CIS issue... oscillating idle right after startup
 

I believe that my auxillary air valve may be the culprit to my problem. I have very liitle experience with CIS components so I would like to get an opinion or two before I buy a new part.

The problem is that after I first start the engine, all is well and it idles perfectly... after a few minutes... say 4 or 5 minutes... the engine begins to slightly oscillate. The oscillation grows after another 3 or 4 mintes to the point that the engine dies.

I have read many posts on this site and I have determined that from my reading the auxillary air valve can cause the oscilation. Does anyone have any ideas to help me test this idea out?

Any help will be appreciated!

Another bit of information is that if I drive the car and keep the idle high for about 5 more mintes and not letting the engine die, the engine then behaves normally until I turn off the engine... let it cool...

i dought it, it either works or it doesnt. sounds like it may be too rich. always eliminate the basics first. start with a tune-up.

Part swapping CIS usually results in expensive mistakes. Your system uses many separate components that do not speak to each other and need to be tested separately to determine what is wrong. The AAV should only let in extra air as the engine starts and should be out of the picture when it is running. To test it simply pinch off both hoses and see what happens. The AAR is a different component and it controls idle speed when cold. Your problem is probably mixture related, perhaps caused by a vacuum leaks. First set the ignition timing and idle to spec, next check the mixture. You can get a rough idea what is happening with a dwell meter at the test connector for the FV. Cold it should be 58 dwell, then fluctuating between 40-58 dwell when it goes closed loop. The injector sleeve orings are a source of vacuum leaks, check for leaks with an unlit propane torch.

Check your control fuel pressure.....
 

Groovy,

Have you checked your fuel pressures (cold & hot)? Your erratic idle speed could be caused by unmetered air (vacuum leak) or WUR beginning to malfunction. Double check the vacuum hoses of your AAR.

Do not replace CIS components unless they are confirmed defective. WUR, CSV, TTS, AAR, fuel pump, etc. could be tested individually and will give you a better idea where the problem is coming from. Avoid guess work trouble shooting and you'll save a lot.

Tony

have a shop with an exhaust analyzer set the mixture for you. you might want to leave the oxy sensor unplugged and set the hot, at idle CO to around 3-3.5%. then see what it's like next morning.

I have a similar problem. My mechanic checked my vacuum unit and I have a leak on the left side. I just ordered some parts from our host including a new Vacuum unit. Also I will change the O-ring seals of my injections(I have no clue how to remove the line yet) Just pull a bit I guess as per the Technical Articles.

We dropped little water on the line and it suck it right away so a air leak there also. After that I assume the car will drive FASTER


:)

Daniell911

pull on the injector not the line. try an open end wrench under the injector for some leverage.
curious....what vacuum unit?

1-237-122-763 Pelican Part

I have 1983 911sc

Yes you are right not the line.

I saw a thread about it using a wrench and a flat screwdriver.
I will pop out. Not screw right?

Little rookie here!

But did some deep work before
See my page below

Daniell911

Do not change the AAR until you have checked everything else. I had the same problem with my 77 Carrera 3 and after trying several things I took it to the local Porsche guru and he fixed it in 30 minutes. I was charged labour only so the AAR was not changed.

Thank you all for your input! I would love to fix the oscilation myself only to learn more about the CIS system. I rebuilt the engine about 18 months ago and did not replace any CIS components except for the rubber intake sleeves on the manifold. I use the car as a track day only vehicle so fiddling does not worry me so much... I just do not want to cause myself more self inflicted expense.

Do you have the equipment to check your CO?

If not there is still one thing I would try first and that is to lean the mixture.
With the engine cold and off, place the 3mm tool into the mixture screw at a position that has the handle pointing towards a referance point. Bump the tool clockwise and then turn the screw 1/8 of a counter clockwise revolution (45*). Remove the tool and start the car. Note any changes in the cold operation and as you drive and it becomes warm. The occilation should be reduced or gone. If not, repeat the above procedure placing the tool EXACTLY where you left off. Do not adjust to a total of more than 1/2 of a revolution (180*).
If this does not work or you are not comfortable with it simply return the screw to the original position. No harm, no foul.

Is there a reason you are recommending adjusting the mixture the opposite way that the factory recommends ? Porsche recommends always making the final adjustment in the clockwise direction. His engine has the lambda system which will display a reasonable indication of the mixture setting on a dwell meter.

Paul,
I think what Rarly means is that the overall way to go to lean the mix is in the counterclockwise direction from the starting point.

Groovy,
On my car I would do what Rarly said. However, since you are new to this car & CIS, it would be better to either first check your control pressures w/ a gauge, or have a mechanic do it. If they are right, set the mixture per JWalker. It is very likely your mixture is too rich, as this is a common symptom, but you should know why it is first.

I'm just reading, not guessing. My point is all of the factory and Bosch publications state to adjust the mixture level from lean to rich. The final adjustment is to be made in the clockwise direction, opposite of his statement. The reason has to do with the screw thread and the lever hinge. I have never heard of a reason to contradict this procedure in 25 years of putting up with CIS.

The CO spec for the engine is 0.4-0.8 %, not 3-3.5%

How are you going to adjust from lean to rich if the engine is running rich?

Read what I stated. You "bump" the screw clockwise (rich) then begin CC to lean. You can then "bump" clockwise again if it makes you comfortable to do so. I also have been doing this a few years. I don't "put up" with CIS. I tune it to work properly.

Again, if you don't feel comfortable doing this don't do it. A mechanic will gladly do it for you for $90/hr.

By following the factory's procedure:

"If the mixture is excessively rich, first turn idle control screw counterclockwise further than necessary and then clockwise to the nominal value.....always adjust CO level from rich to lean" (Porsche AG)

Following your procedure can result in false and inconsistent settings and I do not recommend it. I would not describe that as "properly", Henry Royce would be turning in his grave. I have been adjusting CIS injection since 1978 and would not advise solving a minor problem by creating another.

Paul,

You and Rarly ARE saying the same thing: that to go from rich to lean you must go counterclockwise, but to finish off you should go slightly clockwise. In other words, go slightly past where you want and then come back.

The original spec for CO was based on EPA standards that had to be met for the times. For better running and horsepower, these cars seem to like a richer setting, about 3 or 3.5% as JW says, especially as they age. There seems to be a lot of leeway here, as these engines can run within that range.

Groovy,

I think the best thing is to check your control pressures first, especially noting your cold cp. It sounds to me like either your ccp is too high which results in poor starting (too lean), and someone may have tried to compensate by setting the mixture richer, or it may simply be too rich now that you have completed your rebuild. This may be because your WUR is shot, or it may be simply out of adjustment and can be reset by a process called "knocking the plug" (do a search). If all this sounds daunting or you don't have a pressure gauge, have a trusted mechanic check the pressures for you. You then have a baseline from which to start

Since you have done the rebuild, I think you can probably handle this. I have also been adjusting CIS cars since 1978 and if this were my car, I would do what Rarly said: set the CO about 1/16 turn leaner than now and see what happens. If it improves, but is still there somewhat, try another 1/16 turn. You can always put it back, and, since you live in a place where the temps are not too extreme, this may be all you need to do.

You seem to be missing the point entirely, removing the tool after a counterclockwise adjustment is incorrect and bad advice. Suggesting that it is correct , creates another error. It is impossible to properly diagnose a CIS lambda system if you set the CO to 4 to 8 times the factory spec. That will mask any number of problems and incite emails from Al Gore.

Ok, I'm an idiot, my car runs bad, (Al Gore can kiss my ass) and I have no business giving advice on tuning CIS. Happy?

Subscribing for the entertainment value.:D

Wow, I think this is going well. I love pulling politics into a CIS tune up. Typically, the high/low hunt is a rich condition. I agree that if the only tool you have available is a 3mm wrench then WTF, try to lean it out. What do you have to lose. Now if the issue is to then tweek the screw one more time is a CW direction to maintain some tension in the threads..well yeah sure ok. Thats how you should tune a guitar too you pull the string up to proper pitch by going sharp... but of course guitar tuning is a bit more critical then a gummed up CIS mixture adjustment screw. ;)

I am having the same problem with my CIS, but I know for sure my airbox is cracked, and assuming there is an air leak issue. But haven't gotten around to putting it on. Should I even bother trying to tune my CIS until I'm sure there is no air leaks?

I will absoutely try to adjust the mixture but... here is my next question... the mixture is rich as my mechanic who did the tune did so after the rebuild to keep the engine running cooler on hot days. My "home track" is No Problem Raceway in south Louisiana. The temps, starting now, get really warm. Is this a condition I should attempt to live with or should I have this condition looked at and corrected cause something bad will happen one day while making the Lotus Elise drivers cry while wondering why a 25 year old car caught him again?

Groovy911SC,

If your engine is stock and the lambda system is working, you can see what the mixture correction factor is by hooking up a dwell meter to the test plug under the plastic cover on the left hand side of the engine compartment (green/white wire). If your mixture is substantially off, the duty cycle will read outside the range I gave you.

My advice is that if this is a track car, you should forget about idle mixture and concentrate on mixture under load at WOT. Beg, borrow or steal a wide band meter like an LM-1 and test it during your track time.

wowzer911

It is pointless to tune a leaking airbox, also pointless to speculate about a leak. Swapping the airbox is no picnic, some people insist on a partial engine drop to do it. I've done it both ways and neither is fun. If you are attempting it in place, loosen the box mounts as soon as you can get to them.

You may want to use a propane torch and check for leaks around the airbox and injectors. You will hear the engine note change when it finds the leak. Porsche chose an inferior method of installing the injector sleeves and their seals typically are toast before the injector seals. After 34 years you can get wiggly injectors and the sleeves can pull out and destroy themselves over the stakes when you pull the injectors. You want to consider replacing the runner boots, gaskets and oil switch while you are at it. When its back in, check the bell crank bushings and for full throttle using an assistant. I am always surprised a how many 911's with performance parts have a throttle that only opens 3/4.

Hmmm, I won't be dropping the engine as I just did an engine swap a couple of years ago but kept my existing CIS (w/ good boots and runners). Recently cracked box w/ a backfire as there was not a pop-up valve in it (my bad). I was looking forward to installing the SS airbox (till now). Also my injectors are somewhat wiggly, is this bad, or is a little movement acceptable? I will check for leaks now that you mention it. Throttle linkage has already been adjusted, no problems there.

Also my injectors are somewhat wiggly, is this bad, or is a little movement acceptable?

The leak test will tell you.

Other than the fact that we can't seem to agree on what going "from lean to rich" means, psalt has given sound advice. What mysterytrain and others (incl me) are saying is that it sounds like you might be just slightly rich. While running too lean will make temps rise, a 1/16 or 1/8 turn leaner to cure your idle hunt probably won't change your temps, especially if your mix was too rich to begin with. Also, if this is exclusively a track car, I agree that you should tune for your race conditions, which is not the same as real world conditions.

I personally make a small mixture adjustment when warm weather arrives with a 1/16 leaner and yes the idle occilates at start up and stops after the adjustment. I am curious too why the final turn of the mixture screw should be clockwise and if anyone knows the techical/mechanical answer, please advise us:) If you just put on new carbs or a new FI system and had no idea where you are set, then I can see tuning rich to lean. But binding or ruining something with an allen wrench removal after a 1/16 CC turn is interesting to me.

I've been through the airleak deal too. Think you found it, fix it and then it still idles high. Finally I done a another partial drop and went through the complete CIS to finally correct the leak. It wasnt too bad, but I was proficient in the partial drop by the time I got it fixed.

http://forums.pelicanparts.com/porsche-911-technical-forum/219508-man-teaches-chimp-remove-cis-90-minutes-partial-drop.html

I made a small writeup including non-Porsche procedures like using JB Weld to seal the airbox :) No duct tape though.

Scott,

It has nothing to do with the concept of tuning from rich to lean, like when you are programming an EFI fuel curve. It has to do with the design of the CIS air flow sensor. The basic fuel metering of the system is controlled by the mechanical relationship between the air flow sensor and the control plunger. The air flow sensor is hinged at one end, with a second floating lever that contains the adjustment screw and makes the actual contact with the control plunger. The reason Bosch and Porsche have advised to always make the final adjustment in the clockwise direction is that when you back out the screw, you remove the tension from the screw threads and the adjustment is not reliable. The screw is also more likely to move from vibration when left in this position. It is a mechanical issue, not a "rich or lean" issue. This has be known since the 1970's, mentioned in all the publications, and one of the first thing taught in K-jet 101. Unless you have your head on the desk.

If your 81SC CIS is working correctly, you can use the lambda test port and a dwell meter to make repeatable adjustments with more accuracy than "1/16 of a turn", especially when you make the adjustment using the proper procedure.

I'm putting down a dollar on the WUR heat element.
If the electrical connection is disconnected the idle will oscillate until the heat from the engine heat up the WUR.

Been there and done that by testing the FV with the dwell before and after the 1/16 turn. All your help is appreciated, but we're all not primative Pete's without procedure manuals. :) I think the confusion arose when you called it tuning rich-2-lean, when it has nothing to do with the mixture. Phrasing it as "always make the last turn clockwise to lock the setting" would have been suffice. Thanks gain

Yes, reading is a skill. It is lean to rich, not rich and lean, and that is how it is written in all of the Porsche factory manuals. Reread the posts, you are the only one calling it rich to lean. Then read the manual.

My 930 does that sometimes too, and it is a symptom of being a little rich at warm idle.
But the reason mine is doing it is something is causing the sensor plate to hang a tiny bit.
Sometimes when the engine is off the fuel pumps will come on when turning on the ignition.
That is also because the sensor plate is not fully returning to the rested position it should be in when the motor is off.
Because it only does it occasionally I havn't done anything about it.

Sounds like you need to reset the sensor plate Zero position. There is a good description of how to do this in Probst's book on Bosch fuel injection.

LOL. You are a piece of work my fellow pelican. See below ;)

If YOU read your personal post again, you quoted the Porsche manual as rich to lean. Please read carefully and correct your mistake before you write another condescending post. You seem to have alot of knowlegde and experience. It would be well received and usefull if you could leave out the off color comments.

condescending: To deal with people in a patronizingly superior manner.

Safe, this is not true. If the heat element has failed, or does not receive power, the car will run well after a warm start because the WUR has heat soaked. After a few minutes of running, the top of the engine will cool sufficiently, and the WUR will cool to the point of richening the mixture again. I suppose a hot running engine may behave as you described, but if all else is in order, no.

Doug
75 911S

I'm convinced 'lean to rich' is refering to the mechanical direction of the final adjustment of the mixture screw. Isn't this the same concept as torquing a bolt? If you over torque you back off past the setting and then bring it backup. I still think we are spliting hairs. Lets not forget that this is the same system that requires a hammer to adjust the cold pressures.
Speaking of the WUR, I'm not sure the WUR or the AAR will heat soak from the engine and be able to operate correctly without the heating elements functioning. Remember the elements don't timeout..they remain hot when the ignition is on.

Ron,

Yes, the "why" is a mechanical issue. Porsche chose to write it as lean to rich in the manual, but like many other instructions, it does not explain the reason.