When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Was chit chatting with my neighbor today (big JDM guy who builds some fun stuff) and we were discussing extrude hone porting the S4 intake. I'm not real versed on any of this stuff but he said I might be able to gain a bit by doing that.
Here's what I think. I am not sure if it's correct or not, but I am sure that it's what I think! ;-)
If you can't use much camshaft overlap and headers, then big and slow ports give you decent top end performance, decent fuel economy, and a flattish torque curve without huge peaks and valleys. If you can use headers and and significant camshaft overlap, then you can make more power from the same displacement with smaller and faster ports.
I agree that in a high performance application, with headers and a camshaft with more overlap, velocity through the port is very important....especially if run in the lower rpm ranges.
However, it seems like most engines that have headers and a camshaft with more overlap are not lugging around town at low rpms....most are used in applications that keep the engine in the higher rpm ranges, which dilutes the importance of smaller ports and velocity to a large extent.
I think my confusion about your thoughts may be that I'm not understanding your intended application.
Here's what I think. I am not sure if it's correct or not, but I am sure that it's what I think! ;-)
If you can't use much camshaft overlap and headers, then big and slow ports give you decent top end performance, decent fuel economy, and a flattish torque curve without huge peaks and valleys. If you can use headers and and significant camshaft overlap, then you can make more power from the same displacement with smaller and faster ports.
Please understand that I'm not being critical of your thoughts or trying to be argumentative.
I enjoy your through research and subsequent thoughts.
Was chit chatting with my neighbor today (big JDM guy who builds some fun stuff) and we were discussing extrude hone porting the S4 intake. I'm not real versed on any of this stuff but he said I might be able to gain a bit by doing that.
Thoughts?
Every attempt to extrude hone the stock s4 manifold that I know of has resulted either in no change or a loss of power. This regardless of power level and displacement. The conclusion from this is (my conclusion anyway) that the problem with the s4 manifold is not that it's too small and that the real problem is the shape and extrude honing makes the shape problems worse. Instead of spending the money on extrude honing, if spend it on welding material to runner five and reshaping that disaster of a turn.
Every attempt to extrude hone the stock s4 manifold that I know of has resulted either in no change or a loss of power. This regardless of power level and displacement. The conclusion from this is (my conclusion anyway) that the problem with the s4 manifold is not that it's too small and that the real problem is the shape and extrude honing makes the shape problems worse. Instead of spending the money on extrude honing, if spend it on welding material to runner five and reshaping that disaster of a turn.
Thanks, that's the info I needed. Neighbor was looking at it and said it is either going to help or hinder it.
Was chit chatting with my neighbor today (big JDM guy who builds some fun stuff) and we were discussing extrude hone porting the S4 intake. I'm not real versed on any of this stuff but he said I might be able to gain a bit by doing that.
Thoughts?
I did quite a bit of Extrudahone "trials" on the 3.2 Carrera intakes. These intakes needed some serious work and I was cutting them apart to gain access to the choke areas inside and then welding them back together....a very labor intensive process.
The problem was that the "mudlike" abrasive material made the port exit (at the head) huge, before there was much improvement at the internal choke areas.
This left me with the option of making the port on the head bigger, or leaving a giant "ridge" for the incoming air to hit on.
Neither were good alternatives.
Combine this with the fact that airflow is laminar and really dislikes a smooth polished surface....and the results were really poor.
I performed multiple tests where these intakes actually lost power, forcing me to go back to the labor intensive cutting, porting, and welding.
I see no reason that an S4 intake would not have the same issues.
Here's my opinion of the s3 vs s4 intake and heads.
First, EZK will allow one to run more advance than EZF. To make a fair test of the hardware, one should run high octane race gas in both and set the spark timing to best torque with any knocks or knock retards. Also, to make the manifold active resonance effects comparable, one should set the flappy permanently closed in the s4 manifold for this test. It would be interesting to see what the results would be.
The s4 port is not only bigger than s3, but also better shape. Basically over time with better design techniques the four valve intake ports have become more oval and wider. S4 was a step in right direction. S4 intake valve size is also more suitable for 100mm bore than the smaller s3 valves. What I don't like about the s4 intake port is that in my opinion for that valve size the port proper slightly upstream of the valve guides should be a bit smaller in cross sectional area. I think that s3 is better there. S3 camshafts are bigger which will help at high rpms, but the lsa of 114 degrees is in my opinion too wide considering the valve size. The 106.5 of s4 camshafts is probably a big reason why it makes more mid range torque.
I agree that a too small port is somewhat tautologically too small and the ill effects of too small will be felt at higher rpms. The question is what's too small. When I use pipe max to size the ports and intake runners for 7000 rpm and 450 hp, the actual s4 port and intake runner diameter are out of range too large for a highly tuned motor at that rpm and power levels.
The motorcycle engines that are highly tuned to start with appear to pick up some peak power from porting and part of the porting process is reducing the port size from what the factory made it in the 1980's and 1990's.
Originally Posted by GregBBRD
I agree that in a high performance application, with headers and a camshaft with more overlap, velocity through the port is very important....especially if run in the lower rpm ranges.
However, it seems like most engines that have headers and a camshaft with more overlap are not lugging around town at low rpms....most are used in applications that keep the engine in the higher rpm ranges, which dilutes the importance of smaller ports and velocity to a large extent.
I think my confusion about your thoughts may be that I'm not understanding your intended application.
Well there's always this option (part of he reason it works so well is that it addresses flow turbulence at the front of the TB plenum, and in the 45deg elbow approaching the TB).
Freeing up manifold airflow will allow any downstream improvements to really shine.
Happy to walk anyone through the processes involved.
If you think these ports and intake are big.....you need to find a 968 (3.0 liter 4 cylinder) head and intake to look at.
I find the thought really interesting that Porsche, with their vast engineering skills and their huge testing facilities, not to mention their ability to have outside sources help them with engine development, would make the mistake of making port and intake systems too large....over and over again.
Certainly, if I had been involved in the development of the 968 engine, I would have certainly walked over to the shelf, plucked a stock 928 head off of it, and ran it on that engine, long before I went to the cost of having a completely new head designed and built.
The fact that they used bigger valves and made the port much, much larger must have some basis.
It's not just the 968, but all the 16V heads for the 944 series.
Years ago when Todd first got involved with his own 928 he took my 944S head to compare it to the S4 unit and noted how much larger the ports were.
Good point RE: the base circle reduction affecting relief geometry, it passed through my mind in thinking about my response but at the lifts I'm using the reliefs are a moot point.
How high in terms of valve lift do each of these cams that you have in your stash get by 75.5 crankshaft degrees ATDC? With stock stroke and stock rod length, that's when the piston demands the most air into the cylinder.
Well there's always this option (part of he reason it works so well is that it addresses flow turbulence at the front of the TB plenum, and in the 45deg elbow approaching the TB).
Freeing up manifold airflow will allow any downstream improvements to really shine.
Happy to walk anyone through the processes involved.
I've too done quite a bit of work to the stock intake manifold, over the years....but I'm always combining that work with larger displacement, different camshafts, cylinder head work, etc. It is therefore difficult to separate which change did what.
I have built one engine, in particular, with slightly larger cams and some cylinder head port work with a stock manifold at the client's request (economy move). The engine was disappointing, as I anticipated. Before we could go back and make a proper intake manifold, the car was sold.
I found your testing/results on a stock engine very interesting. I'd certainly like to compare what you have found out with what I have done.
How high in terms of valve lift do each of these cams that you have in your stash get by 75.5 crankshaft degrees ATDC? With stock stroke and stock rod length, that's when the piston demands the most air into the cylinder.
So I've been playing with cams, and for whatever it's worth, 11.2 mm / 10.1 mm of lift and 227/216 at .05" (1.27mm) is more lift and duration than ever ran in any of Anderson's or Fan's 6.5L motors. That would be awfully sporty on a 5-liter.
(Which begs the question of how much lift one can have on a 32V cam before you start to get the lobe too close to the edge of the 35 mm lifter? There's a picture of Ake's somewhere showing him having to enlarge the radius of the cuts in the heads for lobe clearance. That's 'crazy'!
More relevant to your modeling is Jim Corenman's 5-liter GT, which for a time ran Greg's intake in October 2013. Jim's cams are 10.7 and 10.1 lift, duration at 1mm are 220/220, LSA is 108. He's got good headers and exhaust, same ST-Alpha as me. 367 rwhp, 336 torque, STD. 8-10 less for SAE.
So that would be 420ish NA chp assuming 15% driveline loss.
The runner's on Greg's intake are 2-1/8". No idea how long they are.
So using those numbers as inputs, if you play with runner diameters, what happens to predicted hp/tq?
Rob, aggressive race camshafts having a high rate of lift velocity needs lifters of larger diameter than mild cams with low rate of lift velocity. If you cannot install lifters large enough radiused lifters a la GT3 or BMW M5 is the way to go to keep the lift velocity up. High lift velocity is good for power but hard on the valve train. I did some calculations for the stock S3 and the stock GTS intake cams and the smallest safe lifter diameter for both cams is 30mm much smaller than the stock 35mm lifters.
Didn´t I show you the WEB #458 hydraulic cam profile which I have used for the 2V engines? I think this is about the most cam (lift .520" duration 258 dgs @ .050") that can be run on the stock size lifters.
About the nose of the lobe touching the edge of the lifter bore is very common on many engines when installing high lift cams. I first came across the problem when we started to modify the old Kawasaki 900 Fours back in the 70´s. This is usually taken care of by using the die grinder but if you like it made by high precision a cutting tool of the type I made will be necessary.
There are many more problems that can occur when installing high lift cams like valve guides that need to be replaced with shorter ones or the head will not allow longer valve springs for the increased valve lift which call for the need of cutting down the spring seat in the head and/or installing special top retainers for increased spring installed height.
Åke
Last edited by Strosek Ultra; 10-05-2015 at 07:24 AM.
Are the 928 32V intake ports too large? The answer is yes and no. For an all stock engine running small cams with hardly no overlap it works good. If we like to build a high performance 5 liter NA engine the intake ports are too large unless you want to build a very high revving engine which I have not seen yet. The port CSA at the flange (injector cutout not included) is 1700 mm2 or 2.635 sqin. 40mm intake valves (3mm oversize) will match the port size. Porting shall be concentrated to the divided ports and the valve seat area including optimizing of the valve head shape. Everything optimized for maximum power output (cams, porting, ITBs headers etc), we will end up with an engine having max power at about 9000 to 9500 rpm which is extremely high up. If we put the same stuff onto a 7 liter stroker we will end up with an engine having much more torque (probably about 40% more) at lower rpm and peak power will be at about 7000 rpm, a much more driveable engine.
The 9000 rpm engine is not unrealistic at all. A couple of years ago I built a V2 Moto Guzzi road racing engine for a customer. The stroke 78mm is same as for the 928 engine. The bore 95mm is same as for the 4,5 liter 2V engine. The Moto Guzzi is a 2-valve pushrod engine with limitation of how much valve lift can be used. The valves are 50/40mm, lift 11mm, carburetors 41,5mm, stepped headers 1.75" to 1.875" to 2.0". The engine did put out 100 hp at the wheel at 8500 rpm, red line at 9250 rpm, max torque at 6500 rpm. According to the owner the engine is pulling good from 3000 rpm.
The picture is showing an intake port having unchanged CSA at the flange but ported for 39mm valves (398 CFM @ 14mm). I did not bother polish the port which was made for flow testing only. Polishing is a lot of additional work and it does not bring anything in respect of flow but looks nicer.
Åke
Last edited by Strosek Ultra; 10-05-2015 at 10:26 AM.
10-04-2015, 06:24 PM
