2.7 CIS - Higher Compression

[ber911]

I want to bump the compression up on my 2.7 but still retain the CIS. Anybody done this with standard dome JE type pistons?

Not sure what effect this will have compared to the swirl dome design of a CIS piston.

 

Rob

[bumble]

I think you'll find there's not much to be gained - in fact, you could go backwards..  The components (cams, piston shape/compression, port sizes, injectors, ignition etc) are designed to work together.  Change one thing (compression for example) and changes are required elsewhere to realise any benefits.  CIS is practically speaking impossible for anyone other than the factory to change.  There is no greater illustration of this than the Group S Porsches.  All things being equal, the 3L Carreras should easily outpace the 2.7s.  The reason they can't (despite truck loads of cash thrown at the problem) is that CIS is not tunable in the way MFI, EFI or Carburettors are.  When Porsche introduced the '81SC with an extra 20 odd hp, the increase in c/r to 9.8:1 was accompanied by significant changes to the CIS.  The ability to do this is beyond your average home spanner walla.

[Pfitz911]

This is my thinking. I'm planning on getting the bitz efi kit for my 2.7. Then when it comes time to rebuild, I can increase compression etc then. I thought Wayne talks a lot about modifying CIS for performance applications in his book?

[bumble]

Pg 98 - "The CIS system does not lend itself well to modifications in the name of increased horsepower."

Pg 113 re: Top engine picks - Upgraded 3.0L CIS use JE 9.8:1, 964 cams and SSI heat exchangers.  No mention of any CIS mods.  30-50hp increase.  IMHO quite a bit of $/work for a modest gain...

[Pfitz911]

Maybe it wasn't in Wayne's book...thought I read it somewhere...

          Need to buy parts for this project?
Click here to order!
Those of us who own a 911 with CIS fuel injection-that is, every owner of a 911 produced during the decade starting in 1973-suffer, we are told, from the inordinate restrictions of the CIS system, also known as Bosch K-Jetronic injection. Bruce Anderson warns in his canonical Porsche 911 Performance Handbook that "if you are going to make a hot rod of the 2.7 engine, you must replace the CIS injection system with either Weber carburetors or mechanical injection." Regarding the 3.0-liter motors, Anderson states that "The next change [after backdating the exhaust] is to replace the CIS fuel injection with a pair of Weber 40IDA-3C carburetors." Most other assessments from the automotive punditocracy are similar, and CIS injection is referred to generally as yet another drawback to the congenitally defective 2.7-liter motors. In short, the accepted wisdom holds, if you want better performance from your CIS-injected 911, after backdating the exhaust the next step is an obligatory scrapping of the fuel injection in favor of carburetors or, more expensive yet, a customized fuel injection setup.

I thought that this would be my predicament, too, after a cursory investigation into modifying my 1977 Carrera 3.0. I had installed SSI heat exchangers and an early muffler, which are superb additions, but as we learned the first day of economics class, human appetites are insatiable, an observation that applies with great relevancy to the desire of Porsche owners for more power. Badly worn exhaust valve guides forced a rebuild at only 142,000 kilometers-Carrera 3.0's were built with the poor 2.7 valve guides-and so the opportunity to perform some modifications while the engine was apart presented itself.

The drawback to replacing CIS with carburetors is that after doing so, the engine internals are still untouched and the car is only negligibly faster. A set of PMO carburetors-which are extremely well-crafted, and include intake manifolds, linkages, and water shields-costs a hefty $2,600, and that's what you have to spend just to reach the point where the motor itself can be modified. But aside from cost, there are several good reasons not to want to switch to carburetion. When properly sorted, CIS is a smooth and reliable fuel injection system; it starts and runs well in cold weather, gets excellent gas mileage, and, with a pop-off valve, is very reliable. Carburetors, by contrast, run more poorly in inclement weather, can be fussy to set up and maintain, and (the main drawback) get comparably poor mileage. Paying $1,000 for used Webers or $2,600 for new PMO's-a setup that brings you down to 15 mpg and only produces a dozen or so horsepower-is not my idea of a great deal.

After talking to some knowledgeable people, most importantly Steve Weiner at Rennsport Systems in Portland, Oregon, I was surprised to learn that there are a couple of very easy bolt-on modifications one can carry out on the CIS cars that are entirely compatible with the stock injection: regrinding the camshafts to the 964 profile, and adding higher-compression pistons. First, consider the cams.

CIS is famously intolerant of the more radical (high-lift, long-duration) camshaft grinds, which is the main reason it must be discarded to build extremely powerful street and racing motors. Fuel flow in the CIS system is controlled by the rising and falling of a round sensor plate inside a conical housing, or venturi. The more the throttle butterfly valve is opened, the more air gets sucked into the airbox intake and flows up and around the sensor plate, which rises in response to the flowing air. The sensor plate acts as a lever, and the more it is forced upward by the rushing air, the more the piston housed in the fuel distributor falls, metering out precisely calibrated fuel to the fuel injectors. More radical camshaft grinds (such as those in the early 911S) produce more valve overlap than do more mild cams, and thus cause air pulsations to travel up the intake system and batter the sensor plate, interrupting the precise metering of air that allows the precise metering of fuel. "This is called intake reversion," Steve Weiner explained to me, "and is common with long-duration and narrow-lobe center cams. I've seen this in many engines since 1963 when I started [working on Porsches]. You cannot 'see' it except on the engine dyno at certain RPMs. Worst cases create a fog of fuel above the intake stacks."

It would be interesting to know who first discovered that 964 (a.k.a. 911 C2 and C4, '89-'94) camshafts are compatible with CIS injection, but produce about 10 more horsepower and a more peaky power curve. There are two ways to make the 964 setup work: Either acquire a set of actual 964 cams and have the power steering pump drive ground off to make them fit into the 2.7 and 3.0 cam housings, or send you stock CIS cams to Elgin or Web-cam to get reground in the 964 profile (I had mine done through Engine Builders Supply Co. for about $350). The effect of the 964 cams is to move the engine's power curve toward the top end of the RPM range, along with slightly increasing the engine's power. Where stock CIS engines are out of breath by 5,000 RPMs, with 964 cams they pull strongly straight up to redline.

The low-compression pistons of the CIS cars also beg for enhancement, as long as you're willing to buy high-octane gasoline. Most 2.7 motors had 8.5:1 compression (sufficient for 87 octane fuel), including the famed Carrera RS, while the '73 CIS 911T and '74-'75 911's ran only 8:1 compression. All the U.S.-market 911SC's ran 8.5:1 compression until 1980, when they were permanently upgraded to 9.3:1 compression, and the Carrera 3.0's had 8.5:1 compression. Although it may mean removing a serviceable set of pistons, you can get a respectable power increase (especially in torque) by bumping compression up to 9.5:1, or even 9.8:1 if you won't have to pull into West coast gas stations and use their "premium" 91 octane fuel. JE Pistons sells the appropriate piston sets (that include Goetze rings, wrist pins, and clips) for under $1,000, and roughly speaking, you get about a 10 percent power increase per compression point increase. The only caveat about the piston swap is that JE pistons won't work with Alusil cylinders-but those cylinders were rarely used on CIS cars.

The cams can be reground and pistons purchased for a combined total under $1,400, and if your motor is already coming apart for a rebuild, they don't take any extra time to install. The modifications were a big success: Drivability in every respect is unchanged, except that the car is very noticeably more powerful-it's considerably faster than a stock 3.2 Carrera, and there is much more high-end power than before. But as the saying goes, one run on a dyno is worth a thousand opinions. I dyno'd the car on a Dynojet and achieved 203 rear-wheel horsepower and 176 pound-feet of torque-all without doing anything to the CIS. Figuring 15-20 percent drivetrain power loss, this engine produces in the ballpark of 230-240 horsepower. Also, a few months after the dyno run, I discovered that in the federalization process my car had been outfitted with a 2.7 fuel distributor, which surely robbed my engine of at least a few ponies. Nonetheless, considering that the car only weighs a stock 2470 pounds (the 2.7's and SC's are similar), one can imagine how satisfying these modifications are.

[ber911]

I would imagine that the JE pistons mentioned in the article would be a "standard" dome piston and not specifically designed for CIS. 

The latest Group S 3.0 Carrera,s are probably as fast as the 2.7 MFI cars bar one - the green machine, which is in a league of its own.

They do not idle all that good, but then they are race cars.

 

Rob

[bumble]

I think that illustrates the issue of CIS as a performance system.  All things being equal, the 3.0 should have about 10% more power than a 2.7.  It speaks volumes that the 3.0 Carreras are 'probably' as fast as the 2.7s.  One technique I have seen with the race engines is to part fill the dome and CNC a peanut-shaped combustion/squish chamber on the plug side.  For a no expense spared race car it may be an acceptable approach.  A bit extreme for a road car though.

[OBRUT]

The SMSP group S lap record is now held by a C3.0.

However, a C3.0 has a tyre advantage over a C2.7 so we are no longer comparing apples with apples.

[Russ]

stock 2.7 out of breath by 5000rpm? that seems different to the ones i've driven

[bumble]

Thanks Ryan - I'm a wee bit out of date then - it would be very interesting to peer inside some of these engines!.  In any event, it's true to say that CIS is a well performing, highly reliable system for standard or mildly tuned cars, but is a limiting factor for more highly modified engines.

[OBRUT]

Correct Mark.

Only recently have very quick C3.0's emerged in NSW and VIC (both also well driven cars BTW). 

I understand that lots (read LOTS and LOTS and LOTS) of effort has been expended in relation to modifying the fuel head and cam selection.

[ber911]

The motor is in a 1974 911s which has an Sc logbook and would need to retain the CIS in order to run in Group S. Obviously just a fun car that would never be competitive, but if I could get it to do mid 1:55's at the old eastern creek circuit, there would be more than enough cars around to have a bit of fun with.

The dilemma I have is to whether to upgrade the pistons and cams  and be able to run in group S or swap out the motor for a 3.0 which would be inherently more reliable and longer lasting as well as faster, but then only be able to run in regularity, supersprints e.t.c.

 

Rob