S4 intake manifold facts and ideas
05-13-2013, 10:53 AM
#31
Nordschleife Master
Thread Starter
Per Jim's measurements, the steady state flow appears very closely matched as they come from the factory, which on the one hand is surprising because the runners look so different but on the other hand isn't really that surprising because it's a manifold designed by a car factory. They do have flow benches and stuff at the factory R&D department... ;-)
This then pegs the question why Todd Tremel observes lower air flow (and the need to back off fuel) to cylinder #5. If we eliminate the steady state flow as a source of discrepancy, only resonance effects remain to explain observed imbalances greater than just a couple of percent. I guess the next thing to know is by how much he had to back off the fuel and at what rpms. Hacker wrote in his post that the optimal fuel trim varies per rpm per cylinder. This would also be consistent with the resonance effects and not simple flow restrictions being the culprit for the observed flow imbalance. In particular, the runner #5 has that expansion in the area at the turn and then the turn wall, maybe that's reflecting something back to the cylinder.
I have to give Todd Tremel a call this week on an unrelated issue so I might as well ask him that, too.
05-13-2013, 11:08 AM
#32
Rennlist Member
We got a slightly better result than Greg with a 5.5ltr bottom end, but that had ITBs and 12:1 compression and trick trumpets... Maybe the intake is a restrictor after all... Colin, with all due respect, I find it hard to believe that a GT bottom end can put down these kind of numbers...
05-13-2013, 12:20 PM
#33
Nordschleife Master
Thread Starter
We got a slightly better result than Greg with a 5.5ltr bottom end, but that had ITBs and 12:1 compression and trick trumpets... Maybe the intake is a restrictor after all... Colin, with all due respect, I find it hard to believe that a GT bottom end can put down these kind of numbers...
Jim Corenman's result of 350 rwhp at only 6200 or so rpm looks pretty damn impressive. From earlier in this thread. That's a GT starting point with bigger cams, Devek headers, and free-flowing exhaust. Wouldn't be shocked if some time was spent Sharktuning as well... ;-)
Judging from the shape of the curve, he'd made even more power had he run it to higher rpms:
05-13-2013, 12:45 PM
#34
Nordschleife Master
Alex,
I'm only reporting what was made.
Same Dyno put down 285 with a strong stock S4 (91), 220rwhp on a heavily modified 84 engine, and 352.2 on this GT with big cams and full exhaust. I want to get more cars on this Dyno for more baselines. But I cannot afford to pay to test each car on here.
I'm only reporting what was made.
Same Dyno put down 285 with a strong stock S4 (91), 220rwhp on a heavily modified 84 engine, and 352.2 on this GT with big cams and full exhaust. I want to get more cars on this Dyno for more baselines. But I cannot afford to pay to test each car on here.
05-13-2013, 01:51 PM
#35
Rennlist Member
Alex,
I'm only reporting what was made.
Same Dyno put down 285 with a strong stock S4 (91), 220rwhp on a heavily modified 84 engine, and 352.2 on this GT with big cams and full exhaust. I want to get more cars on this Dyno for more baselines. But I cannot afford to pay to test each car on here.
I'm only reporting what was made.
Same Dyno put down 285 with a strong stock S4 (91), 220rwhp on a heavily modified 84 engine, and 352.2 on this GT with big cams and full exhaust. I want to get more cars on this Dyno for more baselines. But I cannot afford to pay to test each car on here.
05-13-2013, 02:02 PM
#37
Inventor
Rennlist Member
Rennlist Member
Note that it's STandarD correction, though. SAE would be likely be <340. (SAE is a B.)
I pulled 320 STD in a '86.5 5-speed with X-pipe and chips.
05-13-2013, 02:10 PM
#38
Rennlist Member
05-13-2013, 02:15 PM
#39
Rennlist Member
I'd certainly share my flow bench data....if it was meaningful. My flow bench simply will not "pull" enough air to get really meaningful data out of an intake manifold....at least not enough air to find the inadequacies of this manifold. I have to extrapolate data to get numbers at 10 inches....and
extrapolation just doesn't work well for meaningful data at high flow rates.
I was going to go over and borrow a friends larger flow bench and do some testing.....when I "hit the wall" on Sean Smith's engine, during two solid months of testing.....
Allow me to try to explain....all "power numbers" are plus or minus 4-5 hp......and all "changes' were re-Sharktuned before testing, after original Sharktuning was performed. Sharkplotter did not exist, at this time, and neither did John Speake's "easy to change" chips....all were individually burned eproms.
6.5 liter stroker engine. 968 Intake valves. Ported heads. 928 GTS camshafts at stock specifications Stock exhaust manifolds. Stock intake manifold. High flow cats with x-pipe and high flow mufflers. 19lb injectors. Higher fuel pressure to add additional fuel. 360hp.
Same as above with 30 pound injector and Sharktuning. 360hp.
Same as above with 100 octane race fuel and significantly increased timing. 360hp.
Same as above with GT specification exhaust cams and different exhaust valve springs. 360hp.
Same as above with GT specification intake cams and different intake valve springs. 360hp.
Same as above at a completely different dyno facility. 360hp.
Same as above with a completely different torque convertor design. 360hp.
Same as above with cams advanced 4 degrees. 360hp.
Same as above with cams retarded 4 degrees. 360hp.
Same as above with intake cams advanced 4 degrees and exhaust cams at std. 360hp.
Same as above with intake cams retarded 4 degrees and exhaust cams at std. 360hp.
Same as above with exhaust cams advanced 4 degrees and intake cams at std. 360hp.
Same as above with exhaust cams retarded 4 degrees and intake cams at std. 360hp
Same as above with "intake spacers". 360hp.
Same as above with highly ported intake manifold (no spacers). 360hp.
Same as above with intake spacers and modifications to "blend in" intake spacers. 360hp.
Same as above with 80mm throttle plate. 360hp.
Same as above with highly modified lower plenum (20 hours of welding and porting). 360hp.
Same as above with highly modified intake "bells". 360hp.
Same as above with stock GT exhaust.....just to see if anything might possibly make a change. 340hp.
$2500 worth of dyno testing and two solid months of R&D and changing parts......
Found out that this stuff isn't quite as easy as one might think, reading Rennlist.
Sanity, my wife, and my bank all required a stop to testing.
And with regards to this thread....I do not personally believe that the intake manifold was a significant "restriction" in the above engine.
By God, this stuff is so much fun!!!!
extrapolation just doesn't work well for meaningful data at high flow rates.
I was going to go over and borrow a friends larger flow bench and do some testing.....when I "hit the wall" on Sean Smith's engine, during two solid months of testing.....
Allow me to try to explain....all "power numbers" are plus or minus 4-5 hp......and all "changes' were re-Sharktuned before testing, after original Sharktuning was performed. Sharkplotter did not exist, at this time, and neither did John Speake's "easy to change" chips....all were individually burned eproms.
6.5 liter stroker engine. 968 Intake valves. Ported heads. 928 GTS camshafts at stock specifications Stock exhaust manifolds. Stock intake manifold. High flow cats with x-pipe and high flow mufflers. 19lb injectors. Higher fuel pressure to add additional fuel. 360hp.
Same as above with 30 pound injector and Sharktuning. 360hp.
Same as above with 100 octane race fuel and significantly increased timing. 360hp.
Same as above with GT specification exhaust cams and different exhaust valve springs. 360hp.
Same as above with GT specification intake cams and different intake valve springs. 360hp.
Same as above at a completely different dyno facility. 360hp.
Same as above with a completely different torque convertor design. 360hp.
Same as above with cams advanced 4 degrees. 360hp.
Same as above with cams retarded 4 degrees. 360hp.
Same as above with intake cams advanced 4 degrees and exhaust cams at std. 360hp.
Same as above with intake cams retarded 4 degrees and exhaust cams at std. 360hp.
Same as above with exhaust cams advanced 4 degrees and intake cams at std. 360hp.
Same as above with exhaust cams retarded 4 degrees and intake cams at std. 360hp
Same as above with "intake spacers". 360hp.
Same as above with highly ported intake manifold (no spacers). 360hp.
Same as above with intake spacers and modifications to "blend in" intake spacers. 360hp.
Same as above with 80mm throttle plate. 360hp.
Same as above with highly modified lower plenum (20 hours of welding and porting). 360hp.
Same as above with highly modified intake "bells". 360hp.
Same as above with stock GT exhaust.....just to see if anything might possibly make a change. 340hp.
$2500 worth of dyno testing and two solid months of R&D and changing parts......
Found out that this stuff isn't quite as easy as one might think, reading Rennlist.
Sanity, my wife, and my bank all required a stop to testing.
And with regards to this thread....I do not personally believe that the intake manifold was a significant "restriction" in the above engine.
By God, this stuff is so much fun!!!!
I think what you have shown by all your testing it is the intake that is restrictive!
05-13-2013, 02:33 PM
#40
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Everyone is entitled to an opinion about everything, on the Internet. No qualifications required.
I'm sure that you have done way more testing, engineering, and fabricating on the 928 vehicle than I have....
I will grant you this:
Obviously, the intake system is going to be a restriction, at some point, however I do not think it was/is the problem, here. (Why else would I be building an alternative intake system?)
In my current theory (and road that I'm following), much of Anderson's improvements were actually a "side effect" of the manifold change, not directly the result of the manifold change to airflow.
__________________
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
greg brown
714 879 9072
GregBBRD@aol.com
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
05-13-2013, 02:54 PM
#41
Three Wheelin'
Join Date: Feb 2002
Posts: 1,663
Likes: 0
Received 0 Likes
on
0 Posts
Tuomo, et al:
Without going into pages of history, I can provide some info about the stock S4 intake manifold.
About 4 years ago, I worked with BC Gerolamy Co. in Rancho Cordova to develop a pair of 928 4V cylinder heads for a "hot rod" project. (NOTE for Rennlisters, this was not Dennis Kao's engine). Using a flow bench and lots of experience and information from BC Gerolamy, the modified heads flowed quite well, but tested using a bellmouth at the manifold connection face. The flow data was relevant for the success of the port / valve work, but did not matter much if going to be mated to a S4 manifold. In short, the whole intake system has to work together, not just the heads.
At this point, we spent several hours doing flow bench tests on the S4 manifold and learned some interesting things relative to this thread.
1.) The flow and velocities through each runner varied enough to suggest creating inconsistent combustion in each cylinder... not acceptable for a batch fire injection engine.
2.) For a stock set of heads, the manifold was "not bad", but if you work the heads, the manifold needs work hand in hand with the heads.... bad results if not worked as a matched set !
2.) Simply "extrude honing" the manifold MIGHT improve one or two of the runners, but would definitely make matters worse for other runners, especially the tight turns ones.
3.) The runner bellmouths on #5 and #8 become restricted with the manifold plenum end plates are mounted. Flow does not significantly improvement with spacers less than 5/8 of an inch. Some smaller amount of spacer thickness helps, but not as much as thought if you are trying to get best flow from the stock cast manifold.
4.) After several hours of hand porting and testing, the manifold was able to flow around 285-288 CFM @ 25inches, but did require some welding to maintain wall thickness in some of the tighter corners. Welding the Mg manifold is tricky given the need to purge the air from runner being welded. Plumber test plugs work well to plug one side of the runner for purging
5.) Even at 288 CFM, the manifold was never able to flow for possible long block HP. The runner area / length models better than observed results. In short, getting the heads to flow is much easier than creating a good working manifold.
Based on the tests done, lots of ideas were generated for future work. The costs of exploring these ideas were quite high given the risk surrounding the likely success of the total engine. After a few years of attempting to win the lottery to fund further research, I folded...
Best wishes to those still working this stuff... it's lots of fun... just a matter of $$$
Without going into pages of history, I can provide some info about the stock S4 intake manifold.
About 4 years ago, I worked with BC Gerolamy Co. in Rancho Cordova to develop a pair of 928 4V cylinder heads for a "hot rod" project. (NOTE for Rennlisters, this was not Dennis Kao's engine). Using a flow bench and lots of experience and information from BC Gerolamy, the modified heads flowed quite well, but tested using a bellmouth at the manifold connection face. The flow data was relevant for the success of the port / valve work, but did not matter much if going to be mated to a S4 manifold. In short, the whole intake system has to work together, not just the heads.
At this point, we spent several hours doing flow bench tests on the S4 manifold and learned some interesting things relative to this thread.
1.) The flow and velocities through each runner varied enough to suggest creating inconsistent combustion in each cylinder... not acceptable for a batch fire injection engine.
2.) For a stock set of heads, the manifold was "not bad", but if you work the heads, the manifold needs work hand in hand with the heads.... bad results if not worked as a matched set !
2.) Simply "extrude honing" the manifold MIGHT improve one or two of the runners, but would definitely make matters worse for other runners, especially the tight turns ones.
3.) The runner bellmouths on #5 and #8 become restricted with the manifold plenum end plates are mounted. Flow does not significantly improvement with spacers less than 5/8 of an inch. Some smaller amount of spacer thickness helps, but not as much as thought if you are trying to get best flow from the stock cast manifold.
4.) After several hours of hand porting and testing, the manifold was able to flow around 285-288 CFM @ 25inches, but did require some welding to maintain wall thickness in some of the tighter corners. Welding the Mg manifold is tricky given the need to purge the air from runner being welded. Plumber test plugs work well to plug one side of the runner for purging
5.) Even at 288 CFM, the manifold was never able to flow for possible long block HP. The runner area / length models better than observed results. In short, getting the heads to flow is much easier than creating a good working manifold.
Based on the tests done, lots of ideas were generated for future work. The costs of exploring these ideas were quite high given the risk surrounding the likely success of the total engine. After a few years of attempting to win the lottery to fund further research, I folded...
Best wishes to those still working this stuff... it's lots of fun... just a matter of $$$
05-13-2013, 03:35 PM
#42
Race Director
great thread....lots of information way beyond my comprehension....but still interesting none the less....
Personally I lean more towards the "testing" side of all this.....since I don't know enough to figure out all the math-engineering stuff I make a change that seems to make sense to me.....and go out and run it.....or race it..... Its been working so far....
BUT the one thing I firmly believe in.....is the value of experts....and utlilizing their expertise to help with your project......there is a HUGE brain trust in this thread.....we are luckly they are so willing to share their knowledge with us
Personally I lean more towards the "testing" side of all this.....since I don't know enough to figure out all the math-engineering stuff I make a change that seems to make sense to me.....and go out and run it.....or race it..... Its been working so far....
BUT the one thing I firmly believe in.....is the value of experts....and utlilizing their expertise to help with your project......there is a HUGE brain trust in this thread.....we are luckly they are so willing to share their knowledge with us
05-13-2013, 04:04 PM
#43
Inventor
Rennlist Member
Rennlist Member
What about restricting the rest of the runners?
Less overall CFM, but a more balanced output, cylinder to cylinder.
A more even AFR might allow for more ignition advance, or at least less knock retard from the lean cylinders.
05-13-2013, 05:21 PM
#45
Nordschleife Master
Thread Starter
By the way, there's more about S3/S4 manifolds here:
Porsche patent # 4641610
