An Alfa Romeo 8C Project

A post about classic cars! Travelingwithtools has been focused on the travel part of late, so I thought it was time for a tools post.

This Alfa Romeo 8C 2300 came to me in 2008 with a badly leaking rear main seal. As I recall, it was a very fresh restoration but there had been some sort of assembly issue.

The normal procedure at work was to thoroughly document the car with photographs before disassembly, so I have a stack of digital images to choose from.

  • Classic pre-war Alfa front end.
  • 8C 2300 rear view, sans axle.
  • 8C 2300 axle assembly with torque tube.
  • 8C 2300 interior.
  • 8C 2300 gearbox and bellhousing.
  • 8C 2300 multi-plate clutch.
  • Running the 8C sans clutch.

After removing the interior to get access, and removing the gearbox (which requires removing the rear axle and torque-tube entirely), there was a surprise: The back face of the crankcases had been shimmed with layers of paper to produce a taper, in order to square the gearbox to the crankshaft centerline. Running an indicator around the bellhousing flange while attached to the clutch showed that it was indeed not square to the crank, and the paper shim corrected it well enough.

You might think that this sort of bodge on a million dollar car would be totally unacceptable, but I can see the scenario that probably played out. The issue was unrecognized until after the engine was completed and installed in the car, while attempting to fit the gearbox. Fixing it properly would mean, at a minimum, removing the engine and setting it up whole on a milling machine, to cut a new face on the crankcases. Or it could be completely disassembled, the cases machined, and then reassembled. I imagine there was a deadline, a debut event to visit, and there simply was not time. The flange should have been checked for square on the line boring machine, back when the babbitt main bearings were being machined.

The leaking oil may or may not have been related to this, but that was yet to be determined. We ran the motor with the gearbox off to visually confirm where the leak was coming from before proceeding.

  • A different 8C crankcase on assembly.
  • Original-style 8C scroll seal, coated with Redline assembly lube,

I pulled off the flywheel and removed the sump, which gives access to the crankshaft and main caps. The rear main journal on an 8C originally had what is called a scroll seal… there is a diameter behind the rear main journal that is larger than the bearing and has a spiral groove machined into it. That lives in a housing with a close tolerance, so that any oil trying to escape through the gap is pumped by the spiral scroll forwards back into the sump. It is a simple, all-metal device that works wonderfully if the tolerances are correct. However, when the engine is on the way to worn out, the crank can move enough in the main bearings to wipe the scroll seal surfaces.

This motor had been converted to a modern lip seal… the scroll on the crank replaced with a smooth seal surface, and a two-piece rubber seal. One half of the seal had become displaced, so there was a large gap for the oil to get past.

Lip seal conversion
The upper half of the lip seal (the reddish brown rubber nearest the flywheel flange) with it’s L-shaped steel spacer .

I can’t say that I remember exactly what the flaw was (it was 13 years ago!), but I do recall that it was obvious, and fairly simple to resolve.

The photos from under the car Illustrate a few interesting things about the 8C. One of the main things that was special about Vittorio Jano’s design was the two-part crankshaft. Previous straight eight engines with one piece cranks had problems with cracking after high RPM stresses introduced strange harmonics. Modern designers would use computational fluid dynamics to figure out where to strengthen the forging, but Janos decided to split the long crank into two pieces, with a spring-loaded damper in the center.

  • 8C 2300 crankshaft center gears.
  • 8C 2300 crankshaft center gears showing the sprung gear hub.
  • An original 8C2300 crankshaft center flange.

The cams were also made in two pieces rather than one long billet, joined at the cam gear in the middle. The gear drive carried up the center between the front and rear cylinder block sections. This center-drive design solved another problem of the long straight eight, which was that the far end of the cam was a long way away from the drive gear when the gear is at one end… as with a Bugatti or Miller.

The 8C 2300s that I’ve worked on all had magnesium crankcases with aluminum main caps. They had two aluminum cylinder blocks with iron liners, and a separate cylinder head that was also made in two four cylinder sections. The later 8C 2900 had a mono-blocks that eliminated the head gasket, the heads cast in-unit with the cylinders. Which makes doing valve work more difficult, but eliminates the possibility of head gasket leaks. Mono-blocks were commonly used by many manufacturers, including Bugatti.

An 8C 2300 engine in process with new blocks in a test fitting. Note the center cavity for the cam drive intermediate gear.

The engine in the picture above was pretty special, in that it was from the car that Enzo Ferrari reputedly last raced in his career as a driver, before turning to manage Scuderia Ferrari. I would love to know the story behind the big diamond-shaped window repair in the upper case. Where was the car racing when the number three connecting rod attempted its escape?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s