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.
Tire smoking torque off the line but peak HP would take a big hit as the exhaust valves close too early. One cam tooth is 7.5° (crank). There is roughly a two tooth range in the cam gear slots (±1). IIRC, folks have skipped three cam teeth without valve damage.
I was suggesting advancing the (S3) intake camshafts only, by one cam tooth on the cam chain, not on the (exhaust) cam pulley.
I believe each camshaft features 19 teeth, meaning each tooth is 360/19*2=38° crank.
Last edited by Thom; 09-24-2020 at 10:13 AM.
Reason: edited to correct calc
I was suggesting advancing the (S3) intake camshafts only, by one cam tooth on the cam chain, not on the (exhaust) cam pulley.
I believe each camshaft features 19 teeth, meaning each tooth is 360/19/2=9.5° crank.
Not correct, the camshaft chain gear has 19 teeth, in other words it is about 19 degrees between two teeth. The camshaft rotates half the speed of the crankshaft. To move the camshaft 19 degrees, the crankshaft need to be moved 38 degrees.
Åke
Kaase continued, "I talked to Comp Cams and they helped me out with ideas on the duration. I had two different sets of cams, and in the end had one set reground, so I had quite a few variations to try out. I tried a lot of different things, and it is basically trial and error. I started out with the biggest cams, and I had those cams in every possible position, with wide lobe separation, narrow separation, way advanced, way retarded, just everywhere, because they are easy to move. It was pretty consistent that the intake wanted to be at around 98 degrees centerline, and when I finalized everything, the exhaust was the same; it wanted to be at 98 degrees. You can say I was at 98-degree lobe-separation angle and zero advance.“ https://www.hotrod.com/articles/jon-...lve-mod-motor/
This is an engine that needs to make maximum power and torque in the 3000-7000 rpm range. To do that with a four valve engine, the optimum LSA ended up being 98 degrees. This is not an engine that needs to have a street idle or the ability cruise at part throttle.
In my opinion, the street car LSA will end up being whatever it ends up being.
The considerations that lead to that start with the camshaft overlap. One first has to decide what is an acceptable idle rpm, acceptable idle emissions level (if any limits), and acceptable part throttle cruise behavior. Then, one needs to decide whether one can use ITBs or a plenum manifold. ITBs can produce an acceptable idle quality and part throttle performance with much greater camshaft overlap than a single-throttle plenum manifold. Finally, with a 90-degree cross-plane V8, one has to decide whether the engine will have long-tube headers or exhaust manifolds. Because of the exhaust blowdown interference, engines with headers can tolerate more camshaft overlap. These considerations give one IVO, EVC, and overlap.
Once the overlap and IVO and EVC events have been decided, then one needs to decide the durations and EVO and IVC. The main considerations here are the redline rpm and intake tract flow capability and tuning. Higher the target peak power rpm, later the IVC and longer the duration. The exhaust duration EVO will depend on the volumetric efficiency and exhaust port size, the basic idea being is that the exhaust duration should be the shortest possible duration that still can empty the cylinder. The short exhaust duration with headers can generate a very powerful rarefaction exhaust wave during the overlap and band-aid bad intake tract. For example, NASCAR engines shorten the exhaust duration dramatically when a restrictor plate is fitted but maintain approximately the same overlap as the unrestricted engine.
In my opinion, when cams are designed in this way, the LSA can fall pretty much anywhere depending on the purpose of the engine. As a personal observation, I would say that engines with low redline rpm, high idle rpm, restrictive intake, and headers will want cams that have tight LSAs. Engines with high redline rpm, low idle rpm, free-flowing intake, and exhaust manifolds will want cams that have wide LSAs.
Finally, with a 90-degree cross-plane V8, one has to decide whether the engine will have long-tube headers or exhaust manifolds. Because of the exhaust blowdown interference, engines with headers can tolerate more camshaft overlap.
Could the blowdown interference be the main reason behind negative overlap cams on most street cross plane crank V8 engines?
I suppose positive overlap cams invariably lead to Harley-Davidson levels of idle quality?
May I ask the LSA and overlap values on your current TT engine?
Kaase continued, "I talked to Comp Cams and they helped me out with ideas on the duration. I had two different sets of cams, and in the end had one set reground, so I had quite a few variations to try out. I tried a lot of different things, and it is basically trial and error. I started out with the biggest cams, and I had those cams in every possible position, with wide lobe separation, narrow separation, way advanced, way retarded, just everywhere, because they are easy to move. It was pretty consistent that the intake wanted to be at around 98 degrees centerline, and when I finalized everything, the exhaust was the same; it wanted to be at 98 degrees. You can say I was at 98-degree lobe-separation angle and zero advance.“ https://www.hotrod.com/articles/jon-...lve-mod-motor/
This is an engine that needs to make maximum power and torque in the 3000-7000 rpm range. To do that with a four valve engine, the optimum LSA ended up being 98 degrees. This is not an engine that needs to have a street idle or the ability cruise at part throttle.
In my opinion, the street car LSA will end up being whatever it ends up being.
The considerations that lead to that start with the camshaft overlap. One first has to decide what is an acceptable idle rpm, acceptable idle emissions level (if any limits), and acceptable part throttle cruise behavior. Then, one needs to decide whether one can use ITBs or a plenum manifold. ITBs can produce an acceptable idle quality and part throttle performance with much greater camshaft overlap than a single-throttle plenum manifold. Finally, with a 90-degree cross-plane V8, one has to decide whether the engine will have long-tube headers or exhaust manifolds. Because of the exhaust blowdown interference, engines with headers can tolerate more camshaft overlap. These considerations give one IVO, EVC, and overlap.
Once the overlap and IVO and EVC events have been decided, then one needs to decide the durations and EVO and IVC. The main considerations here are the redline rpm and intake tract flow capability and tuning. Higher the target peak power rpm, later the IVC and longer the duration. The exhaust duration EVO will depend on the volumetric efficiency and exhaust port size, the basic idea being is that the exhaust duration should be the shortest possible duration that still can empty the cylinder. The short exhaust duration with headers can generate a very powerful rarefaction exhaust wave during the overlap and band-aid bad intake tract. For example, NASCAR engines shorten the exhaust duration dramatically when a restrictor plate is fitted but maintain approximately the same overlap as the unrestricted engine.
In my opinion, when cams are designed in this way, the LSA can fall pretty much anywhere depending on the purpose of the engine. As a personal observation, I would say that engines with low redline rpm, high idle rpm, restrictive intake, and headers will want cams that have tight LSAs. Engines with high redline rpm, low idle rpm, free-flowing intake, and exhaust manifolds will want cams that have wide LSAs.
What you say is very accurate.
The issue, of course, is translating the above to numbers, making a master, and grinding cams with have a chain connecting the two, which determines LSA.
A small error in sprocket to cam lobe location and you can have LSA numbers different than your target.
Could the blowdown interference be the main reason behind negative overlap cams on most street cross plane crank V8 engines?
I suppose positive overlap cams invariably lead to Harley-Davidson levels of idle quality?
May I ask the LSA and overlap values on your current TT engine?
While the cam specs say that many street cams, including those for the 928, have negative overlap, in reality, they don’t. The overlap may be negative at 1mm or 0.05” lift valve events, but it is positive at seat events. The flow area and air flow are there during the overlap. And even if the cams were truly negative overlap, they’d still behave as if they’d have some overlap because the combustion chamber acts as a pressure/vacuum reservoir.
I think that exhaust manifolds and cross plane crank explain why many production 32-valve V8s have relatively little camshaft overlap.
My car has “emission” cams, so 114 LSA and duration somewhat longer than GT/S3. They work well in my engine with turbos, stock intake, and cast compact exhaust manifold. The thread has some more cam values.
One of the things that a MAF injection system does not tolerate is any intake "blowback" through the intake system that will reach the MAF...which is why the opening events are always timed ADTC and the base circle diameter continues so far up the ramps.
The problem, for Porsche, was creating a camshaft that was more aggressive, yet still did not have intake blowback to the MAF. On the GT cams, they moved the opening event as close to TDC as they could, without incurring this.
And from there, Porsche did not have to think very hard or long...
My perspective is slightly different. This is what I believe now, after having been involved in dual cartridge MAF conversion with the stock LH computers:
The special tuning difficulties with the 928 S4 and big cams, compared to any modern single throttle plenum manifold EFI system that also always have some difficulties with big camshaft overlap, don’t in my opinion come from the intake manifold pulsing the MAF. The MAF doesn’t need to measure air flow correctly, it just needs to measure it consistently. Furthermore, if the problems were caused by pulsing, later IVC would give the system big problems, whereas it’s mostly just the overlap that seems give the system problems.
Instead, I believe that the real problem is that LH version we have doesn’t have throttle position sensor or input, it just has the throttle WOT position switch. Because of this, the system needs to infer acceleration mostly from the change in MAF signal and rpm. The stock MAF reacts to air flow changes slowly and smoothing the signal makes it slower yet. The system leans out with big cams when the throttle is opened suddenly, and then usually overcompensates.
On top of that, there’s the part throttle instability in cylinder filling that high overlap cams produce in any single-throttle plenum manifold system.
Instead, I believe that the real problem is that LH version we have doesn’t have throttle position sensor or input, it just has the throttle WOT position switch. Because of this, the system needs to infer acceleration mostly from the change in MAF signal and rpm. The stock MAF reacts to air flow changes slowly and smoothing the signal makes it slower yet. The system leans out with big cams when the throttle is opened suddenly, and then usually overcompensates.
For what little it is worth I came to a similar conclusion during my shark tuner efforts. The system has no predictive element and when the throttle is suddenly opened there is a large [unsteady state] in-rush that the system cannot respond. The system relies on feedback from a rising signal and they use preset compensation parameters for transient enrichment that decay over the first 4 firing pulses [ if my memory serves me correctly]. With the stock cams/inlet/exhaust it was tuned for it works reasonably well but start dicking around with cam profiles and whatever else one throws at it and simply put it is not a "one size fits all". Once the throttle is mashed open full bore the gas velocity initially tries to go sonic until dynamic equilibrium is achieved and it is impossible to meter that initial flow pulse as the control system wiil not even see it.
In a perfect world we would have a brain implant that tells the computer what is about to happen as we think about it so that it can preemptively chuck in more fuel to compensate appropriately - a form of feed forward control! I did wonder if a pressure sensor on the throttle pedal would be sufficient "advance notice" to mitigate this problem and thus initiate a transient enrichment. For sure it would be "quicker" than the stock setup but whether that increment would have a real world value not so sure. Assuming with most drivers it takes 100 msec to get from idle to full throttle opening I suspect such time advantage could be beneficial in throttle response terms.
Put simply- by relying on feedback from a rising MAF signal, rising rpms or a full throttle switch the horse has long since bolted control wise.
For what little it is worth I came to a similar conclusion during my shark tuner efforts. The system has no predictive element and when the throttle is suddenly opened there is a large [unsteady state] in-rush that the system cannot respond. The system relies on feedback from a rising signal and they use preset compensation parameters for transient enrichment that decay over the first 4 firing pulses [ if my memory serves me correctly]. With the stock cams/inlet/exhaust it was tuned for it works reasonably well but start dicking around with cam profiles and whatever else one throws at it and simply put it is not a "one size fits all". Once the throttle is mashed open full bore the gas velocity initially tries to go sonic until dynamic equilibrium is achieved and it is impossible to meter that initial flow pulse as the control system wiil not even see it.
In a perfect world we would have a brain implant that tells the computer what is about to happen as we think about it so that it can preemptively chuck in more fuel to compensate appropriately - a form of feed forward control! I did wonder if a pressure sensor on the throttle pedal would be sufficient "advance notice" to mitigate this problem and thus initiate a transient enrichment. For sure it would be "quicker" than the stock setup but whether that increment would have a real world value not so sure. Assuming with most drivers it takes 100 msec to get from idle to full throttle opening I suspect such time advantage could be beneficial in throttle response terms.
Put simply- by relying on feedback from a rising MAF signal, rising rpms or a full throttle switch the horse has long since bolted control wise.
A couple of points.
If the throttle is mashed open, I believe acceleration enrichment can be tuned in from the WOT switch flipping from 0 to 1. For this extreme event, the LH is basically tunable to immediately respond to the change in controls. It’s the sudden throttle position changes above idle but below WOT that give the system the most problems.
In an normally aspirated car, the throttle bore never goes sonic regardless of the vacuum generated by the engine at throttle closed because the pressure ratio is not high enough. In a turbo car, on could in principle see that during maximum-acceleration gear changes, but the bypass circuit I believe eliminates it.
I believe that the reason why the S4 system works as well as it does is that (a) there is a large volume between the throttle and the intake valves, which slows down the air flow velocity change and (b) the MAF is close to the throttle blade such that no unnecessary lag is added. This was something to manage with the dual MAF conversion.
I sometimes wonder how much more lift and duration one could grind on the S4 cams while still getting them to operate reliably. The tight LSA might make them just right for a S4 engine with (modified) stock intake manifold and headers and low restriction exhaust. Shark tuner would allow a little higher idle yet the peak power would still stay at slightly above 6000 rpm where the stock intake manifold can still work well. The torque might go up a lot.
Bonjour,
Thank you all for these interesting information.
I tried to summarize data I understand from Ake graphics and 928CS in this thread : https://rennlist.com/forums/928-foru...gt-cams-3.html
I hope I did it right... GT looks to breath much better :
My 928 had an engine replaced years ago and I'm finding out it was not replaced with same MY and type as I was told at purchase. So my question is can GT cams be installed in an S4 engine. I only ask because my GT appears to have a S4 engine (M28/42 81H05548) but certainly performs like a GT (305 rear wh HP - dyno'd). It pulls strong and idles unlike S4's. BTW what is the H in the serial number? That is what year is this engine?
The answer is definetely YES. I installed GT cams when I was rebuilding my S4 auto engine but replaced both LH and EZK boxes with GT specs. The difference was just unbelievable. Not sure why Porsche did not deliver the S4s with these cams in the first place...
I have a fresh set of GT cams and was wondering if it could help the Auto engine since it’s lower revved installation as compared to a 5 speed
Do you have any more info on this install you did .
For example did you install an X pipe .
That’s a clearly noticeable increase in HP for auto or 5 speed
I have a fresh set of GT cams and was wondering if it could help the Auto engine since it’s lower revved installation as compared to a 5 speed
Do you have any more info on this install you did .
For example did you install an X pipe .
That’s a clearly noticeable increase in HP for auto or 5 speed
Roger added GT cams to his GTS and loves them.
Last edited by Kevin in Atlanta; 04-22-2023 at 11:29 PM.
I have a fresh set of GT cams and was wondering if it could help the Auto engine since it’s lower revved installation as compared to a 5 speed
Do you have any more info on this install you did .
For example did you install an X pipe .
That’s a clearly noticeable increase in HP for auto or 5 speed
Yes the GT cams have transformed the engine and is definitely a must even with the auto box.
My car being a non-catalyst from the factory I have not seen a clearly noticeable increase in HP from the X-pipe only. Not sure what info you need on the install but I am running a 928Motorsports X pipe + 928Motorsports manifold spacers + GT exhaust mid and rear sections + GT specced EZK and LH by JDS + JDS modified EZK chip to rev even better which gives a pleasantly awaken engine; its eagerness to rev overcomes the drawback of having only 4 speeds. Acceleration is much much better although not a car for 0-100 these days but the true benefit since the mods is top speed which for what is worth is >279km/h on the digital counter on a free autobahn...