996 3.4 Liter = 300hp, Why Can't More HP be Squeezed Out of 3.2 Liter Carrera?

[RandalJ]

Just curious about this. I have read numerous times that Porsche extracted about as much power and torque as possible out of the 3.2, taking into account the need for drivability and fuel economy. So, don't bother trying to extract more power with anything other than forced induction.

So, why is it that the early 996 engine, which has only a smidgen more displacement, about the same fuel economy, and arguably better drivability has so much more power and torque?

I'm wondering whether it is because the 3.4 of the 996 can be run at a higher "stress level" because it is water cooled. That if it were air-cooled, it would have the same limitations as the 3.2.

Or was the first statement pertaining only to bolt-on engine upgrades? Could the 3.2 with different cams, crank, etc, be made almost "3.4 like" in character?

[kojo911]

I would think it's a combination of: Compression ratio (higher in water cooled), RPM (twin cam, four valve), Engine management. I think all of those parameters can be stretch due in part to better cooling.

[Steve@Rennsport]

Its not difficult to get good power at all from a 3.2; its simply not cheap to do and maintain reliability.

Smog compliance also plays a role here so this becomes a limiting factor when that's a criteria.

Like many things,......it just takes money.

[RandalJ]

Thanks for your responses. Tighter emissions standards over time makes sense, and I think I read that one of the reasons why Porsche went to water-cooled was that it was going to be cost prohibitive to carry on with air-cooled, meet emissions, and still up the ante on performance.

[johndaroza]

From what I have seen a 3.2 can make 425hp . Better question is how much money do you have? 425hp= $30,000

When they went to water the head stayed a lot cooler, the engine is not the same and it is easier to make more power

[kenikh]

It's a lot easier to make power with water cooling and twin cams. Cooler engine= higher potential CR. DOHC = RPM adjustable overlap. The list is longer, but these two are the biggest reasons.

[Matt Monson]

I think a lot of it has to do with 4 valves versus two. It is this same reason that to get good power out of a 3.2l you have to go twin plug. The two valve heads really restrict the piston shape.

[kenikh]

Originally Posted by Matt Monson

I think a lot of it has to do with 4 valves versus two. It is this same reason that to get good power out of a 3.2l you have to go twin plug. The two valve heads really restrict the piston shape.

Two things: flatter head dome and central plug location really help flame propagation at ignition, meaning you can run higher CR.

On top of the other things I mentioned, four valves also means you can stagger valve opening to optimize swirl into the cylinder, also improving combustion by optimizing charge velocity and scavenging.

Four valve heads were experimented upon by the factory, but abandoned them because an air cooled motor just didn't have enough surface area with four valves to adequately cool the heads.

[Walt Fricke]

Water cooling allowed 4 valves per head, which greatly increased volumetric efficiency. I think twin plugging may resolve dome shape issues. As to staggered timing for swirl, I hadn't heard of that, but sounds good, as that will increase VE.

Here are some reasonably attainable horsepower figures for 911 race motors:

Air cooled: 110/liter (this is what the factory asserted)
Water cooled GT3 (Metzger type) motors with single throttle body: 165/L
Same with individual TBs: 175/L
M96: 135/L
M97: 140/L

I'm not saying motors like this are inexpensive. And so far I haven't been able to hit the 110 mark with my home built motors. But the good shop guys can reach these targets (and more depending on projected engine hours and the cost of 112 octane fuel).

So the change to water cooling and 4 valves made a huge difference.

[kenikh]

You won't crack 110HP/L without custom designing and tweaking camshafts on a per cylinder basis. This requires a Spintron to measure, then equalize bmep across all cylinders by tweaking cam lobe specs. Once set, you CNC grind cams based on the data analysis. This is how NASCAR achieves crazy power figures, especially with restrictor plates.

[mamut]

Simple, Better Volumetric Efficiency across the board.

[Steve@Rennsport]

Originally Posted by kenikh

You won't crack 110HP/L without custom designing and tweaking camshafts on a per cylinder basis. This requires a Spintron to measure, then equalize bmep across all cylinders by tweaking cam lobe specs. Once set, you CNC grind cams based on the data analysis. This is how NASCAR achieves crazy power figures, especially with restrictor plates.

LOL,..You won't crack 110HP/L with just cams and compression.

For that objective, you need a really good intake system, special heads, special valves, custom pistons, long rods, a really good exhaust, etc. etc. etc.

[kenikh]

^ All of the above were inferred.

[HawgRyder]

Think of the whole exercise like a teeter-totter .... when power goes up...life expectancy goes down.
You choice of course...but be prepared for a shock when screaming down the road....LOL.
Bob