Technical CIS operation & Cam Timing w/ CIS Discussion
 

Hello all.

I've read many posts on this subject, but I have yet to find a detailed thread that explains exactly how a CIS system meters fuel and what inputs is uses to do so. I believe it soley uses vacuum, but I'm unsure.

I'd also like to talk about camshafts while using the CIS system, as it appears most people think that there isn't much of a choice here. I think I can give a lot of suggestions and even test quite a few profiles on my engine, but I'm limited on actual CIS "function" knowledge right now.

I have a feeling that the CIS unit uses vacuum for the majority of it's input. I believe this because I have read that you can't use "s" cams with the CIS system. I bet the "s" cam has a tight LC (lobe center) with more duration then a non-"s" cam. What this does is drop the vacuum signal at idle and off idle running....which "might" hurt CIS performance. Again, I don't have full CIS knowledge yet.

Anyway, I'll go into this more in a bit. Can someone chime in on the operation of the CIS and anything they know about camshaft profiles while using the CIS system?

Thanks!

Jay

You could say that the air meter works on vacuum (differential pressure). Flow through the air meter body causes the air plate to rise which in turn raises the plunger in the fuel distributor. The plunger position w/in the fuel distributor and the fuel control pressure determines how much fuel is injected into the engine.

Reasons why folks have used conservative cams on a CIS engine:

-piston/valve clearance

-intake pulses (reversion) that will result in an oscillating air plate...(see above interaction as it relates to fuel).

I love to see folks wanting to experiment, but I think you should research CIS cam timing advance, and 964 camshaft installation first.

I would venture that the CIS meters fuel according to the mass (inertia) of the air entering the engine and not vacuum. There are numerous sites on the internet dedicated to CIS, read around some are quite in depth. I printed a whole CIS manual that I found on the web in .pdf form.

So, vacuum signal does play an important role with CIS? I've been thinking about a custom ground "s" cam, but with a wider LC to increase the vacuum signal at idle and off idle. I've used this trick on many other EFI engines, especially on MAP equipped vehicles, but I'm still learning about the operation of the CIS.

I truly think that aggressive cams can be used (without low RPM problems), if the right combination of compression/ignition timing/and cam timing is selected.

I admit that I need more knowledge on the CIS system, but it is hard for me to believe that it is limited to just above stock HP numbers. Maybe I just "want" to believe that it could work well in a "hot" engine build, but as of right now I can't see why it couldn't.

Please note that I'm not trying to argue with anyone, but merely trying to gain CIS knowledge and discuss CIS vs Cam profiles with everyone.

Thanks!

Jay

I just found a very cool tech diagram on how the CIS meters fuel.

http://img170.imageshack.us/img170/7483/cis29gm.gif

http://img170.imageshack.us/img170/6...eldissy3vh.gif

It's just like you said....metered off of the amount of incoming air and how far up it "pushes" the plate/arm.

I bet you could made enrichment "lobes" that basically built up the range of motion for the plungeron the metering port. Either that or we can move the sensor plate further away from the arm? Who knows, maybe it is as simple as changing spring rates in the pressure regulator. This way you can keep low end fuel delivery AND have greater fuel delivery at full throttle. I'm betting the spring rate change is the way to go.

Again, I'm just thinking out loud here.

Thoughts anyone?

the mass of the plate and its associated parts + friction will always limit the responsiveness of these cars

I would NOT try to experiment with hotter cams, etc. -- unless the engine belonged to an enemy...

FWIW - I am a known CIS-hater. Death to the CIS!

I have to agree with you on that point. That doesn't mean it can't be modified for less then a full injection swap. Again, just brainstorming. It sounds like everyone is against the CIS though.

It is the work of the Devil

Get your pitchfork and torch and meet me outside the moat...

I thought the Motronics had hot wire air velocity sensors -- like in an anemometer?

Fuel Distributor

With regard to the CIS system - it is limited in the amount of fuel that it delivers. There are openings in the body of the fuel distributor housing alongside the the plunger to allow fuel to reach the injectors when the plunger is raised. The plunger covers these openings when it is at rest (car not running)

When the engine is started, the sensor plate lifts up the plunger and the narrow part of the plunber is exposed to the incoming fuel flow and at the same time to the openings which provide fuel to the injectors. As the sensor plate rises, more of the narrow part of the plunger is exposed to allow for more fuel to flow.

Maximum fuel flow is reached when the full lenght of the narrow part of the plunger is exposed to the fuel coming in at the bottom and the fuel leaving the top of the fuel distributor (both dark and grey areas on the above diagram).

Now, if you continue to raise the plunger, the bottom of the plunger which is the same width as the top of the plunger will begin to restrict the flow of the incoming fuel by closing off the bottom opening (the light grey area on the above diagram).

The diagram shows the maximum output when the engine is running under full load. (both dark grey/light grey areas are exposed to the narrow part of the plunger)

Control Pressure Regulator

Now, if we reduce the pressure on the top of the plunger by allowing more fuel to flow through the control pressure regulator, we would then increase the fuel flow through the fuel distributor. (cold start and run condition)

On the other hand, if we put more pressure on the top of the plunger by restricting the flow through the control pressure regulator, we would then reduce the fuel flow through the fuel distributor. (warm start and run condition)

The control pressure regulator affects how high the air flow sensor lifts.


Hope this info helps

Steve

"A Porsche does more then just go fast in a straight line"

I picked up this book at my local Borders. It covers all the various CIS systems.

Guys, the air plate is counter balanced. It's balance enough that if there were no fuel pressure, you could blow on it, and it will move.

Flow through the throat of the meter creates a "differential" across the plate...that's what causes it to lift! Flow and differential pressure...internal flow...crack a book. No need to beat this one...concentrate on the valves to piston clearance and overlap. You don't want parts hitting and you want to minmize reversion.

Randy, the DME AFM is trap door type meter..not a hot wire.

Getting fuel to the injectors is not a big hurdle. I can run rich enough to wash down the cylinders! And if that's not enough, you can add extra injectors. Fuel and air through the air meter are not your problems.

Getting the engine to breath (at the valve) is your big hurdle. If you are going to go custom grind, start with a 964 cam. Then find out what your clearance is over the range of cam rotation. You might be able to add a little more lift and maybe add some duration until you start to stutter when running because your air plate is dancing!

I've heard that the AFM meter is less affected by reversion than the CIS air plate....but that's not going to help Jay any.

If you can make a custom grind that is an improvement over the 964 grind, you'll have a nice product for the CIS and 3.2 engines!. But be wary of kissing the valves. Has anyone ever checked the piston to valve clearance on with a 964 cam on a 3.0 or 3.2 across the rotation? What is that number? (John - Camgrinder?)

Take a look at the top of a CIS or 3.2 piston. Someone posted a picture in the picture of a CIS/3.2 piston in the engine forum with valve pockets cut into it...seem it might have been less effort to get JE pistons and build from that.

If the ultimate goal is to make a better CIS engine...concentrate on the cams/vavles...use the 964 as your baseline and build from that...until you start hitting parts. Air and fuel can be tweeked easily.

If you really want to take this to the next level, design a pulsation bottle of sort to cancel or dampen the reversion! Put it between the air meter and the throttle body. Frequency range would be limited unless you build a dynamic bottle to keep up with the change in frequency. One drawback that comes to mind quickly is lag...thottle lag due to the volume in the bottle.

I like my CIS engine..it runs well and I can tune it with my eyes closed...but I understand it's limitations.

If you want more....pay up. There are plenty of engine combinations...just have a fat wallet.

Not at all, some reactionaries aside...

The CIS is a relatively simple, very robust system that offers excellent driveability and is a HUGE advance over carbs. It can't stack up with later Motronic systems, but it is pretty good at what it does and is easy to troubleshoot once you understand the principles of operation,

ianc

I agree it is a huge advance over carbs in terms of emissions and fuel economy - maybe warmup too.

It is simply inappropriate for a sports car.

Troubleshooting is not too horrible once you (re) educate yourself on how it works (tho it is still harder than carbs). Unfortunately, one troubleshoots CIS by replacing components, each of which is usually a few hundered dollars in cost. After doing 3 or 4 replacements, tho, you will eventually find the problem. Enjoy!

I do think Souk has some good ideas. Everyone should go back and print out his last sentence and staple it to your chest...

Hmmm... My car seems to like it fine, and so do I. :)

Definitely not. Leaky throttle bushings? Trying to jet correctly for different engines when you've modified? Byzantine linkage mechansims? Trying to synchronize 6 carb throats? Troubleshooting poor running conditions with carbs is an absolute nightmare, even with only ONE!

I suppose one could approach problems that way, if one didn't really understand how the system worked...

ianc

Souk,

I appreciate the input and suggestions. They are truly appreciated and provided me with some great information!

I'll be the first to admit that I am new to Porsche engines, but I am also well versed in engine design, theory, and assembly. I know that you didn’t imply that I lacked knowledge or experience, but I wanted you to know that I am not green when it comes to performance engineering. I have the engineering and hands-on background to discuss this CIS/CAM "problem" in full detail, plus possibly sort a solution with other members.

I understand differential pressure and its effects on airflow (don’t forget intake pulses, etc…). I design and engineer cold air intakes for many major manufacturers on a daily basis, plus I work with Honeywell on filter design for the Fram “AirHog” line of filters. I've spent about the last 11 years designing and producing parts that directly rely on thermal dynamics.

I’m not trying to say “look what I’ve done”, but more to point out that this was a mere attempt for a discussion from a fellow user of this board to fully dissect the CIS limitations and possibly raise the performance bar a bit. I just needed a little bit of background on the basic function of the CIS system first before the discussion went totally in-depth.

I’ve done some searching on the net and these links might help others who want a little more knowledge about the CIS system and how it works.
How the Porsche CIS system works
CIS Troubleshooting Chart
Adjusting Your Warm-Up Regulator

Now, 99% of the threads that I have read on this board say that the horsepower limit is about 210HP (N/A) for a CIS equipped 2.7L engine. I can't pick up a book and read about the "why's" and "how's” of this because there hasn't been a book written concerning this directly (That I know of). If you know of a book that explains in detail why not to use an "s" cam with the CIS unit or why a 2.7L is limited to 210hp with the CIS, then please let me know and I'll buy it tomorrow! That information has come from users of this and other forums and I (and possibly others) would just like to discuss it instead of just passing it off as truth.

While we’re discussing this, I’ll be tearing down my 2.7L engine at the same time. I have the equipment to flow test everything and I plan on doing just that. Once I have all of my flow data, I’ll post all of the data for everyone to compare. I will then design a few camshaft profiles around that airflow data and set the compression accordingly. This is so far ahead of where we’re at in the discussion though.

Along with trying to get a little background on the CIS system, I want to discuss real world experience concerning the CIS unit and large cams from other users on the board. Trust me, I’ve searched and have yet to find a clear explanation as to why it has these “limits” besides reversion and vibrations. Depending on the setup, some reversion can be eliminated using the proper cam profile and compression ratio. Again, I think this topic needs more detailed information before we dive into new cam profiles.

Thanks and I look forward to reading and providing some great information on this subject!

Jay



PS – Souk & everyone else, please do not take this as an attack or as a harsh reply directed at you. I am merely trying to explain myself and my writing style has a tendency to come off as aggressive or harsh. This was not my intent at all with this reply.

Jay,

You'll find that my posts are all harsh! Well they come off like that anyhow. ;) :D Really, you have to wear your "thick-skin suit" when you start discussing technical matters on this board. Not because folks are going to jump down your throat, but because you don't get the facal expressions and body language through the screen. You know how it is when you start discussing tech topics with a bunch of engineers....oh boy...get a big cup of coffee! Heck I've even managed to piss off Randy a few times before. :D

Like I stated in my first post here, I love to see folks going down paths never taken. Engineering can be a real bore if all we do is piece together parts that we already know will work. It's like building a lego project by copying the picture on the box! Woohoo! :D

I have not found one source that lays everything out for CIS. Everything I know about CIS comes from several books and other resources...and from frustrating work on CIS. Heck I should quite my day job and write a book :) Yep...that'll be a hot seller.

Anyhow, I like that you are chasing the hard truth...and armed with an engineering background! If you were standing here I'd hug you man! :D

Now, anything you build is not going to be limited like the factory was. Remember they were building cars for the general public..well, the sports car loving portion. So going to more aggressive cams may result in a car that we all like, and the car may behave in such a way that we can endure. Compromises that the fancy Long Island lawyer would hate...we'd all love! (sorry all you lawyers :D ... I'm not sorry about the Long Island part though :D ;)) If you are going to use new pistons, great! That may take car of one of the biggest concerns (valve contact). Now, there is considerable damping in the induction, but I've never seen anyone publish information/data on the use of aggressive cams. I would live to see where your cam project goes!

If you are building a mean street engine....some of the what you take away from a CIS' driveability is probably fine. Think of all the lumpy engines out there...if you have the time and resources...GO FOR IT! Now that I know you are going to open the engine up and do all the stuff....go wild! Then back down...changing cams ain't that difficult.

Something else that occured to me while I was typing this post...

If you want to go another step further...build a damper for the air meter plate/lever. Dampen it just enough to handle the pulses, but still allow it to move freely when it has to as more fuel is needed. That will not be as difficult as the pulsation bottle idea.

Souk,

Thanks for the reply and for understanding my writing style! I didn't want anyone to think I was being harsh, hence my many "disclaimers" :D

As far as the engine build goes, I just want to build power out of the 2.7L CIS engine and I don't understand why it has such a low HP "limit". A switch to carbs or a different EFI removes said limits, so I'm just curious as to why. From the way this thread has gone so far, it looks like it might be a pretty heated topic too. Should prove to be fun!

I'm concerned with weight and its affects on the balance of the lever. I'm thinking a different sensor plate (lip lower then center of plate) or a different arm might just be the answer to the majority of the problem.

Since this isn't a vacuum problem associated with the CIS (MAP systems), then cam choice should be larger then what people think. It sounds like it is more fuel control at different RPM and cylinder pressure levels then anything else.

The biggest hurdle will be fine tuning the CIS at all RPM points. Just like a carb, this is a mechanical induction, so hard parts will need to be changed/modified to tune the beast. This might mean new springs for the regulator, maybe a new plate/lever, and who knows what else? I think if we can fully understand how small changes in the lever/springs/plate/fuel pressure affect the tune, then we might be able to see much larger HP numbers with the CIS system as well.

Well, this is just my .02 anyway!

Thanks!

Jay

I would not mess with modifying the air plate lever. Like I said, it’s balanced, and it’s designed to react such that air and fuel are within acceptable ratios. Altering the level will require that you calibrate it with the air plate (and air meter throat) and the fuel plunger travel.

You could space the plunger (fuel dist.) closer to the air plate to increase the plungers travel relative to the plate’s travel, but even that will hit a limit…(the plunger/fuel dist limit). That would require a modified lever so that the roller on the lever moves with the plunger. You’ll also have to work out the force that’s required to overcome the fuel pressure force on the plunger. Bosch spent a lot of R&D money to develop the system. Even re-engineering will take considerable effort and resource. (If you have both..spread it around man…I’d love to quit my day job and play with stuff like this!)

A little CIS trick that I’ve yet to quantify (although my 3.0 runs exceptionally well, if I can boast a bit) is to lower the air plate by lowering the stop under the air plate, the idle mixture then has to be adjusted by raising the plunger back to the “before” position. You are essentially lowering the air plate w/o lowering the plunger. The result is a slightly quicker/and higher rising rate of the plunger. Like I said, I have yet to quantify it, but with all the things I’ve done to my CIS engine, I’m not aching for carbs or EFI just yet. (This is the first time I posted this...so if anyone wants to do this, and has free dyno time I would love to get some data! Unfortunately, my track car has been out of commmision for so long..and I've not have any time to play on a dyno :( )

Again, fuel quantity will not be your problem…dynamic fuel control (quality w/r to AFR) will be a problem as you go toward more aggressive cams. The big hurdle is designing a cam that creates just enough intake pulses yet will still allow for a smooth and “linear” air plate operation. You can alter the fuel control pressure with a modified warm up regulator (WUR). Unless you change the air intake quantity, you’ll run rich all the time. We know running rich is not the answer to more power. I’ve been playing with a dynamic regulator idea that can be set for an “ideal” idle control pressure and as the RPM’s change or as boost changes, so does the control pressure. A lower control pressure will get more fuel into the cylinders. Get the idea? The stock WUR is too simple to work the way I want it to work with a highly modified engine…or a boosted engine. You’ll need feedback.

While you are at it, you’ll also want to massage the heads, change the intake runner ID…the list goes on.

I admire your out of the box thinking, but if you are going to spend all the effort and resources…why not go with EFI? OK…I didn’t say that…wait! I did say that. I also said that there are limitations to an engine running with CIS. That’s why MotoMotions is developing carburetors and EFI. Search ITB EFI. The EFI kit will surprise you from a performance and cost perspective.

Souk,

That is basically what I'm thiking as well. We have the same idea, but discussed making it happen using two different methods.

We might be able to machine down the "stop", but we'll have to figure out how much we can remove and what angle to cut at (the plate moves on an arm, so the lower the plate goes the angle of the plate will contiue to change. That plate to stop seal is important for idle quality).

Another option might be a different sensor plate like I said. The weight would be the same, but maybe it'll have a "step" on the lip to allow the arm to sit lower? I'll have a better idea on which way to go once I tear into the system and actually hold the pieces in my hand.

I've designed EFI systems for vehicles before, especially the inline 6. I usually run with the Fel-Pro unit (F.A.S.T. now) for control and then one-off everything else. I'd really like to avoid swaping systems and I don't mind "tweaking" the CIS. Actually, I think it's just my sick mind wanting to tinker and make the CIS work where everyone says it can't. :D

http://img305.imageshack.us/img305/5...ection45re.jpg

Thanks!

Jay

83 SC (Alucil p/c's) with 964 cams, .25 milled off the heads and running 1.0mm deck height measured about .25mm over min p/v clearance.

After looking at the diagram, I think the easiest option might be the "cam" idea on the arm. Take a look at the diagram below and how the idle adjustment arm has that round stand-off on it. By changing the height of this stand-off (smaller), we can actually get more range of motion out of the arm. This might be easier then a new sensor plate and/or cutting down the plate "stop".

It is either that or maybe a new plunger? I'm thinking the "ports" in the plunger might make all of the mods useless....now that I look at it. A new plunger might be needed for these mods to work.

Thoughts?

http://img170.imageshack.us/img170/6...eldissy3vh.gif

That's tight Charlie...1.25mm. So I guess, CIS pistons with 964 cams is the limit w/r to lift?

Jay, the roller under the plunger allows for movement w/o off-axis forces. A cam solidly mounted on the lever will induce some off-axis force on the plunger. Will it hurt? maybe not, but the plunger-fuel dist. tolerance is tight...tight enough to prevent fuel from coming out of the plunger.

Like I side...it's not fuel quantity you need to worry about.

Has anyone ever flow tested the heads? Is more lift really needed? If the heads don't flow more with higher lift, then we shouldn't run more lift. We can change the ramps, nose, and profile of the lobes to help flow.

Min intake is 1.50mm, min exhaust is 2.00mm. The 964 cams timed at 1.8mm measured about 1.75mm intake and 2.25mm exhaust. Still some room to play, especially if you want to play with cam timing as well to make more room for lift.

EDIT: FYI the webcam 20/21 grind and Elgin SuperC2 (both very similar grinds) both have a tad more lift and duration than the 964 grind. These things scream in a 3.2 and also work well in a 3.0. Don't have clearance figures for these though.

We can still make it a roller, but with a smaller O.D. then. Again, these are just thoughts.

I'm not so worried about fuel quantity, but more so about fuel control over a wider range. The CIS has a set "range" it works in and you can adjust within that range for rich/lean conditions. If we can open that range up a little then we might see the improvement that we need/desire?

The more I look the picture, the more I'm starting to think the answer is in the plunger/spring.

Charlie...I just got some 964 cams in...maybe I should have gone with 20/21..but would the 20/21 pass emissions? :( I still have another set of cams I can send out for regrind.

Quick Jay...get to work on the cam! The fuel is simple...just get me some wild cams to work with my DME 3.2 Carrera! And still pass emissions!

What kind of exhaust are you planning on using (SSI's, OEM w/cat)? Cats can hide ALOT and you can most probably pass smog with either cam, a healthy cat, lean idle co% and stock timing (25 degrees btdc). With SSI's its a toss up with either cam, but I wouldnt expect it to pass. The 20/21's are a bit more emissions unfriendly.

BTW: If you want hotter cams, I'd send the cores to Camgrinder and have him grind you the similar 20/21 grind for $500 cheaper.

Yep...John did all my valve train work. He made the 964 cams for me. :) I did ask him about my options for cams with emissions being a factor. I guess I have to build a spare engine and swap it in every two years :D

FWIW, most CIS systems have the capacity to handle more to much more air/fuel than their orig. installation. As an example, there was a fellow running 7psi supercharged on his 4.5L 928 using the stock injection system; I believe he even sells a kit for this application. The guy who used to care for the injection on my 928 before I took that chore over was racing some VW Rabbit/Golf with CIS - same thing, highly modded, but stock injection components.

Now whether this relates to 911 apps or not..

I'm a little dumb in my insistence to hot rod my CIS ( "it can't be done!!") The crux of it is the atomization due to the fuel hanging around the port for a while, compounded by the cam overlap.

Pat

I don't doubt that at all, 928ram. In fact, it is what I'm hoping for! Basically, I'm trying to get to the bottom of why larger cams can't be used with the CIS and why the 2.7L CIS engine is limited to about 200hp.

I think I understand the "s" cam statement now, as it would induce reversion in the intake system. I'm curious if anyone has tried to increase exhaust duration and leave the intake side alone? (I'm not forgetting overlap, etc...). Remember, this cam was designed to promote high port signals for the carbs. Not a good idea when running a single plenum....especially the small one found on the CIS system.

Reversion can be avoided by changing cam timing, increasing plenum size, improving exhaust flow, etc... Obviously, anything you can do to scavenge the exhaust and reduce overlap will aide in stopping reversion. There are side effects though, so modifications must be done wisely.

Does anyone know the exact specs on the "s" cam and the "964" cams?

Opening/closing events -
Intake/exhaust duration -
Intake/Exhaust Lift -
LSA -

Also, has anyone every tried using a higher ratio rocker on the exhaust side (if even possible)?

Maybe this whole CIS thing is WAY off and it isn't the actual system that is bad! Maybe some fine tuning to reduce reversion while using larger cams is the actual answer. Is reversion the major problem then?

Again, I'm still trying to have someone step up and say EXACTLY what is limiting the power output the CIS system? Why is it well known that 200hp is about all you'll get from a 2.7L engine with a CIS, but you can make almost 300hp by changing the induction? After more research, I'm guessing it is a reversion problem only and the camshaft design was never addressed.

I'd guess limiting HP is around 500...the turbos ran CIS

Here are a few quotes that I found -

and another -

Good finds! So, we're on the right track here with overlap/reversion. The direction now slight mods to the plenum volume/exhaust flow/and cam profile.

I think that 200hp limit is going to go bye-bye real soon! :D

Hi,

I haven't seen anybody mention the biggest performance problem of CIS (I have 2 CIS cars and I like CIS for its simplicity and robustness):
It meters fuel more or less by air mass. It measures air mass by putting a circular plate square in the air stream. The aerodynamic resistance of that plate is (more or less) proportional to the air mass flow and used to move the plunger. But that plate is also a big intake restriction. Basically the volumetric efficiency of the engine is limited by it, no matter how well you make it breathe downstream. And as the plate has basically the same aerodynamic resistance in either flow direction (unlike a flapper style MAF sensor) it is sensitive to intake reversion. Porsche (and other CIS users) use the rubber boot between throttle body and CIS plate throat as pulse damper, but that works only so far. I actually measured the effects of that rubber boot on AFR when switching between a new one and an old and hardend one, and they were measurable at certain RPMs. The old boot was still airtight, so it was not leaks but more of the intake pulses getting to the plate.

Regards,
Klaus

I would be happy with 200 Hp! How do I get that with CIS?

I can actually flow the entire system and I plan on doing that rather quickly. Once that is done, we'll be able to see the power numbers the system could "support".

So your Air/Fuel ratio changed quite a bit after seitching the boot alone? How did you measure the change?

Also - Thanks for the info!

I'm still working on that myself. Those numbers are quoted from other posts on this forum.

If we can get enough data, then we'll be able to see what we can and can't do before we spend the money on modifications. That is the point here!

Jay