Finally got to build one myself.

[Neil Harvey]

It's really nice to get involved in an engine that is not complicated and easy to build.

We supplied a customer with all of the parts and machining required and now we are doing the assembly. This was not planned, but we are helping the shop building the car by doing the assembly work. This is finally a fun build.

This is a 2.9L build on a 73 911T/S base engine. It will be interesting how this engine performs.

Mag case, 70.40mm stroke stock crankshaft, the rest our stuff. The crank went through all of the typical upgrades so nothing new there.

The case was narrowed and line bored by a well know shop but it had issues after 2 attempts, so we got involved and line bored it 1st O/S and straight. I'm not a big fan of the narrowing cases.

This build is a follow on from the 2.8L engine we did for Jeff in Australia. It uses our long lightweight steel top guided rods, Mahle bridged pistons with small pins and short compression height. The goal here was to lighten up the masses that move. Its fitted with one of our Crankshaft dampers to help with the secondary imbalance these engine have. The longer rod helps the gas exchange over the top but as the masses involved have a new amount of acceleration to mid stroke, the forces are increased. You don't get anything for free with engines.

Heads are not to radical and the engines displacement and smaller chamber limits valve size and port size. This is a street engine so response is very important. Would be a waste of all the lightening of the masses, if we made the cylinder head flow all wrong.

Cam is one of ours and we are using the new billet adjustable rocker arms.

So far the engine is going together one side at a time, as we are checking a few things. This came to use partially assembled so we are doing our due diligence to make sure all is correct.

I will post more photos tomorrow of the progress and of the parts we are using.

[Neil Harvey]

We designed a camshaft profile just for this size engine but it will be a option for many other engines of similar size. The lobe profile is A symmetrical which helps with the limited air volume these head achieve.

To design a camshaft, we first run simulation studies which give us some idea of the airflow required etc. We have spent a lot of time and effort getting the correlation between theory and actual, correct. This always includes flowing the complete intake system as it would run in the car.
We then take these numbers and use them to get the desired % between the Intake and Exhaust sides of the head where we want them.

These early heads don't have a lot of room for larger valves or port sizes, but as the displacements are typically smaller, we can get away with lower flow numbers.

We have a really good library of new cam lobe designs for these older air cooled engines now. From 2.5L up to 4.1L both air cooled and Turbo. The 2.9L is just one of the new designs we have.

Matched with the new cam we are using the new billet adjustable rocker arms. These have a longer pad length to allow for more cam profile to be used.

From the photo's you can see the valve train is pretty conventional. The cam housing is 4 journal unit matched to the early heads. This requires the valve lengths to be slightly different due to the cam design and the rocker arms used. Nothing difficult, just have to look at the contact patch on the engine of the valve throughout the valve motions. This sort of thing is typical when involved with any engine using custom parts.

The valve springs are OE but the pressures are a lot less than stock as our cam design allows for less forces to be used. Steel retainers are used over Titanium as I'm not a big fan of Titanium in street engines, unless the valve train is inspected regularly. Tool steel is a better choice here, but we had to make a small compromise to fit the budget. Brand new retainers or at the very least an idea of time and crack check used ones is always a good idea.

As I get into this build further I will post more photos and description of the work done. I'm really enjoying myself here as this engine is so much less complicated but it still requires attention to detail and focus.

[Neil Harvey]

[Neil Harvey]

The fun continues.

[Hi_Fi_Guy]

Pure art.

[Helix8]

Nice work Neil and following your progress

(rockers are sweet)

[Black 993]

Beautiful.

[Neil Harvey]

Thanks to all those that have given nice comments.

I don't get to build any of these early engines much anymore, so I'm having blast. I'm having so much fun. I should have more progress to show soon. I don't get to spend all the time I have on this, as I have other stuff that has to be done, but when I can I do.

That said, I have a 930 engine we are about to start on too. It includes all of our own parts, so I'll take some shots of them and post those between this engine progress. This engine uses a modified stock crankshaft, our Head studs, Rods, Pistons, Heads, Valve train, Cams, Rocker arms and the new Titanium case through bolts.

['78 SC]

Neil, when you have a moment, please comment on the knock sensor setup. Is that a 993 bridge? Bosch sensor? How did you mount it.

Beautiful build.

[Neil Harvey]

Originally Posted by '78 SC

Neil, when you have a moment, please comment on the knock sensor setup. Is that a 993 bridge? Bosch sensor? How did you mount it.

Beautiful build.

Yes, it is the 993 bridge. We choose a wide band sensor that is a better choice for these earlier engines. These earlier engines seem to have a different signature than the later engines. It has something to do with the masses involved.

We fit both cylinder banks with knock sensors along with Temp sensors. The temp sensor can be seen in the photos on the # 3 cylinder. From memory we fit to the #3 and #6 cylinders. Once I have the other side assembled proper, the sensor position can be seen.

I'm re checking some flow numbers with the middle cylinder, with the Intake attached again, as we have found that the middle cylinders seem to lose a little air flow in the car. Not really sure what is going on, so we are trying to replicate as best we can on the bench. The middle cylinders always have less pressure when measured and in an effort to get all cylinders to make the same or near same torque, airflow is critical. We found this on our 4.1L development and want to see if its the same for this engine and all.

[Walt Fricke]

Neil
I'd never given a thought to the steel valve spring retainers until I found this on my '82 stock (but used exclusively these days on the track) SC 3.0:




Is this common on a ~30 year old motor? I'm sort of kicking myself for just dipping into the parts drawer for a "good used" retainer, visually inspecting the others, and not replacing all of them.

[Neil Harvey]

I think these are some of those parts that when designed were never thought they would be still in use some 50 years later.

Yes, they are problematic. They do crack and often completely fail. I have seen retainers where the contact with the spring has worn a huge groove in the retainer. This is caused by the loose fit to the spring.

We crack check these every time, but that doesn't stop fatigue failures from happening. Add to this, the high spring pressures often used. Most of the time the high installed pressures are required to dampen the harmonics caused by poor cam designs.

I'm also not fan of Titanium retainers used in street engines. These need regular attention and checking. Some are coated and have good fit to the spring, but most are un coated and have a loose fit. We made tool steel retainers in the past but have decided to redesign then to fit the new spring we are about to sell. The stock springs are a good spring made from good German steel, but sometimes you need different rates and distances when using other than stock cam lift.

[gled49]

I thought stock retainers were powdered metal. 906 steel retainers are a little wimpy and you can see a little pull thru, depth of the keepers. 935 steel retainers are very robust.

[Peter M]

Originally Posted by Neil Harvey

We fit both cylinder banks with knock sensors along with Temp sensors. The temp sensor can be seen in the photos on the # 3 cylinder. From memory we fit to the #3 and #6 cylinders.

Hi Neil,
Just interested in why you use a temp sensor on each bank?

Does that mean you use separate temp/fueling tables for each cylinder bank as well?

Or do you use it as an engine protection measure?

Thanks
Peter

[jpnovak]

That is a knock sensor bridge for the EFI.

One bridge per bank of cylinders. With proper cam phasing and crank angle information you can pinpoint which cylinder has knock.

[Neil Harvey]

Originally Posted by Peter M

Hi Neil,
Just interested in why you use a temp sensor on each bank?

Does that mean you use separate temp/fueling tables for each cylinder bank as well?

Or do you use it as an engine protection measure?

Thanks
Peter

Hi Peter,

I guess for protection. It can be an input for more fuel, less Ignition timing or put the engine into a limp mode. Multiple strategies' can be used here.

This engine uses Oil pressure, Oil temp, Crank case pressure, Fuel pressure and fuel temp sensors along with IAT and the two cyl temps and knock sensors. Lots of info that can be used in running the engine or to collect data only.

The one cylinder head missing is having the pressure sensor fitted for dyno purposes. From our testing to date cylinder 5 seems to be the lowest always. Not sure why yet, so we are looking to see why.

It was my idea to do this so I can understand these earlier engines and their habits. This way we can learn and use this info for developing future products.

[Peter M]

Originally Posted by jpnovak

That is a knock sensor bridge for the EFI.

One bridge per bank of cylinders. With proper cam phasing and crank angle information you can pinpoint which cylinder has knock.

Thanks Jamie, I'm across the use of knock bridges. He's my M130 equipped engine:



Just never considered using temp sensors on two cylinder heads or a crankcase pressure sensor for that matter.

[silverc4s]

Subscribed. Always interested in new work

[ian c2]

Originally Posted by Neil Harvey

Yes, it is the 993 bridge. We choose a wide band sensor that is a better choice for these earlier engines. These earlier engines seem to have a different signature than the later engines. It has something to do with the masses involved.

We fit both cylinder banks with knock sensors along with Temp sensors. The temp sensor can be seen in the photos on the # 3 cylinder. From memory we fit to the #3 and #6 cylinders. Once I have the other side assembled proper, the sensor position can be seen.

I'm re checking some flow numbers with the middle cylinder, with the Intake attached again, as we have found that the middle cylinders seem to lose a little air flow in the car. Not really sure what is going on, so we are trying to replicate as best we can on the bench. The middle cylinders always have less pressure when measured and in an effort to get all cylinders to make the same or near same torque, airflow is critical. We found this on our 4.1L development and want to see if its the same for this engine and all.

Hey Niel
Is this temperature affecting pressure once the engine is running , or a mechanical thing that you are seeing on the bench as the engine is assembled ?

[Neil Harvey]

Originally Posted by ian c2

Hey Niel
Is this temperature affecting pressure once the engine is running , or a mechanical thing that you are seeing on the bench as the engine is assembled ?

The easiest (actually the hardest) performance to find is trying to equalize the performance from each cylinder. In these engines we have calculated at least 10 -15 HP difference in all of these cylinders. That is collective. Seems like an easy upgrade.

Temperature has an effect for sure as does air flow through the manifold. Other friction factors play too. Understanding where and what is causing this is a never ending goal. Seems it all changes from model engine to another as well. It would be way too simple if they were all the same.

I am trying to understand these earlier smaller displacement engines as any added performance would make a huge difference. Measuring cylinder pressures, temperatures amongst other data recorded is a start.

We have developed/tested an application of a different cylinder wall coating on a 997 GT3 engine that has proven to be very successful in lowering the ring friction. I'm hoping this will transfer over to the air cooled cylinders but we first need to understand how we can pull the lower performing cylinders up to the performance levels of the other cylinders. These are very simple engines but still have complex issues I don't fully understand.

You can keep trying ways to add performance to these early engines, but its like "feeding a dead horse". Soon than later you figure out, it wont get up.