How we go about designing a camshaft.

[Neil Harvey]

This is how we go about using the data we collect from flowing cylinder heads and including the engine component spec's, when starting to design a camshaft. There is a lot more to, but this may give some insight into what is involved. Its how the air numbers are used.

As an engine company that sells performance, we must know the cylinder head’s ability to flow air. Flowing Cylinder Head Intake and Exhaust ports give us the insight of what we can expect in performance. But is that adage, garbage in is garbage out. We also must know the engine use. Is it going to be a race engine, street, Turbo or NA. It all is very important.

The flow bench needs to be calibrated, the right engine simulated parts used, and the data then used correctly. Was the same cylinder bore used, any leaks considered in the final CFM number and most importantly if tested at another pressure number, the results recalibrated to whatever the bench was calibrated to. Our bench was calibrated to 25” of water so flowing a head at 28”, the numbers must be reconfigured.

But most importantly, the complete Intake system needs to be flowed. You will always see a drop in the flow numbers. Using just the head numbers in any decision is not correct.

Flow numbers give a good idea of what performance can be expected but these are not the only important data needed. We measure the air speed through the port as well. With pivot tubes, we can measure where the air is faster, where there are any choke points and how the short side effects the air speed. This is important if any port work is done or going to be done. Air speed is the more important than total air flow as well. Turbo engines can have air speed “built” in, but still can have less performance if the air speed is slowed up any.

There are other factors as well to be considered. Rod length plays a part here. If a longer rod is used, it slows down over TDC and pulls slower down the bore compared to a shorter rod. A shorter rod will pull more vacuum per travel down the bore until it reaches approx. 90° of crank angle. But this will show up any Intake system inefficiencies.

Valve sizes, their ratio to each other, the engines bore all play a part here. The back side shape of the valves is important. The guide boss and how it separates the air coming into the port is important. The most important part of any port is the valve seat, and the part of the port just above it. The throat and its taper play a huge role in the speed and air flow. Poorly repaired seats can kill both the flow and speed. By a lot. Its scary just how bad a good port is killed by a seat sunk into the head too far and unblended steps between the chamber “roof” and the top seat angle. I have had heads sent here to have the sealing system fitted that have just been reconditioned. Its less work and a lot less time (money) to just keep cutting. These Porsche air cooled heads are prone to have guide issues. Soft cheap guide material is commonly used, accelerating the wear. The guides are changed and never concentric to the original seat, requiring the seats to be recut. This always requires the seat depth to go lower to obtain a full 45°, or whatever sealing angle is used.

As I stated in the beginning, as a company that sells performance, we must approach it this way. For the DIY at home, you have the choice to follow the professional approach or buy off the shelf and hope for the best. I can tell you the flow and speed loss through a mid-80’ 911 and 930 Intake manifold as we have flowed those, but I cannot tell you what the loss might be through your head over the seat. If you are choosing a cam expecting the air numbers to be something, then you should expect the performance you end up with, as well. Good or bad.

Now that we have collected these air flow and speed numbers, how do we use them in a cam design?

We combine the airflow numbers with the valve motion. A lot of calculus is involved. You can tell a good cam design by who did the work and if they have a handle on basic calculus.

The final lift numbers are based on the airflow at the measured valve lift. If the airflow stalls at 0.500” valve lift, its pointless to build cam with 0.600” lift. Also, piston clearance must be considered here as well, as this effects the possible separation angle. 911 2V cams have fixed separation angles.

The duration is the valve motion in inches (or mm) verses crank angle. Crank angle is the piston position in the bore. We can then plot the CFM on a vertical axis verses the crank angle on the horizontal axis. We get a time verses area plot or time verses area analysis. This was actually developed by a Japanese engineer back in the 60’s.

CFM verses lift tells you nothing. The average flow is a big deal. It’s the area under the curve, the valve motion verses the crank angle or piston position that is huge.

We have tested the 2V 911 valve train on the Sprinton and know what works and what doesn’t. We know the limits here. We know how important valve sealing is. If the valve bounces off the seat it kills the performance. We get a true understanding of the valve train dynamics, spring rates etc.

Valve seal on the seat, ring seal in the bore all make or loose performance. This is one reason why rods are often guided by the piston. It helps keep the piston in the bore straighter. Allows thinner rings to be used, lowering the biggest friction adder.

We took the position that we needed to offer a whole new line of camshaft designs. Most of what is offered today in the aftermarket are copies, bad ones at best, cut and pasted from old camshafts designs. You can always fit an older camshaft design into your engine to get a base line, but nothing will beat a cam that has modern technology built in, designed for your engine type.

Our designs are A symmetrical, the opening and closing of the valve is different. We accelerated the valve off the seat faster than we close it. We slow the valve down and place it back on the seat with far less force. This allows lower spring seat pressures, helping with “turning” the engine. The downside, they cost more to produce. It all must come together as one solution to the performance puzzle.

There is more to this than what I have written. An engine is the sum of all its parts. Some are happy with the performance of their engine as is. Some will be happy with what’s offered now, in the aftermarket. Some may want more.

Remember, you cannot fix a bad head with a camshaft.

[piggdekk]

So Neil I've got a question for You.
I've een looking at cam profiles lately to build a gr.2 2.3 engine and I was quite surprise to see that center lobes even on aggressive cam shafts (300+ duration) are closer to 110 degrees. My experience is with italian and german 4L engines, and in that case performance cam shafts tend to have center lobes close to 100deg. Porsche engines seems to work differently. Am I wrong? Are they always biased more towards torque than HP?
thanks
luca

[Classic 911]

Following


Sent from my iPhone using Tapatalk

[Old H2S]

Good information, lots of variables to juggle.
My question is I am building a stockish 3.2 ss, this has been done to death because it is easy for the home builder not to screw up too bad..
New jugs, pistons, 9.5 CR, heads done my known good guy, no porting, fly cut .010
single plug, CIS, plane Jane numbers expected...We have seen many dyno sheets and they all come in close to each other so what cam for the daily driver do YOU like?
It is easy to go down the rabbit hole and twin plug, up the CR, port port port and CARBS, over cam to make the most it can be at 8000RPM but the price goes up quick.
How fast do you want to go?

[Brookieslayer]

This is all pertinent stuff.
I like the cam / valve speed to the seat discussion. It does not get brought up very often, but it is very important.

Thanks for the insights.

[Jeff Alton]

Great Post Neil,

Our next build with needs for performance cams, I will be contacting you to discuss all of this to see if you can help us optimize the build.

Cheers

[Neil Harvey]

Wasn't trying to sell anything. Was just answering a question regarding how and why we use air numbers when designing a cam shaft.

If anyone has questions they can PM me and I am happy to answer anything required.

[Black 993]

Great post, Neil. I love this stuff. Many thanks for the contribution.

[Rosco_NZ]

Interested to know Neil whether Jeff’s EFI 2.8 came out hp wise where you thought it would be .. what you designed the cam for ..??

[Neil Harvey]

Originally Posted by Rosco_NZ

Interested to know Neil whether Jeff’s EFI 2.8 came out hp wise where you thought it would be .. what you designed the cam for ..??

Not really sure about final HP numbers. Jeff was having issues with Exhaust and was remapping certain areas of the map as well.

I did talk by email last week and he was very happy with the performance. Better to contact him directly.