Print

Email

Designing and Building a World-Record Beating Porsche V8

The task I had given myself was to see if I could get enough hp out of a Porsche 928 V8 to break one or more world land-speed records.

By Carl Fausett

Page 1 of 2

We would build for the Blown Gas Modified Sports (BGMS) class where the record is currently 231.5 mph. Another record we were after was the Porsche marquee record for the 928 of 205.6 mph.

Our math told us that – because we knew our final drive ratio, tire diameters and engine redline; and if our estimates for vehicle frontal area and coefficient of drag were correct – we would need 750 hp at the tire just to meet the current record. With a 15% drivetrain loss, this meant we would require 900 hp at the engine to net the 750 we needed at the tire. (Remember, the class is Blown Gas Modified Sports cars – there is a “Fuel” class as well and they get to run anything they want in their tanks such as alcohol, nitro methane or whatever.)

Cars in the Gas class must run the event gas provided – everybody gets the same event gas right there at trackside and then your filler neck is sealed by the officials. Using just straight gas, making 900 hp from this small V8 would be a little more challenging. Also, we didn’t have to make 900 hp once, but repeatedly and for a long time. Unlike a dyno pull of 6 seconds or so, we would be at wide-open throttle and full load for almost 2 minutes each run – and have to do that several times to get our Bonneville 200 mph license and (hopefully) establish a new world record. The stresses on the engine during these runs are staggering.

Over the last decade I have carefully modified and built each engine in my race car and have learned the “do’s and don’ts” peculiar to supercharging the Porsche 928 engine. I had already built a 650 hp motor for my race car and it had performed admirably at the Pikes Peak International Hill Climb where it netted us a podium finish in the Open Division and made us the fastest 2WD car in the class.  

But for Bonneville, 650 hp wasn’t going to get it done. Every system in the engine would need attention if we were to make the hp needed and survive long enough to repeat it.  Some of the parts I would need existed, but had to be adapted or modified to the task. Many others didn’t exist at all and would have to be engineered and fabricated from scratch.

Compression Ratio and Displacement
When designing the engine for this event, I had a choice of displacement options and compression ratios. The displacement decision was made easy by looking at the current Bonneville land speed records. The engine size with the lowest mph record was the lowest hanging fruit on the tree, so we felt it was ripe for a change. This meant I would build for the “B” engine class – engines from 6.11L to a maximum of 7.19L. By boring and stroking the 928 it landed squarely in the middle of the “B” engine class at 6.54L.

Compression ratio is another matter. When building for boost there are two schools of thought: build with higher compression and run low boost, or build with lower compression and run higher boost. In theory, both engines can produce about the same horsepower, but I am a member of the group that believes that the lower CR motor will be safer and less difficult to tune.

With lower compression, I have a larger cylinder volume to fill with air/fuel mixture, a safer burn, and (arguably) higher outputs. We decided on a 8.5:1 compression ratio, and about 15 to 18 psi of boost. The custom pistons were made for us with molybdenum coated skirts and ceramic crowns.
 
Doing the Math
After a little engine math we learned that 900 hp would require the capacity to move 320 cfm into each cylinder and 260 cfm out at ambient pressure (1 bar); or an average of 2,560 cfm for the complete V8 engine. Then, if we ran about 2 bar of manifold pressure, we should hit our numbers.

I began at the combustion chamber and worked outward from that point on every component and system all the way to the air filter on the intake side and to the exhaust tips on the other. Every part of every system had to have as much capacity as the next and none less than 2,560 cfm.
 
Cylinder Heads
Usually, porting a set of heads is up to the skill of the tool operator as each intake and exhaust port is hand-milled to open them up. In many cases, this is the only way it can be done. But no matter how expert the operator, making the intake and exhaust ports match from cylinder-to-cylinder is just about impossible. And with that irregularity, variables from cylinder-to-cylinder are being built in that will negatively affect the engine’s smoothness, power and longevity under high loads.

To get more cylinder-to-cylinder consistency, we developed a CNC milling program on a 5-axis mill for optimizing the 928 heads. This digital milling process guarantees not only maximum flow, but repeatability from cylinder-to-cylinder.
 
Valve Sizes and Seats
In concert with our CNC porting job, we upgraded from stock 36 mm to special 39.5 mm intake valves, and also larger 33 mm exhaust valves. We opted for stainless steel as the material for our valves which allowed us to have the same valve weight as stock, even though the valves were larger. Titanium is also available for high-rev engines but we didn’t think we needed it as we would be staying under 7,000 rpm. Special beryllium-copper valve seats were used to improve the thermal conductivity of the seat and to reduce valve bounce at closing.
 
Valve Springs
At 15 pounds of boost, our 65.2 gram intake valve with a head diameter of 1.899 sq.in. will have a virtual weight (physical weight plus air pressure loading) of 28.63 pounds. This meant stronger springs were required to close the intake against the boost pressure. Other challenges also included controlling valve float from cam lobe toss and seat bounce at 7,000 rpm, yet avoid valve spring stack with our high-lift (.442˝) camshafts.

The solution was a custom set of valve springs made from chrome-vanadium wire, shot-peened, heat-treated and stress-relieved. Titanium retainers and clips were found and used to hold them in place.
 
Camshafts
Of course the large valves and porting alone won’t move the right amount of air without the cams to lift them. Unique to the Bonneville application, we needed our max hp to come in at 7,000 rpm where we will have our greatest aerodynamic resistance. A combination of special cams and intake runner lengths were used to tune the power band to make this happen.

Here we have a certain disadvantage compared to our friends running pushrod motors. Because a pushrod motor can use a 1.4:1, 1.5:1 or even 1.6:1 rocker arm ratio, they can throw their valves open further and faster than a OHC, OHV setup like that in the 928 engine. I had a special cam grind made with absolute maximum lift that would still fit within the 32v Porsche 928 heads.
 
Intake Runner Development
Owing to the size and type of injectors available to Porsche when the original manifold was designed, a fairly large intrusion into the intake runner exists to make room for the old Generation II injectors. The advent of the modern thin-body Generation III injector allows us to reduce this intrusion into the runner and permits more air to travel though.

In addition, we canted the runner away from the injector slightly, which had several benefits. It further reduced the intrusion of the injector into the runner wall, it smoothed the transition from the runner into the head, opened the radius of the turn, and guaranteed proper aim of the atomized fuel plume at the back of the valves.

There are two methods of “natural supercharging,” that is, increasing the velocity of the air at the back of the intake valve. The first is the Helmholtz Effect, (using proper inlet bells to reverse the pressure wave in combination with intake runner length to time the pulse so it arrives at the back of the valve just as the valve opens); and the second is runner taper. Even though we would be boosted, these fluid dynamics still apply and needed to be factored in to achieve maximum engine output.

The inner diameter of the runner at the head was set once the head was fully ported.

Working up from the finished ported head, we established the diameter and taper desired in the intake runner, the length of the runner, the injector angle and fitment, and the mounting surface for the plenum.

Computer Aided Design (CAD) of these intake runners were a must. Here I had the assistance of an excellent Solid Works designer named Ryan Silva. CAD modeling the intake runners gave us the opportunity to try many iterations, and make changes to transitions and radiuses rapidly. We “hit our marks” on the third manufacturing model, and went right to pressure and thermal testing the runners. Changes and adjustments to fitment were made via Rapid Prototyping and fed back into the computer model.

The material we chose was SLS Glass-filled nylon for the intake because of its many fine properties and low single-unit cost. But the finished intake needed testing under heat and pressure to make sure they would not fail in the field. The runners were bolted to 928 heads and heated to 200 degrees F, then put through 8,640 pressure cycles (0 to 40 psi to 0 = 1 cycle) without failure, and with only .002˝ deflection. They would hold our 20 psi without a problem, and more if we needed it. (For more information on the nylon intake manifold, see sidebar on next page)