cutting out at high rpm's

[joe hurst]

Morning all from the U.K, I am hoping to find some helpful insight from my American cousins on this issue...

 

I have recently purchased a 2000 986s and have a problem that I have as of yet been unable to resolve...I will list the course of events so far and any input you have would be most helpful.

 

 

I apologise in advance for the rambling first post, and my use of 's' where you good folk would use a 'z'...;-)

 

OK, so the issue is that the car ''cuts out'' like hitting the rev limiter at approaching 6k rpm..

 

key off and back on again ( even whilst moving) and the car fires up and is fine..( under 6k) fuel economy was also terrible ( 17mpg) car is on 130k, has a FPSH up to 90k and then specialists until then, last service was at 128k ( minor) car is currently on 131k

 

No CEL.

 

when this initially happened I plugged in my generic scan tool and graphed, MAF, and Lambda's and found the bank 2 fuel trim as +32% and the Maf reading very erratically, this made sense as a likely culprit as it ''felt'' fuel related or DME stepping in as parameters exceeded.

Maf replaced ( genuine Bosch) bank 2 o2 sensor was replaced ( genuine bosch), all sensors now reading normally , fuel economy improved by 100% I thought I had it licked but alas on the next drive it did it again!

 

so I changed the in line fuel filter which was the original!..

 

I next did a full service, and checked with a mirror and torch the induction piping past the air filter, no blockages at all, I also took the opportunity to inspect and undo, re-positon and re-tighten all of the induction, I then tested for vac leaks at idle with the old carb cleaner trick.

 

idle remained and always has been rock solid at around and indicated 6-800 rpm. in fact the car drives VERY well until the point it cuts out.

 

next was a throttle body clean and replacement ( bosch ) spark plugs , the correct ones , the old plugs looked worn but fine, the result is a further improvement to the snoothness of the engine abut the main problem still persists.

 

all battery and earth connectons have been checked as well...and all look fine.

 

now onto yesterday, I had a thought about DME relays so switched around the type 53 relays in the boot , I of course disconnected the battery to do so, the result, well it revved out all the way!

 

once again I thought I had it licked, but the problem came back..I have now ruled out DME and DME relay and am thinking that the disconnection of the battery re-set some learned parameters which allowed the engine to rev freely, or that the problem is temperature related..? new temp sensor on the way but I really don't think its that..for £15 worth a shot though!

 

so my current thinking is to investigate rotational sensors ( cam 1 and 2 and crank) I don't think its crank, as heat related crank sensor faults would stop the car from re-starting straight away so I am moving towards the idea cam sensor 1 as now being the most likely cause as I assume TDC is calculated by the DME on CYl 1 ?

 

I am happy it's not a fuelling issue, the engine has revved freely and pulls like a train when it does so I am happy its not anything ''mechanical'' or timing chain or tensioner related

 

other elevant info, the alternator is new, ( fitted by previous owner) I have ruled it out as I have established the fault was there before this was changed having spoken to the Porsche specalist that maintained the car for the last 4 years..

 

I have also replaced the expansion tank ( horrible job),slave cylinder, reverse light switch, boot lock solenoid,removed the front bumper and cleaned rads and condensors..I have inpected drive belt, and tensioners and idler pulleys, all looks good, car does not overheat and is in fact great to drive under 6k ( ish ) rpm's

 

I have taken it too a friend with full porsche diagnostics and the only codes found ( and cleared) were the ones relating to MAF, and o2 sensors and have since been cleared.

 

no CEL has ever been present but it does work! lights up and goes out as it should on start up.

 

so any thoughts ? the fact it did rev out yesterday implies to me it's a sensor going bad either unabe to cope at higer rpm's or it's temp related but any further suggestions appreciated.

 

it is booked in at al local Porsche specialist week after next for a live diagnostic session with all sensors graphed whilst I replicate the fault.

 

In the meantime I plan to pull the fuel pump and check the tank and gauze filter for ''stuff'' and try and replicate being able to rev the car all the way to try and further pin down the culprit...I can graph cam and crank signals I think so maybe this will offer some further clues.

 

 

many thanks in advance for your insight and help

 

best

 

Joe

 

P.S compression test done whilst the plugs were being replaced, figures were as follows, 143,140,137,140,141,139

 

don't know what they should be but am happy with those figures in terms of being nice and close together

 

 

 

 

[wizard]

Do you know if the DME is the one which originally came with the car ... and is it manual or Tiptronic?

 

 

[joe hurst]

I cant be 100% sure but on inspection I would say fairly confidently it's never been removed, the car is a manual.

 

I can see where you were going though...?

Edited September 16, 2018 by joe hurst

[mikefocke]

There are two possible MAFs for a 2000 (in both Bosch and Porsche parts) and the latest one requires a ECU reflash to update the standard parameters if used as a replacement.

(see table towards the end of this) Was the new MAF the exact same part number as was on the one replaced? I presume the old one matched the ECU and at one time the pair worked together over 6k RPM.

 

By the way, what does FPSH translate to?

[joe hurst]

FPSH = full porsche service history..

 

thanks for the info on the MAF...

 

pretty sure Its the right one, also don't see how the maf would cause a complete stall, that feels like the engine losing timing...key off and back on again and its all good..worth unplugging it though to see if it still does it, problem is in limp mode I am not even sure it will reach the required rpm under the same load, so it may be a pointless exercise diagnostically speaking, also the car runs like a ''champ'' under 6k rpm, and I have had it over that point 3 times in a row after swapping over the boot relays at the weekend, I think if it were the Maf it wouls run rough across the range..

 

also the car was only just up to temp when I floored it, so I am also considering as said before a cam sensor breaking down when warm.

 

Mafs do some weird stuff agreed, but not shut down at a fairly consistent rev point...thats more a lack of ability to deliver spark and or fuel at the right moment.

 

I will investiagte though as it's another new and helpful piece of info, so thanks..?

[mikefocke]

Thanks. Added to Porsche Acronyms List. Model to model they change, location to location too.

 

I only mention the MAF possibility because suppose the ECU has some built in parameters that say at certain revs it should do thus based on the inputs it is getting from the MAF and the O2 sensors. If the characteristics of a different MAF give the ECU different readings at a certain airflow, then the ECU could get all confused and suggest the wrong air/fuel mixture.

 

Any codes? Got access to any Porsche specific diagnostic hw/sw?

[joe hurst]

no CEL, and at present no codes.

 

codes present were Lambda and Maf specific, cleared and not returned,however problem persists.

 

also the problem has stayed exactly the same with both original and replacement Maf, I think it may be worth mentioning to the specalist I have it booked in with, but honestly I think the maf is a red herring.

 

after it happens you can feel and hear the engine is all ''out of wack'' this clears after keying off,and keying back on again.

 

I understand where you are coming from but normally a maf issue results in erratic running if faulty, ( inconsistent or implausible signal) or cutting of fueling because airflow cannot be established.

 

what it wouldn't do is affect timing, although it could cut out, however it would be more like hitting the rev limiter with timing maintained rather than timing lost.

 

the more i think about it the more I think its a loss of timing...The fact it has several times gone all the way up the range implies its not mechanical..I am trying to re-create the circumstances in which it did this ( after switching boot relays) however I think that also might be a red hrring and that the battery disconnect is more of a significant component of that sequence of events and maybe cleared some learned parameters or pending codes which were then re-learned..subsequently causing the fault to re-occcur.

 

so either voltage to ECU is being interupted, (possibly fuel pump drawing too much voltage causing it) or when the car is warm a rotatonal sensor is going out of whack.. cam(s) or crank

 

if I had to surmise I have revised my initial thoughts on it being cam sensor bank one, I now think its cam sensor bank two, because crank shaft sensor failure would present as hot stall and cold start issue and with this problem the car re-starts everytime, even whilst moving...whilst it doesn't 100% rule out crank it implies a cam sensor.

 

cam sensor one would be ''more'' critical in ignition, I think the ecu could establish cam positions from cam one and crank perhaps until  very high rpm when cam sensor two input becomes critical.  idle is rock solid, and the car runs very well under the 6k mark, for all intents and purposes consider it a normal car up until 6k.

 

these assertions are all absolute conjecture of course ! I plan today to get a new 53 relay, put it in position one, measure fuel pump pressure,ecu voltage and see what I see.

 

sadly I can't read cam sensors, but I may throw a new one at bank 2 as a diagnostic exercise...I can also measure the voltage off the connector, and just a removal and inpsecton of the bank two cam sensor might reveal something.

 

it' a riddle for sure...but very much doubtful its a ''serious issue'' as the car would not ever have revved to the limit ( it has and pulled like a train) and wouldn't run so well under 6k..

 

thanks for input so far, will keep you all posted.

 

 

 

 

 

 

Edited September 17, 2018 by joe hurst

[wizard]

I have the ME 7.8 on a MY 2003 car.

 

From memory, there's a DME re-adaptation procedure following battery reconnection. I believe it's ignition on for one minute without starting the car : ignition off/key left in the switch for 10 seconds. Accelerator not to be touched throughout the process.

 

What effect that has I know not, other than something to do with e-gas perhaps, but there's nothing lost by trying the re-adaptation for the ME 7.2 (?). 

 

P.S. Subsequently found this in my file:-

 

DME 7.8
DME 7.8
PEIN2DME78
0
Printed in Germany – 14, 2003
Note on adaptation
The DME control module must perform a learning and adaptation
routine for the throttle adjustment unit if:
♦The power supply to the DME control module is interrupted
♦The DME control module plugs are disconnected
♦A new DME control module is installed
♦The throttle adjustment unit is replaced
♦The DME is programmed.
To do this:
1. Switch the ignition on for 1 minute without starting the engine.
Do not actuate the accelerator pedal (for instance, make sure
that there is not a carpet pressing on the pedal).
2. Switch off ignition for at least 10 seconds.
The following conditions must also be observed, otherwise learning
is not possible:
♦Vehicle is stationary
♦Battery positive voltage between 10 V and 16 V
♦Engine temperature between 5 °C and 100 °C
♦Intake air temperature between 10 °C and 100 °C

Edited September 17, 2018 by wizard
P.S. addition

[joe hurst]

still have the issue,

 

diagnostic road test this week, on the plus side the fault will not be difficult to replicate..

[JE 17]

Joe Hurst: glad you started this. Mine is doing exactly the same thing, hitting brick wall at 6k. I was going to just throw a fuel filter on it but you have elevated the game a bit.

Awaiting your test results!

 

[joe hurst]

41 minutes ago, JE 17 said:

Joe Hurst: glad you started this. Mine is doing exactly the same thing, hitting brick wall at 6k. I was going to just throw a fuel filter on it but you have elevated the game a bit.

Awaiting your test results!

 

sorry for your troubles, but glad im not alone...!

 

not maf,not lambda'snot fuel filter,or spark plugs, I think it's losing timing.

 

hence I think cam one ,two or crank, if not that then i'm going fuel pump, and finally DME or wiring harness suffering a voltage drop.

 

It's weird.

 

 

Question,

 

when yours does it if you leave it running is the timing all out of whack ? and soes it resolve on key off and back on again ?

 

maybe we can sorth this between us !

[JE 17]

I haven't checked timing.

I just bought this '99 about 10 days ago. I've only recently tried running up towards redline, and every time it dies out just past 6k. No codes.

Love the car, but this is a bit weird.

 

[Loren]

The car needs to be check with a Porsche specific scan tool like Durametric or PIWIS.

There are many Porsche specific codes that will not show up on generic code scanners.

[joe hurst]

I don't mean timing is out, I mean the DME loses signal either from critical sensors, ( crank and or cam position sensors)

 

I did once sort it briefly, I disconnected the battery and swapped the relays in the boot around, there are 4 of the same type, this cleared it for around 30 mins but it came back.

 

try it and see what happens.

 

I suspect when i disconnected to swap the relays around some learned parameters were cleared from the DME, or something VERY odd is going on with the DME or supply to it.

 

anyway.

 

1) disconnect battery

2) swap relays

3) re-connect battery

4) go for blat

 

see  what happens !

[joe hurst]

3 minutes ago, Loren said:

The car needs to be check with a Porsche specific scan tool like Durametric or PIWIS.

There are many Porsche specific codes that will not show up on generic code scanners.

Hi Loren,

 

yup, am aware of this, all set up with a local Porsche specialist.

 

will report back when /if it becomes clear what is going on.

[Ahsai]

No direct experience but I wonder if it has anything to do with the knock sensors.

 

[joe hurst]

8 hours ago, Ahsai said:

No direct experience but I wonder if it has anything to do with the knock sensors.

 

worth a shout as they have a part to play in timing adjustment along with cam and crank sensor, temp sensor , air intake temp sensor etc etc

 

good logic !

[joe hurst]

things have progressed,  I have it definitively down to one of two things...

 

 

sorry to be crptic but I don't want to reveal until i have a definite answer..and there are several routes to said answer...

 

JE17, I will have an answer for you in the next few days...!

 

best

 

 

Joe

Edited September 21, 2018 by joe hurst

[joe hurst]

OK, this is now SOLVED.

 

I took the car to a really good Indie, NOT Porsche but with excellent diagnostics, he agreed all of my tests so far had ruled out MAF, O2 sensors, I had changed the fuel filter, also replaced one fo the Lambda's and the MAF anyway.

 

after multiple live data sessions replicating the fault the mechanic in question, ( Lee at on the road car care in Chertsey) had it down to either Throttle body or the DME itself.

 

as there was a P1508 code i thought the worst and assumed DME, however a had a niggling doubt as it ran so well under the RPM where it cut out.

 

I ordered a used TB from Steve Strange at Douglas Valley breakers and fitted it this morning.

 

The car felt happier straight away, idle was a bit smoother but I put this down to battery disconnection , and subsequent resetting f learned parameters.

 

Idle had never been a point of concern but its smoother none the less, the engine felt a little torqier and as as the test drive progressed I started to feel a little more confident that something was indeed different.

 

so onto the acid test.

 

some quiet roads that allowed me to test whether I could get the car above 6k, when resetting the codes in the past it has let me do it, but only once or twice before the dreaded cut out came back.

 

this time I had the car all the way up the rev range 7 or 8 times in a row with no hint of an issue, I am considering it resolved but will of course keep an eye on it over the coming days.

 

a clue was that when the car cut out the data recording was showing a minus voltage on the TB, in fact exactly the same voltage it shows when unplugged.

 

so there was a short in the TB internals somewhere, TB loom was also inspected and confirmed fine.

 

JE17, that where I would start with your issues, by the sound of it it's almost exactly the same as mine.

 

best of luck with it ad report back and let me now what happens..

[Ahsai]

Well, you never told us about the P1508, which correctly pointed to your throttle body. Glad you got it sorted though.

 

BTW, just want to comment we should really trust the DME codes (most of the time). In this case it's easy for the DME to see the fault because there are TWO independent sensors (really potentiometers) inside the throttle body so if one of them acts up, it's obvious by comparing the readings of the two.

[joe hurst]

well..

 

yes and no.

 

becasue P1508 is potentially one of three things...

 

Maf,DME or TB, now seeing as the Maf and Lambda are related in terms of fueling adjustment I had to go there first...as it does and did feel ''fuelly''

 

finding bank 2 running a long term fuel trim of 32% confirmed that Lambda was borked that along with a flat line line on the sensor on durametic live data, fuel economy improved when replaced but cut -out not solved...I intially thought fuel supply to bank two was sufficiently differnt from bank one to cause the ecu to step in and cut power...

 

Maf was also ''suspect'' as reading oddly on both durametic and scope, changed and whilst it improved things it still never solved the problem I was hunting down.

fuel filter was also the original one! ( inline not in tank as its an early S) and at 130k miles that cannot have been helping..

 

The TB was reading normally until cut out at which point it gave the same voltage as it would unplugged, but that would also happen if the DME was at fault internally (confirmed with an ECU specialist in the UK) In addition P1508 is porsche speak for ''replace'' DME.

 

you also have an issue that it's difficult to replicate live conditions with the TB off the car, we couldn't see a specific fault even with an oscilloscope ! some oddness yes, but nothing you would bet importan body parts on.

 

My chasing things through was very useful from both a sorting things that needed sorting perspective, and getting to know Porsches, ( I'm an Alfa guy really) and sometimes things don't fail in an expected way.

 

The TB was my penultimate diagnostic/ fault solving step towards replacing the DME, as having ruled out and checked all over possibilities, loom etc, earths, battery cables, plugs,coils,packs etc.

 

I looked up symptoms and codes on forums and got multiple possibilities for all the codes thrown FYI there were Maf and Lambda codes too, and they are often present after the cut outs despite having changed both Lambdas and the maf, that would make sense as the misfires resulting from a ''TB disconnect'' would throw multiple codes.

 

as the car is very early S, it has variocam, not variocam plus and only one cam sensor and two rather than 4 lambdas which did make things easier from a fault chasing perspective.

 

not once did P1508 on any of my multiple searches bring up TB as likely culprit with symptoms like mine, ( hitting a brick wall at 5800-6000 rpm) in fact I found only one and Porsche told the guy to change the DME which he elected not to do, choosing to ''drive the car under 6k'' which is NOT a solution.

 

hopefully this thread will help someone else out as honestly, a month of searching, both online and speaking to Porsche specalists and doing my own investigations has not moved things along any quicker than they might have done by me just doing what was needed in order of most likely culprits.

 

diagnotically speaking given what I found on the history of the car and actually on the car, on the live data, and via the codes would have led anyone in the direction I have gone, only if you have seen exactly the same thing before with a clear resolution would you go TB first.

 

and I could find no such thread..

 

what seems to be the case with Porsche is less owners actually work on them , meaning a smaller pool of expertise and experience..with Alfa's 90% of owners on the forum self maintain, meaning collective knowledge is vast.

 

I suspect the Ferrari or Maserati owners sites are similar to Porsche in this respect...worth remembering as I plan on a Maserati GT next!

 

 

[Ahsai]

I misunderstood your post. I thought you were saying P1508 points to the TB per Porsche. I looked it up and the documentation of P1508 is thin. The only troubleshooting step is to replace the DME like you described. The only clue I saw now is the majority of the P15xx codes are related to the throttle body. Of course it's obvious now after your experience. It's not a very common failure for sure and good job tracking it down. I'm sure this thread will be useful for others in the future.

[joe hurst]

its confusing, as many P1508 codes are tracked down to the maf as TB, i suggest anyone with this issue unplug the maf and run without to see if the fault still occurs, if it does then there is your answer...It's NOT the maf.

 

things get even more confusing when you are getting multiple potentially related codes, at this point you only have one course of action, live data with post drive analysis and preferably fault replication thrown in, then remove implicated sensors and put them on an oscilloscope...

 

I did all of this and still couldn't definitively know whether the TB, or the DME was at fault...only diagnostic route left was to replace TB with good one.

 

In the end this was how I got my answer but Maf,lambda's, crank and cam, knock sensors, coil packs, HT leads, and supply too all of the above have to be checked first to comletely rule them out..as sometimes thins can be a combination of small problems combining too create a bigger one.

 

I always have one eye on this when diagnosing issues on a car.

 

you can take some guesses, crank wouldn't re-start until cool, rule out.

cam sensor fault would affect idle, rule out

knock sensors would retard timing and engine would ''ping'' before cut out..rule out if not happening.

maf..well anything can happen with those but unplug and replicate fault, if so then rule out.

replace lambdas, if they need it, cheap enough and was due on mine anyway.

 

its still a puzzle but not one I wish to engage in again.

 

not on a Porsche anyway!

[JE 17]

Joe, sorry to take a while to respond.

I haven't yet obtained the diagnostic program.

The biggest thing is...it's revving fine now! Of course, this could just be temporary, but a change in usage is what I believe responsible I had read elsewhere online that these engines dislike running below 3000rpm. I had been trundling through town in the low 2000s. Now, I make a conscious effort to stay above 3000 in those situations, and it seems to have taken care of the problem.

I hope.

Thanks for your efforts. I'll have a reference if that is not the case.

 

 

[joe hurst]

Mmm

 

I suspect it will come back, there is really no reason why an engine ''shouldn't like running below 3k'' unless engine diagnostics are adapting in an ''odd way''

 

there have been further developments this end as well,

 

it came back.

 

so after more reading of codes and live diagnostics and lots of research we have come to the conclusion that at some point in the cars life it has been run for a significant amout of time with a faulty throttle body..

 

we have found evidence from another fly by wire set up that faulty jacking motor or potentiometers in the TB can cause voltage spikes which damage the throttle drivers in the ECU/DME..

 

changing the TB will solve the problem temporarily as the car goes throught the adaptation process for the new part, after a period the fault will return...and in fact if you were to change only the ecu and leave the old TB in place then the new DME woud go the same way as the old one.

 

I will be able to confirm this as the following tests/steps are going to be undertaken.

 

1) swap out the pedal potentiometer ( it tested fine and looks fine under load but who knows ?) this is because the DME uses relative positions of pedal potentiometer and TB to establish desired engine torque/load ( so I'd like to rule it out)

2) re-set all adapted parameters and drive for several cycles with live diagnostics and see what happens. ( I don't think this will resolve it but its a logical step) if/when this fails..

3) I then have ready to go a New DME/Immobiliser kit and a brand new throttle body, this is my last resort, but if the problem hasn't been resolved with a pedal potentiometer and a Thottle body, Maf,  and Lambda's on the current DME then I think it's safe to assume the DME is at fault.

 

wiring has been checked and found to be good, I think our current hypothesis is a good one and having spoken to several ECU specialists they agree it does look very much like a recurring DME fault, add code P1508 and changing the DME is EXACTLY what Porsche would do when seeing this code.

 

I will keep you all updated. should have a resolution this week.

 

 

Edited October 8, 2018 by joe hurst