Audizine - An Automotive Enthusiast Community

Results 1 to 25 of 25
  1. #1
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    Question [SOLVED] electrical fun: the engine shuts off (B7 S4)

    Guest-only advertisement. Register or Log In now!
    SOLUTION: golly, gang, it was Old Man Fuel Pump Relay all along! see the solution update.

    howdy,

    my 2007 S4 Avant (manual, 113k mi, owned 4 years) hasnít really had any electrical problems until this winter. iíve already done a fair bit of diagnosis, but i still donít feel like i have a solid answer. iím going to be doing some valley pan work (gaskets + oil check valves) and i want to know if thereís anything i can try while iíve assumed the position.

    the symptom is that the motor just dies without any drama. the first time it happened there was a brief CEL and some codes, but since then there hasnít been any CEL action. ODIS, however, shows a whole bunch of faults every time it happens (more on that later).

    the symptoms in detail:

    • the motor dies without any unusual sounds, the revs just drop and the battery light turns on.
    • the lights donít seem to dim, the radio keeps playing, headlights remain on.
    • at the worst it was happening 5-6 times a week, occasionally twice in the same day.
    • the first few times it happened it was always within a few minutes of a warm start, when traveling in a low gear at street speeds.
    • later, it also happened a few times within a couple minutes of a cold start, usually while still warming the car up or maybe 3-4 feet down the driveway.
    • it takes several minutes for the car to start up again, but it will turn over normally in the meantime.
    • turning off high-draw electrical devices seems to help, but itís not a guarantee (could be spurious).


    the conditions:

    • the big thing i noticed is that it started when the weather was cooler (35-50F at night) and it would usually happen in the morning or evening, when itís chillier.
    • it hasnít happened in over a month, as the weather has warmed up.
    • itís always at low RPMs, probably <2.5K.
    • it has never happened at high RPMs or on the freeway.


    diagnostics:

    • the battery and alternator test out fine, and most fuses seem OK (i havenít dug around for the obscure ones).
    • battery voltage doesnít seem to sag excessively when turning over (Optima red top, about 1 year old).
    • i havenít tested the battery or alternator at the moment of failure.
    • thereís no unusual parasitic draw.
    • the ground on the firewall in front of the brake reservoir seems OK, but i havenít checked others.
    • the coil pack harness and injector wires are snug and connections look good.
    • the one OBD instance and ODIS diagnostics (the software VAG-COM emulates) pretty much always say the same things.


    so hereís the meaty stuff - DTCs from ODIS:
    Address: 01 System name: Motronic Fuel Injection and Ignition System ME 7 Protocol versions: KWP2000/TP20
    P003700 (P0037): O2 Sensor Heater Contr. Circ.(Bank1(1)Sensor 2) Low
    P005700 (P0057): O2 Sensor Heater Contr. Circ.(Bank2 Sensor 2) Low
    P026100 (P0261): Cylinder 1- Injector Circuit Low
    P026400 (P0264): Cylinder 2- Injector Circuit Low
    P026700 (P0267): Cylinder 3- Injector Circuit Low
    P027000 (P0270): Cylinder 4- Injector Circuit Low
    P027300 (P0273): Cylinder 5- Injector Circuit Low
    P027600 (P0276): Cylinder 6- Injector Circuit Low
    P027900 (P0279): Cylinder 7- Injector Circuit Low
    P028200 (P0282): Cylinder 8- Injector Circuit Low
    P142100 (P1421): Secondary air inlet valve-N112 Short circuit to ground
    P142500 (P1425): Evaporative Emission (EVAP) Canister Purge Regulator Valve -N80 Short circuit to ground
    P143500 (P1435): Secondary air injection pump relay -J299 Short circuit to ground
    P147200 (P1472): Leak diagnosis pump - EVAP emission ctrl system Short circuit to ground
    P157200 (P1572): Left Electro-Hydraulic Engine Mount Solenoid Valve -N144 Short circuit to ground
    P157600 (P1576): Right Electro-Hydraulic Engine Mount Solenoid Valve-N145 Short circuit to ground
    short to ground! low circuits! fear and loathing!

    i realized something very key after sifting through several baffling wiring diagrams: all of these devices share a ground.

    i donít know whatís up with the electro-hydraulic engine mount solenoid valve errors, because i thought those were an RS4 thing, not S4. maybe whatever is pissing everything off triggers those ECU lines, too? or is ODIS labeling my engine mount sensors the wrong thing?

    possible causes?

    thereís one idea that seems fairly likely: a load on that ground circuit is shorting out, everything on it freaks out, and the ECU shuts down (thereís not much to do if all the injectors are gone).

    if so, whatís shorting? i suspect the secondary air inlet valve or injection pump relay because it dies early after a start. i donít know the SAI cycle in detail, but i imagine it changes state as the engine is warming up. it only happened when the weather was colder, so it may be that the coil in a solenoid or relay has some cracked insulation and the cold weather closes tolerances just enough to cause a short when energized.

    i donít suspect the EVAP system because that should be running all the time, right? and there are no misfire codes, so i donít believe itís an injector (unless coil packs only trigger misfire detection). i donít think itís a loose ground because it never correlates to a bump, driveway, rough road, or other jostling.

    another idea is that the alternator's voltage regulator or some other part of the electrical supply is very, very briefly dropping out (like, less than a second), which might look like a short to ground to some sensitive systems. however, i think there would be more widespread and varied DTCs in this case.

    so, does this sound familiar to anyone? are any of the involved parts easy to test or cheap enough to just swap when i do the valley pan work? can anyone fill me in on the SAI system and cycle? (aside from that it should be removed, iíve already heard that.)

    thanks,
    - emilio
    Last edited by emilio; 08-19-2016 at 11:22 AM.

  2. #2
    Senior Member Two Rings viridia's Avatar
    Join Date
    Apr 13 2011
    AZ Member #
    73886
    Location
    ham best meat
    Items for Sale

    I wouldn't dwell too much on those DTC unless you're absolutely certain they're caused by your specific event. You're clearing codes and these appear each time? Or, for instance, what faults would display if you purposely stalled the car?
    87 4k, 90 f350, 91 v8, 93 v8, 04 s4, 0a3 transmission

  3. #3
    Established Member Two Rings
    Join Date
    Feb 20 2015
    AZ Member #
    316344
    Location
    SLC, Utah!

    Sounds a bit similar to a problem I was having with my RS4 last summer, but the whole car would shut down, no power steering, etc, always at low speeds in city traffic. We tried a bunch of stuff and could not solve the problem- the car was throwing all sorts of phantom codes. About 6 months ago we replaced an intake runner flap solenoid and I have not had any issues since that was done and my car is driven daily. I have no idea if that is your same problem, but something to consider.
    '08 RS4 Cabriolet

  4. #4
    Veteran Member Three Rings IslandHydro's Avatar
    Join Date
    Nov 29 2008
    AZ Member #
    35790
    My Garage
    Sun EZ-Sport AX LWB Bent
    Location
    WA State

    I'd look for a ground lifting

  5. #5
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    hello, thanks for the responses.

    Quote Originally Posted by viridia View Post
    I wouldn't dwell too much on those DTC unless you're absolutely certain they're caused by your specific event. You're clearing codes and these appear each time? Or, for instance, what faults would display if you purposely stalled the car?
    yes, i pulled and cleared OBD once and DTCs at least three times, and they all said the same things (plus those errant window sensor DTCs that always seem to pop up). that's an interesting question about DTCs when stalling out; i'll give that a shot today.

    Quote Originally Posted by Blake-O View Post
    Sounds a bit similar to a problem I was having with my RS4 last summer, but the whole car would shut down, no power steering, etc, always at low speeds in city traffic. We tried a bunch of stuff and could not solve the problem- the car was throwing all sorts of phantom codes. About 6 months ago we replaced an intake runner flap solenoid and I have not had any issues since that was done and my car is driven daily. I have no idea if that is your same problem, but something to consider.
    hrm, that does sound related. i haven't seen any DTCs for the intake, but it could be the same sort of issue on a different circuit: a relay or solenoid's coil energizes, there's a short or low resistance, and the ECU bails.

    Quote Originally Posted by IslandHydro View Post
    I'd look for a ground lifting
    i'm not sure what would lift in such specific circumstances; i'd expect that to be related to hitting a bump or something. maybe if there was some sort of heat expansion that only takes effect when it's cold enough and early in the drive... it might be a stretch, but i might as well see if i can find where that ground is.

    - emilio

  6. #6
    Veteran Member Three Rings q_dubz's Avatar
    Join Date
    Dec 16 2008
    AZ Member #
    36417
    My Garage
    A quick S4
    Location
    EAST COAST
    Items for Sale

    mine does the same but only for a split/minute/hair of a second


    I can't replicate it and it only seems to happen once every few thousand miles.
    I don't listen to what they say i can't do. ;) #willraceforcookies #gapcity #gapcounty

  7. #7
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    Quote Originally Posted by q_dubz View Post
    mine does the same but only for a split/minute/hair of a second

    I can't replicate it and it only seems to happen once every few thousand miles.
    does it just cut out for a moment but keep on going after that? that might actually be a ground lift (i doubt it in my case) or even just a slightly loose connector somewhere. has it ever thrown a CEL or DTC from it?

    - emilio

  8. #8
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    well, i tried stalling it out to see if any DTCs popped up, but nothing did.

    i guess i'll start investigating these parts, where they are, and how much they cost. i'm still going to do the oil check valves and related gaskets, so maybe there are some things i can swap around there. otherwise, the engine pull is going to have to wait until i find a new job!

    - emilio

  9. #9
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    update: it happened several times in a day after some rain, but it wasn't particularly cold. i figured it was time to look under the ECU, since i've read reports of water ingression and corroded fuse sockets.

    so this weekend i went through the incredible trouble of removing the ECU cover, including cracking my windshield when removing the insanely suck wiper arm. inside, i found no problems whatsoever. not a streak of water residue, not a smudge of grease or speck of pollen, not a flake of corrosion. all fuses looked perfect, but since i'd read reports of people having an ambiguously "bad fuse" under there (wtf is a "bad fuse"?) i swapped all the 15A and 20A fuses just to see. the challenge is that it's so intermittent and not reproducible it could be months before i see if it's fixed.



    i seriously doubt it was a "bad fuse" (whatever that is). it could have gotten cold enough that night to trigger the problem, but i think the rain is interesting since it's happened mostly in the cold, even when dry. i can't really find any signs of significant water build-up under the cowling, though there definitely are some places that look like water was around (i'm pretty sure water is supposed to drain through there, anyway). i cleaned out some leaves and such, but i don't really have any better idea of what the problem could be.

    - emilio

  10. #10
    Established Member Three Rings VinnysS4's Avatar
    Join Date
    Nov 12 2015
    AZ Member #
    364349
    Location
    Oregon

    what i would suggest... Try taking each of those circuits that the codes are throwing and seeing if you can disconnect each part and see if that helps. i'm suspecting a bad component that is causing the short to ground. Sometimes one problem can present itself with other un related codes. Pretty common. I have actually had a "bad fuse" before. The metal in the fuse was cracked and for the most part remained in contact/continuity. every once in a while, and god knows what caused it, the fuse would break the continuity in the circuit and (cant remember what circuit it was on) i would lose functionality of said circuit. This poses a few problems. one being the obvious lack of use from said accessories, and the other was the resistance that is caused in the circuit. Its like having a bad ground. yes it does have continuity, but there is not a good metal to metal contact point. In order for the component to function properly, it pulls extra juice. This is caused by the resistance in the dirty/corroded/cracked connector. This can cause a multitude of things to go wrong, especially in a can bus system.
    Try the old train of through and look for a short to ground. There are several methods of doing this. There is a guy on youtube with a channel called Schrodringer's Box. Look up some of his videos. he has one or two that may help. The problem you will face is replicating the problem and associating it to a certain component at that exact moment in time. I wish you the best of luck. i'm no expert, but if there is anything i can help with, feel free to PM me.
    good luck!
    Vinny
    1997 VW GTI VR6 (sold), 2003 Audi RS6 (sold), 2010 Audi A4 2.0T (RIP Totaled on 11/2/2015 ) 2005.5 Audi S4

  11. #11
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    thanks for the response, Vinny!

    yes, since they're all on the same circuit i've sorta been dragging my feet on buckling down with the multimeter until after i tried some easier things. well, after cracking my windshield to look at non-existant issues i'm about ready to do the tedious stuff, lol. at least i know one place to start: a while ago i disconnected the ground at the firewall, in front of the brake booster, and the insulation crumbled. i put some heat-shrink over it, but there could be an intermittent short if it's crumbling further along with some other wire in the bundle to the ECU. ugh, wiring harness work.

    as silly as "bad fuse" sounds, i still think it's a possibility given that it happened (mostly) in the cold. some kind of microfracture on the fuse or wire, just enough temperature change or vibration at the right part of the engine starting to cause a break... it's plausible. that's why i figured i'd at least try swapping those fuses in case there was something.

    i also looked at the cost to replace the solenoids & valves on the circuit, in case it's a short in one of those, but that would be way too expensive to do on just a hunch (>$500).

    i just need some car-sized temperature testing chamber... there's gotta be one around the Silicon Valley! i wonder if Tesla will rent me some time in one?

    thanks again,
    - emilio

  12. #12
    Veteran Member Three Rings q_dubz's Avatar
    Join Date
    Dec 16 2008
    AZ Member #
    36417
    My Garage
    A quick S4
    Location
    EAST COAST
    Items for Sale

    the code i would always throw was terminal 30 low voltage intermittent.
    I don't listen to what they say i can't do. ;) #willraceforcookies #gapcity #gapcounty

  13. #13
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    so i did a bunch of diagnostics today and came away with the likely culprit.

    the problem happened several times this week, despite mild and dry conditions. yesterday it happened and was severe enough the i couldn't start the car again within 10 minutes (i usually can). twice it started for a few seconds, but ran a little roughly and then died again. luckily, i was about 50 yards down the street from my house and got help to push it into a parking spot. today it started up and i pulled it into the garage.

    there were more of the same codes as above, plus a couple more (MAF, more SAI stuff). i dug into the DTCs and looked at the mileage of the last occurrence, and all the SAI stuff reoccurred again few kilometers after everything else. then sifted through the Bentley, ODIS, and Etka to put together a picture of what the problem could be; i briefly tested the evap system via ODIS since it was easy, but nothing failed. on to the secondary air injection system.

    testing the SAI electronics boils down to three components: the secondary air injection pump (V101, near the alternator), the secondary air valve solenoid (N112, lower front of the intake manifold), and the SAI relay (J299, under the ECU). the easiest to get to is the solenoid, so i tested this first.

    it's possible to disconnect the solenoid in place if you're familiar with the connectors and have the right tools. i pushed down on the connector to relieve the pressure, used a long probe to open the latch, and then pulled out the connector with some angled pliers - easy. the amount of work i've done to the front of this engine...

    this solenoid should normally be around 19-28 ohms. the engine was warm when i was testing, and the solenoid measured zero. hours later, after cooling down, it measured 24 ohms. bingo! there's the possibility that the solenoid is damaged because of some other issue that's causing the car to die, but i think the difference in resistance with the temperature is a pretty clear link.

    so i think this is what happens: the SAI is normally supposed to run for a period after a cold start, but if it's been driven there's a period where the engine is warm enough to short the solenoid but the exhaust has cooled enough to trigger the SAI circuit. if the solenoid is switched on while shorting, all other components that share the same or an adjacent ground freak out and the ECU shuts off for its own safety.

    the solenoid is not crazy expensive (079906283, $25-50) but it is on the front of the intake manifold and would need service position to replace (sigh), or even removing the intake manifold... and i just had the intake manifold out last week. i hope i don't have to take it out again, because you know what that means:



    - emilio
    Last edited by emilio; 05-07-2016 at 10:41 AM. Reason: clarity

  14. #14
    Established Member Three Rings VinnysS4's Avatar
    Join Date
    Nov 12 2015
    AZ Member #
    364349
    Location
    Oregon

    Lol. I really hope this is the case. That should make for an excellent diagnosis and track down. Keep us posted!
    Keep up the good work
    Vinny
    1997 VW GTI VR6 (sold), 2003 Audi RS6 (sold), 2010 Audi A4 2.0T (RIP Totaled on 11/2/2015 ) 2005.5 Audi S4

  15. #15
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    sigh well, disconnecting that solenoid stopped the car dying... for a while. it's still disconnected, but the car died when warming up this past weekend. there were many of the same DTCs, including the SAI relay.

    i'm 99% sure i properly tested that solenoid to be bad, so my paranoid hunch that there are other problems is correct. the next easiest & cheapest thing to replace is the SAI relay (US$10), which is still a pain to get to (i've already cracked my windshield). i really hope it's not the SAI pump, which is pricey (>$200) and a bit of a pain to get to. at that point i might just consider deleting the SAI.

    the big question is: what caused the original short? are multiple things failing on one circuit because one has damaged another?

    - emilio

  16. #16
    Senior Member Three Rings Lucca M's Avatar
    Join Date
    Jan 17 2014
    AZ Member #
    139266
    My Garage
    Past Bmw M3 coupe Euro 1998
    Location
    New Jersey
    Items for Sale

    Do sai delete and get rid of that shit. :)
    2004 Black B6
    JHM LWFW, Stage 4 Clutch, Short Throw Trio, IM Spacers, JHM 91,
    034 Street Density Engine, Transmission, Snub Mount & Zero Tolerance Bracket, Rear Diff Inserts,
    PowerFlex Diff insert, 2.5" 200 cell downpipes, Neuspeed 2,75" catback, Cross Drilled/Slotted Rotors,
    Rns-E, B7 steering wheel, BBS CH 18x8.5, ST Coilovers, Ecodes and many many more...

    Insta: luccamentone

  17. #17
    Veteran Member Four Rings xhackerekx's Avatar
    Join Date
    Jul 23 2007
    AZ Member #
    19741
    My Garage
    04S4B6_07A8D3 08Q7_03A4B6_VN670
    Location
    Scranton , PA / Poland UE

    Quote Originally Posted by Lucca M View Post
    Do sai delete and get rid of that shit. :)
    or buy my work excellent for $50 +shipping
    u need help with immobilizer when u installed new or used ecu or cluster included correction millage ? I can did this for you .

  18. #18
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    i was waiting for someone to make that suggestion - but that's making the assumption that the SAI system is the source of the problem, which i'm still not sure about. plus, ripping out those SAI valves would be way more work than i'm already dealing with. plus, i'm in California and that shit is never worth the trouble (there is zero advantage to removing SAI besides tidying up the engine bay a tiny bit).

    - emilio

  19. #19
    Established Member Two Rings
    Join Date
    May 01 2015
    AZ Member #
    330530
    Location
    Mont Clare PA

    You're getting a whole lot of fuel/evap system DTCs from what I'm seeing... have you checked your low pressure fuel pump/fuel filter? Thats what caused my car to shut off and not want to start back up/run. If it did start, it would just die out shortly after.

  20. #20
    Senior Member Three Rings SprintBlueWorld's Avatar
    Join Date
    Sep 16 2013
    AZ Member #
    123267
    My Garage
    Chrysler T&C LE, Honda Accord
    Location
    Pa
    Items for Sale

    Do the symptoms rule out the the Crank Position Sensor? These are known to fail over time especially in hotter weather. Also the above post, fuel filter or pump is considerable. Especially the filter easy to replace cheap part could make a big difference.
    RS4 Pedals, KN Air Filter, JHM 93 tune, JHM Intake Spacers, Stoptech Slots & SS Brake lines, JHM 6-Rib Pullys, 034 Snubby Bracket, Dimple/ECS Oil & Diff Magnetic Drain plugs, Apikol Diff Mount, 034 Carrier Mount Inserts, Avante Garde M310's, Catless DPs, Corsa RSC Cat-Back, 034 Sway Bar & End Links

  21. #21
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    howdy gents, thanks for your replies.

    i've been digging around a lot and i think i've made some progress. to answer your question, i haven't seen any problems with fuel delivery and all the evap codes are electrical (low/short to ground) versus flow (gross leak, insufficient flow, etc.); the fuel filter was replaced a couple years ago, and i always run major-brand top-tier gas. however, there's likely some relation to the evap system... more on that later.

    i doubt it's the crank position sensor since all the affected items are on the same or similar circuits that would much more logically be explained by an electrical fault, rather than a mechanical one that just kills the whole engine. i believe there's some computer test or data that can be used to gauge the quality of the crank position sensor, so i may do that just to check it.

    so, on to the diagnosis: things were getting worse for a while - much worse. a couple times the car died and took >20 minutes to restart, including one time when i left it at a market and returned a couple hours later. this was all after completely disconnecting the SAI valve and pump relay, so i can rule those out.

    so it was time to look at everything that was throwing codes, disconnect most of them, and start reconnecting them one-by-one:

    • secondary air valve: disconnected and the issue persisted, now reconnected.
    • secondary air pump relay: disconnected and the issue persisted, now reconnected.
    • evap purge valve: disconnected and the car ran fine, reconnected and the car died shortly after two cold starts.
    • evap leak detection pump (LDP): disconnected and the car ran fine (without the evap valve), reconnected and the car still seems fine, even with a cold start. i'm not sure if the LDP will ever be triggered if the ECU detects that the purge valve is an open circuit, though.
    • lower/downstream O2 sensor heaters: disconnected and the car ran fine (without the evap valve), still disconnected.
    • all fuel injectors: on a separate circuit. the ECU pretty much exists to control the fuel injectors, so it's not surprising it throws a fit if something electrical takes down the ECU. i doubt it's them, though, since i did once remove the fuel injector harnesses, cleared the codes, and received the same DTCs with them all disconnected.
    • mass airflow sensor: on a shared or similar circuit, not yet diagnosed.
    • brake controller: a very sensitive component on a separate circuit, and because of that sensitivity it's not surprising that a code shows up after an electrical disruption.
    • electro-hydraulic engine mounts: these are usually on a shared circuit, but do S4s even have these? i thought they were RS4-only. if so, this is a ghost DTC and it supports the idea that the ECU is freaking out over something on a shared circuit.


    of course, this didn't all come easily. fun fact about the leak detection pump: there is no dedicated fuse for it. it's on the same circuit as a bunch of other stuff - all the stuff that's failing - so you can't just take the LDP offline by pulling a fuse or relay. there's literally a wire runs from the LDP/reed switch connector all the way to a connector at the ECU harness. so not only is it a pain to diagnose it, there's nothing protecting the ECU except for whatever is built into the LDP:

    doin' it raw (V144 is the LDP, J220 the ECU, box 28 runs to the One Fuse to Rule Them All):



    so, then to disconnect the LDP you have two options: disconnect at the pump, or pull pins from a wiring connector near the ECU. to pull the pump's connection you have remove the driver's side rear wheel and strip out the fender lining, so i opted to disassemble a connector. this was probably only slightly easier than digging into a wheel well, but it was definitely a lot cleaner.

    poppin' pins for the LDP and reed switch (the red connector is T17d in the above diagram):


    rear O2 sensor with heater pin removed (FYI the car gets super bitchy when the downstream O2 sensors are completely disconnected, which is why i just pulled the heater pins):


    after all this, though, it may be that it was one of the easiest connectors to get at the whole time: the evap purge valve, located right on the back of the airbox (i'm still not sure this is it, but it's the best possibility). i made two mistake in diagnosis: the big one was that i was disconnecting things while the car wouldn't start and assumed that it would start again right then; if it still didn't start, that must not have been the problem, right? wrong. the ECU likely has some resettable internal fuses that can take anywhere from 2 minutes to +4 hours to reset. during that time it didn't matter what i disconnected because the ECU was probably still pissed off. so early on i tested the evap valve while the car was dead, it didn't start, and i eliminated it as a possibility. it wasn't until i decided to take a step back and start from scratch that i moved past the second mistake:

    the second mistake was to disconnect one thing at a time when i was starting out. most of these systems are activated shortly after startup some of the time: evap (it tests fuel tank pressure, then might pressurize tank or purge the canister), secondary air (runs for 65 seconds when the O2 sensors are cold), and possibly the O2 heaters (i'm not sure wether the O2 sensor heaters are always active or just when the sensors are cold). the lesson is to do what VinnyS4 says: if you have a bunch of stuff on one circuit that's freaking out, disconnect everything and gradually reconnect them one by one, then in combinations (in case it's a circuit load issue versus just one component). it was much easier to rule out things when they're the only one on the circuit. unfortunately, i couldn't take the whole circuit offline because it also runs the ECU... i seriously might rewire the ECU to have its own fuse, plus fuse the LDP line while i'm in there.

    there's a third, minor mistake: i probably mis-tested the SAI valve. i thought i tested it twice at 0 ohms when hot, and i thought i was being really careful with the probes, but i haven't been able to replicate those readings. i guess i'll be returning the one i bought!

    another lesson is to give sporadic issues time. when reconnecting things i made sure to give each of them a couple days, starting at different times and temperatures, before eliminating them as a possibility. i really wonder what a dealer or mechanic would have done, because it can take hours or days for the problem to resurface. the evap purge valve tests fine with an ohm meter (if that's really the problem) so they may have just ended up at "it's a bad ECU." unfortunately, i still consider a flaky ECU a possibility, but i'm willing to put the time in to make sure it's not that pricey little box!

    - emilio
    Last edited by emilio; 06-01-2016 at 11:26 AM.

  22. #22
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    i fixed it - for real, this time! so it didn't turn out to be any component of the SAI, evap, LDP, or injection systems. (i actually fixed this a couple months ago, but waited to be really sure... and then forgot to post the update.)

    the problem was the fuel pump relay. i replaced it with this 70A relay from Amazon (8E0951253 , aka 614, aka J17, aka 7M0951253A). this is the 70A relay with four blade contacts.

    the major problem with diagnosing this is that the fuel pump relay doesn't throw any codes. it runs right off the ignition switch and the ECU has no idea what it's doing, it only knows that status of what's connected to it, which leads to...

    the other major problem is that this runs a shitload more than just the fuel pump. "fuel pump relay" is such a poor description, it really should be called the "half the engine relay".

    so, what does "fuel pump" relay J17 actually power?
    • the fuel injectors
    • Intake Manifold Tuning (IMT) valve N156
    • Secondary Air Injection (SAI) valve N112
    • engine mount solenoids N144 and N145 (do the B6/B7's even have these, or are they RS4 only?)
    • Mass Airflow (MAF) sensor G70
    • evap purge cannister valve N80
    • exhaust flap valve actuator N220
    • Leak Detection Pump (LDP) V144
    • Secondary Air Injection (SAI) pump relay J299
    • ALL of the Oxygen (O2) sensors: G130, G131, G39, G108
    • oh, and the fuel pump G6



    "D" is the ignition, "J17" is the FP relay, and the numbered boxes run to the rest of the engine.

    after tediously disconnecting and reconnecting most of the items in this list i found that different components would kill the car. this led to two possibilities: the ECU is dying, or there's an overload condition (too much power draw causes the problem). i decided to go back to the wiring diagram and scrutinize it more, and then i noticed that while many of the devices throwing codes shared a ground, all of them are powered by the FP relay. i would have had this wrapped up much sooner if i had looked at the diagrams more systematically. always thoroughly check your wiring diagrams!

    the fuel pump relay is kind of a pain to get out, though it's still nothing compared to getting under the ECU. if you pull off the fuse panel cover you can see the relay deep under the dash, just visible past the top of the fuse box; it should be a grey relay with "614" on it. you can't easily remove it through there, though; it's best to remove the footwell cover to get more working room. it's still a pain, of course, and i had to do a lot of pulling and prying with small hooks and screwdrivers to wedge it out of the very tight relay socket.

    since replacing it i haven't had a single problem keeping the car running. (i still get touchy startups, but that's likely the starter solenoid.)

    the worst part about this isn't really the labor, or the parts (i returned ones i didn't need, and the relay was <$10), or even getting stuck in a couple parking lots. nope, the worst part is that i cracked my windshield getting to the ECU one time. so really all this took to fix was a $10 relay and a $500 insurance deductible... well, at least my windshield was already pitted and i was planning on replacing it, just not now, heh.

    - emilio
    Last edited by emilio; 08-19-2016 at 09:06 PM.

  23. #23
    Veteran Member Three Rings q_dubz's Avatar
    Join Date
    Dec 16 2008
    AZ Member #
    36417
    My Garage
    A quick S4
    Location
    EAST COAST
    Items for Sale

    weird, I've replaced mine with no luck.
    I don't listen to what they say i can't do. ;) #willraceforcookies #gapcity #gapcounty

  24. #24
    Established Member Three Rings ven0m's Avatar
    Join Date
    May 10 2015
    AZ Member #
    332001
    Location
    Slovenia

    Thank you for the write-up and being persistent with the problem.


    In regards with not throwing a code: usually if you disconnect the heating-circuit for the O2 (which is ran in parallel with the low-pressure pump) you should throw a code. At least that happened to me.
    Jeremy Clarkson: "So when you were saying that it wonít slide, what you meant was, ĎI canít slide it.íď
    James May: "Yes."


    Audi RS4 b7 (Sprint Blue) - Recaro Bucket Seats - Flat Bottom Steering Wheel - OZ Ultraleggera 10x19'' - Michelin PSS 275/35/19 - Milltek Sport Cat-back & Piggie Downpipes - JHM Short Shifter

  25. #25
    Established Member Two Rings
    Join Date
    Apr 18 2012
    AZ Member #
    92081
    Location
    Bay Area, CA

    Quote Originally Posted by q_dubz View Post
    weird, I've replaced mine with no luck.
    hrm, what codes are you throwing? does the car just not start, or does it die after a minute?

    - emilio

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


    © 2001-2016 Audizine, Audizine.com, and Driverzines.com
    Audizine is an independently owned and operated automotive enthusiast community and news website.
    Audi and the Audi logo(s) are copyright/trademark Audi AG. Audizine is not endorsed by or affiliated with Audi AG.