CIS FI question for the experts out there

emarsh · 07-05-2003, 06:17 PM

I spent a good part of the day today reading my Bentley manual so as to get a better understanding of the care and feeding of my '79 911 SC. I read the sections about the fuel injection system. Its an elegant pre-electronic FI setup but there is one thing I don't understand. How does the system adjust the F/A ratio for throttle opening?

From what I see as air flow increases it opens the air valve which increases the fuel flow. Only thing is, X volume of air can either be caused by the engine at a higher speed and a very small throttle opening or the engine at a lower speed with a greater throttle opening. In one case it is desirable for the system to be closer to full rich and in the other case a lean mix is desirable for cruising.

Can someone explain to me what is the missing piece of this puzzle?

Thanks,

Eric

beepbeep · 07:38 PM

There is no puzzle. CIS doesn't care about throttle at all, it just cares about air flow.

There is slight "accelleration enrichment" though, done mechanically. When you step on the throttle, more air flows trough the barn-door which briefly overswings and thus provide extra fuel in the begining of accelleration.

Other than that, it just injects as much fuel as air is flowing trough the system.

emarsh · 07-05-2003, 07:28 PM

Quote:

Originally posted by beepbeep
There is no puzzle. CIS doesn't car about throttle at all, it just cares about air flow.

Sounds like the fuel air ratio is rarely optimal in that case. I'm surprised.

beepbeep · 07:39 PM

Well no, it's trying to keep it slightly rich of stochiometric to be on the safe side...so it's quite optimal for most of the time. It's not as fuel-efficient as lean-cruise optimized EFI but it's light years better than carbs in that department.

emarsh · 07-05-2003, 07:39 PM

Quote:

Originally posted by beepbeep
Well no, it's trying to keep it slightly rich of stochiometric to be on the safe side...so it's quite optimal for most of the time. It's not as fuel-efficient as lean EFI cruise but it's light years better than carbs in that department.

How about at full throttle when it counts?

beepbeep · 07-05-2003, 07:47 PM

As i already mentioned, it doesn't care about throttle-position at all. All it does is to measure air-flow mechanically and inject fuel proportionally to that. Full throttle means more air through barn-door, and it just squirts enough fuel. That's it. The shape of the cone is carefully chosen so it coresponds to air/fuel "map"... there is no more to it.

emarsh · 07-05-2003, 07:56 PM

It would be very interesting to see some dyno runs at different throttle openings to see what the F/A curves look like.

I'm putting together a wide band o2 meter for my motorcycle. I may have to plug it into the p-car one day.

Hladun · 08:14 PM

In a CIS system the air/fuel mixture control works independently of throttle opening. All it does is measure the air flow (at atmospheric pressure) and match the required fuel. There are enrichment tricks like the WUR and air plate over swing, but these are second order effects.

So what about the closed throttle/high rpm versus open throttle/low rpm dilemma? What must be considered here is the effect of manifold vacuum. The first case has high manifold vacuum which means the engine is "pumping" low density air. The second case has low manifold vacuum and more dense air. So the result can be that at closed throttle/high rpm and open throttle/low rpm the engine is passing the same mass (weight) of air and hence the CIS unit injects the same amount of fuel.

Remember that a throttle modulated engine works in a "choked" condition most of the time. That is, it has manifold vacuum that limits the torque it can produce as combustion occurs. Looking at it another way, if you ran with wide open throttle the engine would rev up until it matched it’s load (or to distruction if it is unloaded; as some Porsche drivers find).

And you are right, a CIS system (aside from the WUR and over swing) isn’t as good for acceleration as a carburetor accelerator pump that throws in a bunch of raw gas. Max power is achieved at around 11 to 1 air/fuel which the CIS unit will probably not achieve if it’s set at Porsche recommendations. If your emission controls are removed you can set the CIS richer for more power.

Superman · 07-05-2003, 11:35 PM

True, the amount of gas precisely matches the amount of air. No problem there. But CIS does have a "throttle response" problem that is due to the elasticity of air. So, it does not work as well in transition. CIS systems do not provide the seat-of-the-pants torque during throttle transition because the air in the air box is elastic. On the other hand, CIS does just fine on dyno runs because dynos measure maximum power the engine can produce at a given rpm and throttle position. That gives the system enough time to quit transitioning and reach equilibrium.

emarsh · 07-06-2003, 06:18 AM

Quote:

Originally posted by Hladun
And you are right, a CIS system (aside from the WUR and over swing) isn’t as good for acceleration as a carburetor accelerator pump that throws in a bunch of raw gas. Max power is achieved at around 11 to 1 air/fuel which the CIS unit will probably not achieve if it’s set at Porsche recommendations. If your emission controls are removed you can set the CIS richer for more power.

Of course if it is set rich for more power then it will be rich in cruise mode too and gas mileage will suffer.

I'm really just surprised that something was not done to make it go richer at full throttle. From what I can see it would be easy enough to do by linking the control pressure regulator to either the throttle or manifold vacume.

The people who designed these cars are plenty smart so I'm sure that the design was intentional. Perhaps it was for emissions reasons?

I'm not planning on playing with this as my Porsche is not my hot rod. If I want power I'll bring the 'Busa out. But I've been playing with vehicles for over 35 years now and like to understand how things work. When I studied the design of the FI system it just seemed to me like this was a hole in the design. I may well be wrong. But it got me curious.

Eric

Hladun · 07:08 AM

This link was supplied by jason2guy in a previous post. It's the Bosch CIS manual and it has an excellent description of the system;

http://www.phat-gti.com/downloads/boschtech-12d.pdf

On page 28 & 29 there is a discussion of acceleration response. These are second order effects, but they are not trivial. The cone shape has an effect and is discussed on page 24.

However, if you study these diagrams you will realize that for 30 years the vacuum lines on Porsche's have been incorrectly connected and have negated the WUR acceleration feature. I'll do a post on it soon.

And yes Eric, if you set it rich it will use more fuel...power costs. I've set my 78SC at 5% CO (about 12.5 to 1) and it gives good power with no noticeable effect on fuel consumption.

In its defense, remember Porsche used the CIS system on its turbo models for a long time. No one ever accused Porsche turbos of poor power or throttle response. It was emissions, not performance, that killed this system.

emarsh · 07-06-2003, 07:17 AM

I downloaded the pdf and will look it over when I have a little free time. Looks interesting.

350HP930 · 07-06-2003, 07:27 AM

Quote:

Originally posted by Hladun
In its defense, remember Porsche used the CIS system on its turbo models for a long time. No one ever accused Porsche turbos of poor power or throttle response. It was emissions, not performance, that killed this system.

Yes, so true and one of the reasons why I am trying to build up my 930 and keep the CIS.

One thing that has not been mentioned in this thread is the fact that CIS equipped with an O2-Lambda circuit does richen its mixture at WOT.

The lambda system goes from closed loop control (which helps to stabalize A/F ratios during throttle transitions and such) to a full rich control state when the TPS thinks that the throttle is wide open.

emarsh · 07-06-2003, 07:51 AM

Quote:

Originally posted by 350HP930
One thing that has not been mentioned in this thread is the fact that CIS equipped with an O2-Lambda circuit does richen its mixture at WOT.

True for 1980 and later. Not for my '79.

Eric

350HP930 · 07-06-2003, 08:06 AM

Whoops, I didn't think about that.

Does your warm up regulator have a hose that connects to the intake manifold?

If so I think the pre-lambda used a WUR that adjusted controll pressure based on manifold pressure to enrich the mix for WOT.

My WUR also connects via a hose but it is for boost enrichment.