennsportsystems.com
Old 02-19-2011, 08:21 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #27 (permalink)
Moderator
 
Bill Verburg's Avatar
 
Join Date: Dec 2000
Posts: 23,230
Garage
There are other ways to design an H6 even fire exhaust, the best alternate being the tri-Y or 180* exhaust and it is correct that the length and diameter of the pipes as well as the merges are going to be a black art to derive, much more so than in a normal header collector setup , the principles are well known it's the implementation that is difficult as what needs to happen in a tri-y is for the rarefactions to be smeared into the compressions in such a manner as to encourage flow as opposed to discourage it or at least to have the rarefactions and compressions alternate rather than coincide. Again the effects are limited to relatively narrow rpm bands, In an even fire collector it's easy and natural to get the rarefactions and compressions to line up as that is a natural consequence of equal length primaries. tri-y's are generally thought to be best for torque, separate collectors for hp, individual, as on a drag rail or P51 for ultimate hp

here are the 6 pulses that are dealt w/ in our engines, this dos not show pressures only the timing of events in the collector, this is what happens when 6 equal length pipes are connected in a single collector, as you can see sometimes 2 compressions are trying to be in the same place at the same time, this is bad for performance and sometimes a compression and rarefaction come together, this is good. But over all this is a poor design



Once again I'll take the time to go though the known facts of the matter.

first there are 2 different broad firing patterns used in internal combustion engines, the firing pattern used is determined by the crankshaft design and angle between the cylinder banks,

they are odd fire and even fire, each has a different exhaust geometry requirement, Porsche knows this, when they build an odd fire engine they connect the banks, the best known example is the odd fire 910 flat 8s, on even fire engines the banks are kept separate

you need to understand a fundamental difference in the exhaust requirements between odd fire and even fire engines
90 deg crank, odd fire 90deg block, US V8 also Porsche 928
Left__collector: B-B-S-S-B-B-S-S
Right_collector: S-S-B-B-S-S-B-B

or another way to hear it L-R-R-L-R-L-L-R

the significant feature is the consecutive firing of each bank into the same collector

here you can see the summation of pulses in an odd fire collector, you can see the periodic overpressure that needs relief from connection to the opposite bank


180 deg crank, 90deg block V8 Ruf V8, Ferrari V8, Porsche RS Spyder V8 ,even 120 deg crank Porsche H6, 60 or 180* V12 etc
Left__collector: B-S-B-S-B-S-B-S
Right_collector: S-B-S-B-S-B-S-B

or L-R-L-R-L-R-L-R-

the significant feature is the alternate firing which allows the collector to breath

here you can see an H6 even fire collector, note the magnitude of the pressure which is reduced by the alteranate appearance of high and lows in the collector, there is a very slight periodic overpressure but it's magnitude is so small as too be of no concern




On an odd fire engine power and torque are increased by joining the 2 sides because otherwise each collector would need to be much larger or suffer from congestion due to the consecutive pulses of gas it receives. This does not happen on an even fire engine. On the odd fire engine the position and or geometry of the connection between sides doesn't matter a whole bunch. Early examples just put a T between the pipes anywhere, newer examples use a merge X to smooth the flow and reduce turbulence

alternatively on an odd fire engine you can use a180* or bundle of snakes exhaust, in this design a left cylinder goes to the right collector and a right cylinder goes to a left collector






there are different phenomena that are occurring in the exhaust, the 2 most important being gas flow and acoustic pressure waves.

gas flow is not continuous or homogeneous, it is lumpy and chaotic but still consists of a series of sinusoidal high pressure followed by low pressure zones that move through the system. The best flow is from reduced turbulence and wide separation of the exhaust lumps to minimize a pressure zones interference w/ it's cohorts . In a Porsche 6 there is a clump of hot, expanding gas delivered to each collector every 240 deg. of crankshaft rotation(a cyl. fires every 120* but the firing alternates sides), the clumps of hot gas expand and slow down as they pass through the pipes, fewer bends and adequate pipe size reduce the slowing but they tend to pile atop each other as the revs build, another thing that happens is the flow becomes more turbulent as it moves through the system, fewer bends turns and transitions reduce turbulence. The most efficient exhaust w/ the lowest pressure build up and the least pumping losses will make the most hp. This is achieved in an even fire engine by keeping the pipes as straight and unconnected side to side as is possible. The other big thing is the acoustic resonances that are set up in the pipes, very similar to those in a musical instrument. These can be useful but only under a set of very restrictive circumstances. The pipes have to have the correct geometry, this includes length, width, transitions, terminations. The resonances have to have something to work w/, that is the cams, which have to have enough overlap duration to suck fresh charge into the cylinder due to partial exhaust port vacuum, the desired effect only occurs in a narrow range in the engines rev band and there is a corresponding range where the pressure at the exhaust port isn't the desired low but rather high, impeding flow.

In a carefully crafted system these co-factors of gas flow and acoustic signals can work very well to enhance power and torque but the key is carefully crafted, not willy-nilly joining as on an odd fire engine. This is not to say that any odd fire doesn't benefit from the same gas flow/acoustic properties as an even fire, just
that the connection between banks only needs to be somewhere, where or the geometry is of no importance.

Further there are several different ways to construct these systems depending on whether hp over a limited range of revs or torque over a wider range of revs is desired.

for HP on an even fire engine you keep the sides separate for torque you join them but you join them in the way that the 3 pipes come together in a separate system.

This double collector setup is called a tri-Y or 180* system.
here's a great example,


the 3 pipes from one bank come together in the first collector and then the 2 sides are merged in the second collector, there is a whole science to the design of the collectors and Burns does make some great ones. What the tri-y does differently is smear the low pressure range out so that the signal is spread out more but is not as strong at any one point of the rev range, this broadens the torque curve

heres' an exaple of a Nascar tri-y



and here's a Burns custom setup for an even fire Ferrarri V12



Lastly a muffled system has a different termination or end than an unmuffled system, I mentioned earlier that transitions and terminations alter the acoustic signals and generate turbulence. The minute a muffler is slapped onto the best header in the world most of the acoustic benefits are lost, there are tricks that can be used such as extending the inlet pipe into the muffler cavity but this is only of limited usefulness.

For a muffled system w/ smog cams the best design goal is free flow, minimized turbulence and light weight, for an unmuffled system you can take advantage of the acoustic signals to generate low pressure at the exhaust port over a narrow range(keep them separate) or wider range(triY)

to go back to the o/p's original question, for a transporter a tri-y would be best for torque but a proper implementation is going to be difficult to come by, why not buy one of Steve's and let us know how it turns out.

For the rest of us that are more concerned w/ track/street performance a separated collector setup will be better.
__________________
Bill Verburg
'76 Carrera 3.6RS(nee C3/hotrod), '95 993RS/CS(clone)
| Pelican Home |Rennlist Wheels |Rennlist Brakes |
Old 02-20-2011, 08:29 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #28 (permalink)
A new breed
 
Twisted Minis's Avatar
 
Join Date: Jan 2010
Location: Windsor, California
Posts: 23
Here is the design we ended up using.

__________________
If you can find it cheaper, I can fix it!
Old 11-06-2011, 07:56 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #29 (permalink)
Registered User
 
ShakinJoe's Avatar
 
Join Date: Jul 2008
Location: St. Petersburg, FL
Posts: 2,468
Diggin this
__________________
'87 Carrera - 2400 lbs of Track Beast!!
'88 Carrera Cab - Too nice for the track.
'85 Targa - Salvage title that was not caught!
Old 11-06-2011, 08:34 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #30 (permalink)
A new breed
 
Twisted Minis's Avatar
 
Join Date: Jan 2010
Location: Windsor, California
Posts: 23
Here are some better pictures.





__________________
If you can find it cheaper, I can fix it!
Old 11-08-2011, 11:21 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #31 (permalink)
Registered User
 
skinnerd's Avatar
 
Join Date: Aug 2001
Location: Portland, OR USA & Bangkok, Thailand 50/50
Posts: 2,545
Garage
Pls post a vid or sound clip once you get this up and running.
Looks great!
__________________
"Are you out of your Vulcan mind?"

Doug
1973 RSR inspired genesis powr'd by dilithium crystals (3.6 RS Vram)
Old 11-09-2011, 01:05 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #32 (permalink)
 
Registered User
 
jluetjen's Avatar
 
Join Date: Oct 2001
Location: Westford, MA USA
Posts: 8,596
Garage
Great analysis and explanation Bill!
__________________
John
'69 911E

"It's a poor craftsman who blames their tools" -- Unknown
"Any suspension -- no matter how poorly designed -- can be made to work reasonably well if you just stop it from moving." -- Colin Chapman
Old 11-09-2011, 02:24 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #33 (permalink)
Registered User
 
Join Date: May 2007
Posts: 251
mid-engine exhaust design

Thanks all for your input on this subject and as you can see we (Seth) built something.

I will post up dyno and sound video when it is running, the before and chip optimized version.
Old 11-10-2011, 06:51 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #34 (permalink)
A new breed
 
Twisted Minis's Avatar
 
Join Date: Jan 2010
Location: Windsor, California
Posts: 23
Exhaust and bracing from above. Double Cab is done for this visit, and going back home soon. But I will try to get some high res pictures before it leaves.

__________________
If you can find it cheaper, I can fix it!
Old 11-18-2011, 03:09 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #35 (permalink)
Reply

Thread Tools
Rate This Thread
Rate This Thread:

 


All times are GMT -8. The time now is 03:21 PM.


 
Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2018 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website -    DMCA Registered Agent Contact Page
 

DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.