Chasing rough running issue - 3.4 C2 - new stuff!

So my '99 C2 isn't quite running right. I've written this topic a handful of times as it's so vague and hard to describe it pisses me off but thought I'd throw this out to the hive mind. Any leads, suggestions or commiserations much appreciated. It starts on the button and has plenty of power (although it's not been on rollers) but it doesn't idle as smoothly, sound as crisp or rev as cleanly as my fiancee's 987 2.7. It just doesn't have that howl that I hear from other 3.4s - like this one: https://www.youtube.com/watch?v=CXlSZ8aG9Bw This is basically throughout the rev range. I've been thinking I might have an air leak, and I've found some leakage around the brake booster take off but I understand that this should only affect idle and low revs - as the revs rise the proportional reduction in unmetered air should make the issue go away, but it doesn't - it's still a bit rough all the way to the redline. I know there is a tiny tiny blow around the manifold to cat connection - only visible when I block the exhaust tip and spray on soapy water - could this be causing rough running? it's on the list to sort. There are no codes logged that I can read on my OBDII reader. I'm thinking maybe it might be slightly elderly o2 sensors? Either that or I start going down a rabbit hole of cam deviations etc. I had fitted new spark plugs and coils as they seemed a good candidate but they made no difference at all! Before I bought the car it had a new AOS and MAF so I don't think it's either of these. This does feel like it might in ignition related. I'm going to take it to a Porsche garage locally to have the DME interrogated but this feels a bit like admitting defeat....

You need to look at fuel adaptations if there is a fault with no codes . off hand .. Vario cam slippers have worn .. Clean out the throttle housing .. Check maf sensor voltage .. should be 1 ignition on but not engine running . Fuel pump , regulator .. a fuel pressure check .. Slightly off with a high lean value for adaptations .. a fuel injector leaking . I'm afraid an early 996 has basic fault codes .. your car will be not 100% but not enough to generate a fault code to point you in a direction . Later cars have 2 cam sensors so you can check cam deviation , yours doesn't i'm afraid . It's the kind of job that i would listen to , drive the car , get a feeling for it then give you an idea how much it will cost for me to look into it .. not always cheap when there is no direction to go . I still think getting a Porsche guy to have a drive and say what he thinks is your best way forward .

Thanks deMort, I was hoping you'd supply some input. Some things to look at for sure. You're down in Brighton are you not? I might see if I can get the car down to you at some point, I used to live there so an opportunity for a visit is always welcomed if I can't get anyone in Saaaf London to point me in the right direction.

Just to chip in as I have a 987.1 3.2 and a 996.1 3.4 cable throttle, they are very different on the throttle response, the 987 being more willing to respond to tiny throttle inputs as you ascribe to being an Egas car, whereas the 996 is a little more lazy being an actual cable throttle car and obviously older technology. The Boxster seems to want to rev more quickly to the redline, yet the 996 just pulls more and harder and feels less finessed to me, again, as it is older tech :grin: Just my thoughts :thumb:

That's a fair point and it's something that has crossed my mind - there is basically a generation between the engines after all, even if they're the same basic block - variocam vs variocam+, cable throttle vs egas... What really proved it to me is the sound in that clip I posted - having changed my back boxes I did the same thing (so hard to resist revving the engine when the back boxes are off!) and the noise was distinctly different, much less smooth as mentioned.

OK, as I have SORNed my 996.1 and Daily driver Merc over lockdown and not touched them for 7 weeks or so, and I`m so bored now and lockdown is easing, I`ll get the 996.1 started , the battery will be flat so I`ll have to charge it up, and then make a video of revving them both as a comparison for you, it might take a day or 2 as I need to get my battery charger back as I lent it out, along with my jump starter to my kids :grin:

Bank 1 variocam solenoid is prone to sticking and causes idling problems and alters performance through the rev range. Deffo worth getting both solenoids checked with a Porsche diagnostic tester if you haven't already.

Would be interesting to see. I'm running some noisy Dansk back boxes which makes the comparison slightly tough but I will try and do the same thing at some point. This was the main reason I was looking to get the car attached to some diagnostics - I suspect this could be an issue, but I understood the variocam only activates above 4000rpm? Would it affect low rev performance also? I'm tempted to pull them both out and check them manually, or just unplug them to see if there is a noticeable difference in performance. Would this be safe?

I had a duff variocam solenoid. Symtoms were a rough idle and poor low rpm torque. The exhaust connected to the bank concerned pulsed differently to the other one. Replacing the solenoid cured it. There were no code that pointed directly to the fault, just misfire codes that could have been caused by many other things. You're fine to unplug the solenoids when troubleshooting. The one that you unplug and the exhaust note is unchanged at idle is the duff one.

Also -if it is a variocam solenoid then when you start the car it will hunt up and down a couple of times before settling at idle speed. When the solenoids are good this hunting doesn't happen and the revs rise once on starting then dip once slightly lower than idle and then it goes straight to a steady idle with no further hunting.

Interesting will definitely give this a go. I had a similar issue on my M3 and unplugging the solenoids made a distinct difference so very useful for troubleshooting.

A little update about this. I've been logging various data while I've gone for a few drives over the last few days. The results of one such run can be seen below, might be interesting. This is all done via a cheap bluetooth code reader, a crucial tool these days and a smartphone app - carscanner pro in this example. Note the x axis is time - not RPM! This is a good example of taking a snapshot of the fuel trim data is not necessarily that helpful - it does move around a fair bit. Above shows the long term fuel trims for the Left Bank (blue, bank 1) and the right bank (Orange, bank 2). It shows high fuel trims on both left and right banks, slightly higher on the left bank. A very useful indicator is that the fuel trims are high when the car is idling (with an effectively closed throttle, more air is sucked in via leaks) but drop as the car is driven as proportionally more air is taken in via the throttle, a good sign of a vacuum leak. Having carefully check the hoses I found a leak around the brake booster take off. This was missing an o-ring (number 4) here: I also wasn't happy with the seal around the aluminium tubing in the connector itself, it appears to be missing an internal o-ring and so I added one here which seemed to fit nicely - maybe this was lost at some point in the past? Interesting that the brake booster is on Bank 1 - the one showing the highest trim. Will log my drive this afternoon to see if the fuel trims are improved. MAF sensor information also added for DeMort do look over. Seem pretty good to me. I'm leaning towards this being a variocam issue of some description. EDIT: Is it possible to remove the solenoids while the engine is in the car? Would like to take them out to test them with a 12V supply

I believe it is possible to change solenoids in-situ. Need to get the car on a ramp of some sort. You will need to lock the cams in place before removing cam cover - don't be fooled by solenoid cover plates as easy access. If you can lower the engine slightly, then it will give some crucial space to work. An indie or diagnostic tool can turn off the solenoids or check fault codes. In theory you could unplug one solenoid as see is if it changes engine behaviour.

Yeah I see it involves removing the cam cover. Definitely not what I want to do at this stage, I'll just have to rig up a tester. Was fooled by the seal plates, it looks like they might just slide out. Good news is that air leak appears fixed, trim values very steady. Going to have a play this weekend I think.

Young man i really don't think you need much help from me .. you are well on the ball at finding faults and have already found what i would consider to be a major one . A missing O ring at that point is more than a small vacuum pipe disconnected , you need to either clear a fault code from the dme or generate a code then clear it to reset the fuel adaptations .. ie pull the maf sensor then clear the code for it . This puts you at ground zero and you can monitor fueling from there on . A smoke machine is brilliant at finding faults such as this .. never had one at OPC .. use the one we have a lot these days ... always makes me happy dragging it out of the cupboard ! Ref Maf sensors , i've scoped a good and a bad maf back to back pre porsche days .. i couldn't tell the difference from either graph .. The trouble is scopes take a snapshot .. rather like rounding up or down numbers .. an over view , as such you wont see the true picture . With porsche i work with what it feels like , what the symptoms are .. what the reading is at idle or .. Ignition on , engine not running and i want to see 1 volt on the signal wire .. the limits are 0.9-1.1 volts but if reading 1.5 then its already out and will continue to be so for the entire rev range .. over fueling . I also hate g/sec .. all porsche stuff works in kg/hr .. well other than the very new cars .. i have to convert it .. rough amount is 16 kg/hr on idle .. it alters with the a/c on or off though due to engine load so 14 - 19 kg/hr is in limits ( hot engine ) . Ill let you convert that lol . Solenoids and some testers can disable them .. you listen for a change in rpm .. you can unplug them and r/t but thats a dubious method .. it wont really point you in a direction unless its dead . To remove is as said .. take off the cam cover .. by doing that you can inspect the slippers as well . Overall though .. you had a minor fault .. you have found and fixed a pretty good fault so i would monitor it .. We can after all get obsessed and try and fix every little bit when the car can adapt to most issues .. when it can't then it lets you know .. you get an eml