Rough idle and fault code pending

Folks, Have a bit of an issue I could do with picking some brains over. Rebuilt the car around 4 years ago and at the time it had an occasional lumpy idle once warm. Never had a CEL come on or any problems and after initial engine checks after the rebuild I was advised to drive it and see if it : A tripped the CEL for me or B went away. 10,000 miles later and lots of fun and it�s still there. Because of lockdown I�ve decided to get back to it and attempt to find the cause. So started with the simple stuff like plugs, coil pack, leakdowns, compression and vacuums and so far were all good. It�s bank 1 only and it�s not short on power. The plugs were looking a bit mucky but I�m not getting through oil so I�ve assumed that�s just because I so a lot of short journeys. I�ve tried to attach a video for you guys to look at, basically it only starts once warm and stupidly I hadn�t realised it�s when it comes off choke.... and I know before the choke comes off it�s open look so that made me think I have a fuelling issue once I�m on closed loop.... last week I managed to find a p1115 code, so am now wondering of the lambda is not working correctly.... but surely it would have fallen over completely by now and given me a CEL?? Because they are cheap I�ve ordered one for pre-cat, but could also do with what the resistance should be as I�d like to check the one I remove. Oh and a few years back I thought I have narrowed it down to an injector so swapped the suspect one out but again not a change. Ive also got a cheap OBD so took some screen shots incase you guys know what you�re looking for, and a vacuums pic on idle just for reference and it�s looks stable enough to me....... so anyone out there with opinions then chime in. In the longer video loads then you can see the inconsistent steam once it�s off choke on the right hand side exhaust. Likewise on the other video you can see it�s solid as a rock on choke.

Second video. Shows right exhaust caughing

Did u put new injectors or o2 sensors in with the rebuild? If not i would replace all of them and also do an intake smoke test. your scanner is not really telling us anything. We probably need to see all the o2s, fuel trims and cam deviations. A durametric would help u.

nope only changed one injector and none of the O2 yet. Have a new pre cat 02 sensor, but I'm thinking I should change both on that side as well. i have been trying to work out if a proper code reader is worth the money as I'm in the UK and Dura are very expensive,.

I credit you with a scientific approach, but this is over the top. The DME uses mainly the MAF sensor and the O2 sensors to determine how rich or lean to make the engine. There is no choke so don't get things confused. I suggest you replace all O2 sensors and the MAF sensor. After that, disconnect the negative lead on the battery for 10 minutes to reset the DME. Then go for a 30 minute drive with complete warmup and operating temperatures so the DME can check everything out.

Cool thanks for the feedback. Interesting you mention the MAF... I assumed that would give both banks an issue? Also had seems some communication about the EVAP but again figured that would be an issue across the engine rather than one bank. When i say choke, I guess what i was meaning was that on initial start up fuel gets poured in regardless of 02 sensor feedback right? SO to your point I wonder if an 02 sensor is having a brain fart. Im of the opinion they are probably 20 years old so are pretty much toasted either way

I think you have a dead O2 sensor heater & I'd deal with that first. P1115 Oxygen Sensor Heating Ahead of TWC, Bank 1 - Open Circuit Test conditions battery between 9.5V and 16V, time elapsed after engine start 10 seconds or greater. 1) Check connector for corrosion 2) Check sensor heating. Disconnect connector. Check resistance between pins 3 and 4 on sensor, should be 3.0 to 3.5 Ohms at 20C. 3) Check voltage supply. Start engine, measure voltage between pin 3 (car side) and ground, should be battery voltage. 4) Check wiring, stop engine, check resistance between pin 4 and pin II/9 (on the DME), should be 0-5 Ohms. These cars only run open loop for the first 10 seconds. It's all closed loop from then on, even at WOT.

great detail thanks mate. Will replace o2 and also run the voltage tests. really goo dot have the numbers to check for

Just be careful throwing parts at it. This method can become way more expensive then getting a decent diagnostic tool or paying a shop 2 hrs of diagnostics. I think u r on track with a single or multiple lazy o2 sensors and possible the maf too. Was that replaced? Its a part that should be factory or i think its oem bosch. The maf can do weird things to this motor. Im pretty sure this is all u need and im sure they ship accross the pond. https://www.pelicanparts.com/More_In...BoC82wQAvD_BwE

Thanks, your right they ship overseas. Still significantly cheaper then UK sourced, even after freight... so have ordered done. Can also use it on my 958 Pig so its not a bad investment i guess.

Anyone know what the brown cables are ground for? Just found them loose when I was inspecting..... Found a loose ground and unsure what they are serving. Either way I�ve tightened it up, bu am interested to know what the two brown cables are serving as a ground for

OK so manage to replace the knackered old Lambdas, also took the decision to swap out the AOS as I had some oil in the plenum, and didn't want to confuse things with a dodgy or failing AOS. While doing the AOS I and also found a loose coil earth eyelet connecting to the head... probably work hardened, so sorted that. So basically a few issues that all needed sorting, so time well spent. Got the Durametric software installed (epic) and most of the faults have cleared now, so happy days, but still a bit of a lumpy/rough idle. Took a screen shot and band one looks miles out right? Took your advice and tested the vario cams while it was running and activating B2 made the idle really lumpy for a few seconds, but pressing the B1 didn't have any effect, and based on the screen shots from Durametric, I would say I have an issue with the vario.... Any ideas how to do more tests while in the car? I locked the car in TDC and popped off the cam cap and the timing mark looks bank on to me so don't think timing has slipped.... Only thing i can think now is that the vario is stuck open. Would welcome your thoughts guys

So we need to your cam deviation numbers. Also injectors can cause a lumpy idle. The fuel trim numbers r good for diagnosing if u know how to read them. Flat six has a few vidoes in this. U can also leave the duremetric hooked up and data log the variocams. They should jump at 3400-4k depending on the year of your car.

Thanks for the pointer. I�m new to the durametric so learning as I go, but the screen shot I did above was showing 20 on the cam deviation or am I reading it wrong? Reason I ask is because my understanding is that the limit is +/- 6 so 20 is miles out of wack!!!

U r not looking at the correct thing. Someone with a durametric will have to guide u through it.

You are correct that the reading for "Cam Position 1 deviation" should be less than 6, so 20 is way out there. Did you take this reading after the car was fully warmed up? If so, you should buy a timing tool and actually check your cam timing rather than look at it.

Thanks mate. I do have a timing tool and was sure it couldn�t be 20 out without a catastrophic failure haha. So out of curiosity I popped the cam caps, locked the crank on tdc and used the timing tool. It�s bang on. So that makes total sense cos 20 is a mental number to be real. Basically leaves me with vario or the cam timing sensor. Next up is connectors and testing for resistance and voltage. Just need to research what the reading should be and then worst case I�ll pull the vario and test that. I�m actually relived cos a timing slip or a real 20 reading would have been a real concern. If you know how to test a vario or cal sensor that would be magic.

A 20 is possible if the variocam chain between the cams slipped a tooth. Not sure if thats possible but with a failed variocam and very worn pads it might be? U can test the seliniod with a 9v battery and see if it clicks but it should only engauge at 3400rpm so i doubt its your problem. The oil passages in the varicam itself can clog and cause failures. it kinda looks like u r not getting all the data from the dme. How long has it been since u disconnected the battery? It took a while the last time i had the bat discoed for all the numbers to come up, maybe 2 warm up cycles. It its been a while maybe disco the bat and see if a hard reset helps?

Isn't there some odd setup that only uses one cam position sensor?

Quotes source post: 17317870

Interesting you mention that. Aren't some ROW/Euro setup cars that only use one camshaft sensor? I thought I read that somewhere.

Just to clarify, so you have a 3.6 motor? I didn't see that posted earlier but you said that you "tested the variocams" which I'm assuming you mean you activated the solenoids which change the lift using the durametric. Assuming yours is a 3.6, you have two solenoids which affect the cam. The lift solenoid which it appears you activated and the actuator which changes timing. You might watch the specified and actual inlet cam position using your duramteric as you change rpm to see how well those track each other. Using the durametric to actuate the lift solenoid at idle should tell you if its working correctly. Another option is to switch the solenoids between bank 1 and 2. They are a bit of pain to get to on bank 1 but if you lower the engine down to the sub frame (like following instructions for changing AOS), you can access the two bolts holding the solenoids in place.. If it is an issue with a solenoid, your problem will follow the swap. I'm sure there is a more efficient way to diagnose this so hopefully someone who is more proficient at diagnosing this can chime in.

Correct, 3.4L USA LEV cars have two cam sensors although they are on the same signal wire (DME figures out which is which bank), ROW ( Rest of World) 3.4L only have one cam sensor on bank 1. So it makes it more difficult to check vario-cam function and to diagnose issues..

Sorry guys my bad should have said. It�s a 3.4. Tooth slip on the cams is an option and pretty simple to check. I see what your saying about the vario and 9v but because the engine tone didn�t respond on the durametric I wondered if it was stuck open give a dodgy cam setting for idle. The cam sensor by the AOS on mine so I�ll have a quick look and see if I can test or identify any oddities and then take it out for a cycle or two. I disconnected the battery overnight while doing the AOS so assume that was long enough ? Not sure if it help at all, but when I cleared the codes it seemed to idle perfectly for about 5 seconds and then got lumpy again.....

If you have a 3.4, you do not have a lift solenoid. That's only for variocam plus (3.6). I'm not sure what you are actuating with Durametric. I'm surprised that even shows as an option.

Quotes source post: 17318039

well this explains y there r no cam 2 numbers. yes that should be long enough to reset the dme. Since u just did the aos i would inspect the plug and wiring for the soleniod and cam sensor. Also the 3.4l soleniods are on/off not variable so if u got no change then u r in the right track if it is stuck open or shorted they have been known to burn out the transistor in the dme. There r some threads with pics so snapping off the dme cover may be wise

Quotes source post: 17318039

@Porschetech3 Seriously? Maybe this explains why I can�t get bank 2 deviation ratings on my USA �99 that had a motor replacement at some point early in its life. The car starts without depressing the clutch pedal and also passes emissions without a functioning secondary air pump. So maybe it�s a ROW engine through and through and not simply a DME flash?.....

Quotes source post: 17319769

Well its entirely possible but not very probable. The 99C2 USA cars with DME 5.2.2 had a hard time displaying bank 2 cam values regardless,, 99C4 , 2000 and 2001 with DME 7.2 were much better at displaying bank 2 cam values. With you passing emission with no SAI, you definitely have an ROW tune it...

this is great help gents. thanks. Ill get to measuring resistance tonight on the solenoid and believe it should read 13ohm so that's a good starting point. The engine is lowered so I can get the cam sensor out and bench test that. If I come up blank ill pop the scav pump off and check the cam dots align with the silver chain links as i could have ballsed that up during the rebuild I guess. IM pretty sure the Solenoid is the issue as i would have though it would have made the idle worse when i activated it on Duremetric, rather than having no effect. ITs a Cabriolet, so i am assuming the DME is same location as coupe? Is the solenoid is shorted i will defo check the transistor. Have you guys ever test with a 9V the vario in position with the engine off? Wanted to know if it should click when in static or if the car has to be running

Engine off is fine for the test.

Ok just spent an hour on the garage doing some tests and couldn�t get the vario to click. So pulled the cam cover and took the solenoid out. Not sure weather I�m pleased or upset but once I had it out it was clicking away happily with the piston moving 3mm in and out. What I did notice was that the plunger was set with the indentation facing the plunger (as the manual appears to show) but on other threads I see folks saying it�s smooth side to the solenoid.... see images below. Also there are no seals on the shaft but I�m assuming that�s correct. Any thoughts on plunger direction would be greatly appreciated.

Spent the evening trying to find mages of the plunger and found a conversation saying the spring sits in the recess..... in which case the plunger was reversed and the centre dot is the spring end! doesn't explain the 20 cam dev but does explain the vario not working on the durametric test right? I checked the cam dots against the silver links and its all good so thats not an issue. Does the Vario sit actuated with 12v when running and then go off at 3.5k revs or the other way around? Vario Cam Adjuster Plunger question re: P1324 code - 986 Forum - for Porsche Boxster & Cayman Owners

Be careful with your ansysis of assembly. The bank 1 and 2 are opposite because of the rotation of the chains. Im not sure if that pertains to the plungers but i know the pads r opposite.

Quotes source post: 17321653

ahhhhg *****! I didn't know that. the devils in the detail hey! I'll do some more research if that's the case and make sure I didn't make more errors on the vairo cos i changed the pads on the rebuild and don't remember seeing bank 1 or bank two labels when the parts arrived..... The fact thank bank two is running perfectly has to be good news though so i cant have missed the parts up that badly.... fingers crossed. Is there an easy way to check i have the right one in B1? and haven't mixed them up?

Where the guys educated me. There is another link in the thread with more info. https://rennlist.com/forums/996-foru...-backward.html

Those links were really useful. So latest summery is I checked the chains and timings perfect, found the plunger from the solenoid was reversed (must have been my fault), checked the guides and they are correct, checked the vario part number and it is defo the correct one for bank 1. found a small piece of plastic in the cam sensor plug which meant it wasn�t driven all the way to click in.... so basically two issues, the plunger and the plug......all I can see I have to do now it put it back together and see if one or both of the errors go away.... the only other option I have is to take a risk and report the whole vario and that�s an expensive gamble hey... unless you guys have other ideas? if it�s still saying deviation of 20 after a couple of drives, then I guess I have to assume the vario assembly is jammed open and that�s the cause of the lumpy idle. will revert back tomorrow and let you know the outcome

NEWS FLASH - it�s only bl00dy working!!!!! De Jeeper, if you were closer I�d buy you a beer mate! I've been chasing the lumpy idle since I rebuilt it 5 years ago. so as a ref the plunger must have been revered and casing the huge deviation. It�s running like a sowing machine! I could scream with euphoria. think I�ll do the clutch and flywheel next. Can�t be in a situation where I don�t have a project on the go with her!

Man second thead this moning that i am added value. i think ill go for an hour or so drive this morn to celebrate.

Quotes source post: 17325038

I'll vote for you on the "Most Improved 996 Rennlister Award" !! I remember when you first started posting, you were very comfortable with a wrench in your hand, and had a great understanding of dynamics and mechanics of automotive functions, but your 996 knowledge has grown considerably also with it your willingness to share/help others.. I celebrate you my friend !!!

Quotes source post: 17325173

Thx. I read just about every thread just to increase my knowledge base. I also perfer to help on the questions that go unanswered for a while. Sometimes i just need a mental challange.