"Rev-ability" of CIS vs. EFI
 

I recently swapped my "20 footer" 81 Euro SC with my dad's "garage queen" 88 Targa for about 6 weeks. Over the course of the 6 weeks that I had his Carrera, one of the things that bugged me most, was how slow his engine revved up. In my SC, I can stab the throttle for a fraction of a second, and it will rev to redline immediately. On the Carrera, I would have to hold the throttle down for more than a second, and it was noticeably slower to rev to redline. This was especially annoying when trying to rev-match while downshifting. With my SC I can tap the throttle with my heel and get the revs I need, with the Carrera I would have to hold it down to get the engine to rev up enough. Why is this? Is there something wrong with the Carrera, or is that just the way EFI makes it behave. I think a sports car should have a fast acting throttle for rev matching. I know with my Audi A4 1.8t, it is also very slow to rev up, but that could just be the nature of that car and not the EFI.

Any insight would be helpful. My dad is talking about getting rid of his Carrera, and I could have the opportunity to trade up. But I don't think I could stand to drive it after being used to my SC.

It needs a Steve Wong chip. It will wake up nicely. Hard to compare a euro to a domestic 3.2, but you can get it better.

The throttle response on a bone stock Carrera is indeed slow. It's even more so on a G50 Carrera because of the hefty clutch assembly, when compared to a 915 trans. car. Do a search and you'll find numerous ways to wake the car up. The first start is typically the E-PROM chip, which really does help. A freer-flowing muffler and/or premuffler cat replacement really helps as well. The combination of the two on my car made a substantial difference.

Don't overlook maintenance issues like loose intake manifold nuts, loose throttle body elbow or airbox connection, vaccuum connections for the fuel pressure regulator and damper, sticky MAF swinging door,...... check to make everything is in working order first.

My engine is one step further with a lightweight aluminum clutch from Kennedy Engineered Products (KEP). This and the Dougherty Racing DC20 cams seemed to really help the throttle response even more. Blipping the throttle is just that now- a blip and not having to hold the throttle to bring up the revs. A chip modification is a must for this to avoid stalling problems when the revs drop back to idle map.

I'd not recommend the KEP clutch because the pedal effort is high, the cost is high, and there's high potential of breaking the clutch release shaft pivot points on the trans. bellhousing due to the added spring clamping force of the clutch housing. I plan to remove my clutch this winter and instead have my stock clutch lightened by a specialty shop- if I can find one.

Something is likely wrong. One second is a long time to rev to redline.
How is the power? It should be slightly better than your SC, too. If not, something is wrong. Start with the basics compression, vacuum leaks, tune up, etc. I would try a chip after all of the basics have been covered and double checked.

Doug

The Audi 1.8t? any turbo has some lag, so it's not really comparable.

The SC has much better throttle response than the Carrera. The chip will help, other stuff, slightly.

You may have a leak somewhere, but, SC will always be better.

The 1.8t suffers from a poorly mapped drive by wire throttle. Same as the 2.0T (though not as bad)

CIS is lazy, so it a motronic 3.2 or 3.6. EFI with ITB's is not even comparable. It is like carbs or MFI-- instant throttle response.

Cheers

If you don't mind getting your hands dirty, you could always tune out the laziness of the 3.2 as I have done and tune in better throttle response than CIS with different injectors.

I have experimented with many different saturated injectors with different flow rates and have come to the conclusion that the stock injectors are (believe it or not) too big for good throttle response and, in reality, tend to bog performance throughout the rpm range. So far the injectors that work best and really wake up my 86 3.2 are GM Multecs that are found on Chevy 3.1 V6 engines. The have a superior spray pattern than the stock pintel type Bosch injectors, so they idle much smoother and have more than enough flow on the top end to convince my seat dyno.

http://forums.pelicanparts.com/uploa...1194546203.jpg

http://forums.pelicanparts.com/uploa...1194546258.jpg


Which also happens to be the same injector design with a four hole director plate that is currently being used on the 996 M96/03 engine.


The only other modification needed to make the saturated injector work on the 3.2 engine is a start relay for additional fuel for warm starts.

Cheers,

Jose'
74 911 w/86 3.2

That was my first thought. Although I changed the oil, plugs and adjusted the valves while I had it. I spent enough time in the engine compartment to inspect everything and it all looked tip top.

The thing about the power is, I have a euro SC, so I'm only down about 10hp compared to the Carrera, and my euro SC w/o sunroof is significantly lighter than the Targa. The power of the Carrera feels stronger than the SC above 4500rpms, but below that it's a dog. It really makes for annoying city driving. I found it really irritating to not be able to pull ahead of a minivan unless I rung it out, and that is so obnoxious when driving in town. I will have to see if I can get him to add a chip before he parts with it and see how much that helps. I think it will be a year or so before he decides to part with the car, so I've got some time to play with it. I definitely think it can be woken up a lot and be made as fast or faster than my SC. I just find it hard to believe that an archaic CIS system is more pleasant to drive around town than a 3.2 with EFI.

I had a slight throttle response issue at one time. It was something that was a new problem though. I found the little plastic bushings had deteriorated over the years, and for the most part, they had finally crumbled away and it was not the same "feel" as before. I replaced all of the little bushings and all was well again. The bad bushings created a little slack in the linkages. The response on my 1984 3.2 - 915/67 trans, does not show the type of delay that you are talking about. The throttle response is immediate, and crisp. You can "blip" the throttle, and it instantly raises the RPM rate. I have seen some engines that do not have mechanical controlled throttles, that seem to take a little longer to respond, like the totally electrically computer controlled systems, they seem to take some time for the engine to actually get to the expected response in throttle movement, this is even more prevalent on the turbo engines, but they still have pretty fast response. On your car, I would check the throttle linkage, from the pedal to the throttle body to check for any type of slack in the entire linkage system. It almost sounds like it has some slack in it, and it is actually having to overcome the spring tension, along with possible slack, then once the slack is out, the throttle can respond. Hopefully I am making this clear about the slack and the spring tension, because it sounds to me like you are having to overcome the spring tension, with the accelerator pedal, and then once the spring tension is overcome and the slack taken out, the throttle will now increase the engine speed. This would give a slight delay in engine response, as to accelerator pedal movement. Have you actually checked to see what kind of response is made when you actuate the lever by hand at the throttle body? I don't have much seat time in a G50 car to really know, how much difference there is between the G50 and a 915 type transmission car to know about any throttle delays. Just my experience and $0.02. Good luck!

Unfortunately, to a certain extent, that is the nature of EFI vs. CIS.

Think about what is actually happening when you step on the gas.

In CIS, you open the throttle plate, which causes the sensor plate to rise due to more airflow. The instant that happens, through a direct mechanical linkage, more fuel is injected.

With EFI, when you step on the gas, the airbox flapper valve opens, which sends a signal to the brain, which then consults a variety of other input from various sensors, makes some calculations, and arrives at the correct fuel map. It then adjusts the duty cycle of the injectors to match that theoretical calculation, and then you accelerate.

It's just going to be slower due to the nature of the system. I too found this out when I sold my SC and bought a Carrera. There are things that can be done to mitigate this problem (such as a chip), but my feeling is that a Motronic car will never have the instant snappy throttle response that a CIS car has.

You learn to live with it after awhile. 3.2 Cab makes a good suggestion though: the simplest and easiest thing to check is whether you're actually getting full throttle, so do that first. Good luck with it!

ianc

Mine is a hybrid system using the 3.0 crank with a 3.2 top end minus the barn door AFM. With JE pistons, Sachs clutch, and a very light crank pulley it is very rev happy, much more than the stock CIS.

I am really know nothing about Motronic, but I have built and "programmed" a Megasquirt for another project. The MS uses an older style processor that is capable of keeping up with an engine turning 16,000 rpm. I seriously doubt the issue is with the time it takes to compute and send out signals, and then have them acted out. If Motronic in the 3.2 is in fact really sluggish, I would be much more inclined to believe it is because the air plenum, fuel map, linkages, etc were designed as an overall compromise and not tuned for best response.

On an ITB or 911 carbed engine the throttle response is largely gained by the reduction in air volume that the butterfly is controlling. Ever heard an F1 engine wrap up? Definitely EFI and unbelievably fast... other factors are at play, but the speed of a computer is not one of them.

Of course, these are my beliefs... YMMV

"I seriously doubt the issue is with the time it takes to compute and send out signals..."

- Exactly. Moreover, the air flap is a terrible and slow (poorly responsive) way to measure air flow - a hot wire is incredibly faster.

The DME is programming is the real suspect to me. It is simply not designed to give max. throttle response. Look at what after-market EFI systems can do.

Both CIS and the DME suffer from the lack of individual throttle bodies also. Of course, there is an adv. to a single 'body' also and that is the dynamic tuning or effective intake length used by P AG later on...

I've noticed that on CIS cars, a leaner mixture seems to allow a quicker rev up. Whether this is actually true or not I don't know, but it certainly seems that way.

I've never compared it to an EFI engine.

True enough. That's why the injectors I'm using work so well.

A 10 min ride to Novato might change your mind. The invitation is still open.



Cheers,

Jose'
Rewood PCA since 1976

Carreras stock typically have less throttle response than CIS because not really because of EFI vs CIS, but because the later Carrera cars are programmed to run leaner and cleaner in the part throttle regions and anything below 4000 rpm. Adjust a SC to run as lean as a Carrera, plug in the O2 sensor, and it will also have poor response and power. On the other hand, reprogram a Carrera to run richer under load and acceleration, give it some more timing, and the overall torque and response increases noticeably. You could even go the next step and run open loop just as the Euro Carreras and do a reprogramming, and the car runs like a 'freight train' - as quoted from the U.K. guys on the Impact Bumpers forum. It's effectively the same as when running a SC adjusted richer, with the O2 sensor disconnected.

Then how can you explain how later 911s like 993s and 996's which must run even leaner programs at all rpms for even tighter emissions and economy can have such good throttle response?

And if our poor throttle response was due to such lean programs, then why would we be having so many failed smog check issues?

I wouldn't characterize the throttle response of stock 964s, 993s, and 996s as anything fantastic, however, they do make more power for the following reasons: increased displacement, more advanced ignition timing, twin plug ignitions (964s and 993s), dual stage resonance intake systems, improved header exhaust systems (993s and 996s), Variocam (996s) and 4 valve per cylinder heads (996s). It is also a misconception that the later cars run leaner, on the contrary it is the opposite. Analyzing them under various loads, you can see where the factory increased fuel delivery under acceleration and full throttle to increase throttle response and power, at least to the extent where they can still meet emissions testing. That's not to say, that with further fuel adjustments in select areas, the later cars' power and throttle response couldn't be improved further.

As an example, illustrated below are the dyno charts of two cars, a completely stock 3.2 Carrera with a premuffler (red curves) and a completely stock 996 3.4 (blue curves). The air fuel chart below shows how lean a 3.2 runs (~14.0:1 @3500 rpm), vs the 996 (12.4:1 @3500 rpm). Their more efficient and larger 400 cell metal cat systems allow the later cars to produce lower emissions.

http://www.911chips.com/dyno996.gif

"With EFI, when you step on the gas, the airbox flapper valve opens, which sends a signal to the brain, which then consults a variety of other input from various sensors, makes some calculations, and arrives at the correct fuel map. It then adjusts the duty cycle of the injectors to match that theoretical calculation, and then you accelerate."

ianc

Rather naive statement as usual!

"Then how can you explain how later 911s like 993s and 996's which must run even leaner programs at all rpms for even tighter emissions and economy can have such good throttle response?"

stirj

Basically the only logical post of the thread. Then we have those who use the thread with
their usual hyperbole to promote performance chips as the answer.