ECU fault codes - 34 knock sensor

the new symptoms are reliable and wont go away. the car is fine and is running better than ever for the last few days except:

1. i am prevented from revving the engine up at all, as soon as it gets hot. it idles perfectly, but goes whap..............whap.............whap..........whap instead of revving up. and this is past idle. so from about 1000rpm it does nothing - even if the accelerator is completely open.

so i hauled the ecu out to check for fault codes. i got code 34 on mode 3. this is the knock sensor. so i followed the instructions for checking the sender. the sender is fine, and the continuity of the cable seems fine.

so i reset the computer and unplugged the knock sensor this time. the car ran like it did before, and once it got hot (about halfway on nissan temp gauge - and about 100-110 degrees on my guessometer) it stopped me from going over idle at all.

any ideas - what about unscrewing knock sensor, and wiring it up, but keeping it out of the engine block?

i am very close to getting rid of this problem the ol'skool way!!!!

please help if you can.

you need to keep the knock sensor plugged in at all times... the sensor uses a shielded wire, be very sure that the signal wire does not short with ground (I had this problem).

If the knock sensor is not connected it throws the ECU into a different map (for safty so it over fuels and retards engine).

My personal opinion i think you might have an ignition problem spark at greater RPM, its easy to get spark at low RPM I think your ignitor or coils are dying down higher up.

The wobbly RPM can be numerous things with the stock ECU it uses a whole bunch of things to try and stabilize RPM, I'll personally just double / tripple check all the connectors.

VC I am not pretending to know what your problem is, this is what I found, I had a miss on about 4500 revs, it didn't want to rev higher because it started "spattering" or spitting whatever you want to call it, this is after I readjusted my spark gap from 0.6 to 0.8( BCR8ES) bacause the plugs showed me that the car is overfueling and the plug was not burning the carbon off.

I then replaced it with other ones I had but set the gap to 1mm, then it started to miss on idle and a slight miss on higher revs, I then went down to BCPR7ES and put my gap on 0.8, she seems fine now.

My point is, try hotter and colder plugs with different gaps, maybe you get lucky, I know I did, before I never had a problem with the BCR8ES plugs but that was because of the 0.6 gap, all that went wrong was overfueling, car drove like mad....then played with gaps and stuff and things started going wrong, I think somewhere you are loosing heavy spark.

Sorry if I typed alot of shit but these stupid little things made differenses to my car like the restrictor washer in the vent pipe that caused smoking on idle, was at turbo guys, tuners blablabla, took the washer out gone.

Small things keep chekking and see the resolt of changing something.

thanks guys - the more shit the better draco.

but you guys missed the fact that the car is running sweetly now. i have no wobbly idle, and i have no misfires.

i did change the plugs draco, back to the ones i had before the service. these are the ngk bcp6, and they were gapped to 0.8mm. so its going exceptionally well until it get warm, well, lemme say hot rather.

as soon as its hot, something fucks out and wont let me rev. i've heard toby saying that coils can give problems that only manifest when they are hot, so i originally thought it was this. but if there was a spark problem i would get a fault code on the ecu for the power transistor.

i have a fault code for the knock sensor - and as far as i know - that means that the knock sensor is faulty. with a faulty knock sensor the ecu will go into a different map like vlade mentioned. but whatever map this ecu is going into, doesnt even let me go from 1000rpm to 1100rpm. it idles beautifully, but goes whap whap whap above idle.

i dunno.

i have 3 spare coils - maybe i will try those this evening when i get home, but i hate that engine at the moment.

Ya when it is fine again you will love it again...it's a twisted world.

hang on, if you guys know please clear this up for me.

a fault code on the ecu means that a sensor or actuator or solenoid or something is faulty, NO?

or does it mean that the engine has detected a problem (like engine knock) and therefore has a different map loaded.

i think its because the ecu thinks my knock sensor is faulty, but i may be wrong?

why do you run such small gaps? :?:

What boost do you run daily?

because i was having little issues when the standard gaps were installed.

changed it and it seemed to run much better.

while i was testing coilpacks, i was bitten on my hand by the spark from a spark plug, and to tell the truth - i didnt feel strong at all.

i always thought that it would throw me across the garage, but i didnt even let go of the coilpack. i just stood there, asking my appi to turn off the engine.

my appi says that once the same thing happened with his golf, and his hand hurt for some time after that!

i dunno - maybe those coils are too blame, but wouldn't i get a fault code on the ecu for the transistor then?

and why is the fault code for the knock sensor there then?

i've heard toby saying that coils can give problems that only manifest when they are hot, so i originally thought it was this. but if there was a spark problem i would get a fault code on the ecu for the power transistor.

No you wouldnt

i've heard toby saying that coils can give problems that only manifest when they are hot, so i originally thought it was this. but if there was a spark problem i would get a fault code on the ecu for the power transistor.

If i understand you correctly it sounds just like when i ran while the sensor was disconnected. The car ran ok, just hard to pull away, but with revs it was fine...albiet a little underpowered.

The ECU only picks up serious faults on sensors. it basically has to be dead.

You could try fooling the ECU into thinking the knock sensor is there. A 470K resistor should do the trick.

Daniel wrote:

You could try fooling the ECU into thinking the knock sensor is there. A 470K resistor should do the trick.

i thought about this too, but i wasn't sure how to do it. thanks daniel.

so get the fault code, and i assume the sensor is stuffed. so i perform the test for it as per the manual, and it is apparently fine. go figure. it is a nissan after all!!!

if i clear the ecu's memory and start the car - the fault code is displayed immediately. but the revving up problem only starts after about 15-20 minutes of "spirited" driving!???!

while i was testing coilpacks, i was bitten on my hand by the spark from a spark plug

BWAAAHAHAHA 😂

I was hit by the coil while checking timing. It wasnt bad at all...so i dont think thats an accurate way to test coils either 🙄

i guess its not an accurate test - but i always thought that the number of volts from a coil should hurt a bit when you get shocked!!!

its supposed to be close to 40 000 volts isnt it?

i mean an electric fence is only 10 000 colts!!! sorry, volts - but thats funny enough.

The ECU also has a real time mode. I suggest that you drive the car while cold & get it warmed up & as she starts getting hot get a dude to watch the ecu or get somebody to drive the car while you watch the ecu & see the codes that come up.

The car goes into a limp mode if there is a problem with your AFM. maybe the heat from the engine bay & stuff causes problems on a failing AFM.

Just a thought.

WooHaa wrote:

The ECU also has a real time mode. I suggest that you drive the car while cold & get it warmed up & as she starts getting hot get a dude to watch the ecu or get somebody to drive the car while you watch the ecu & see the codes that come up.

The car goes into a limp mode if there is a problem with your AFM. maybe the heat from the engine bay & stuff causes problems on a failing AFM.

Just a thought.

its a good thought - we never got around to do the realtime test last night. will do that tonight.

i thought it might be the AFM too, but when the problem reared its ugly head, i unplugged it and the engine just died immediately. so i dont think its that.

A knock sensor fault will limit boost to 0.5ar and probably run a different timing map however the caqr will still drive normally.

U suspect a fuel issue ie: running lean.
While the car is in the warm up phase it runs a richer mixture untill hot when all temperature compensation is inactive.

Some more info would be helpfull. Does it emit black smoke while its going" whap whap"? If so it will indicate a rich mixture if no smoke it may very well be lean.
Stick a gas analyzer in the exh or a WB sensor and see what the mixtures are doing.

thanks calibrator.

i didnt think that the car would not be driveable with a knock map engaged. its very strange - i've never seen anything like it.

we have checked the fuel pressure while the car exhibits the symptoms, and its normal. the spark seems to be ok, because it idles fine.

its basically at any point above the throttle resting position that it will not rev up. also the "whap whap" symptom (i have no other way to describe it) is one that is very regular. it has a beat, a rhythm. whap then pause then whap then pause. while its "whapping!" it sounds like a normal engine running on four cylinders. the pause part sounds like an engine turning but not combusting any mixture.

caver - would you be willing to do a dyno tune without keeping the car for too long? i mean like about 2 or 3 working days?

A dynotune or piggyback normally takes a day unless there are issues.

The Calibrator wrote:

A dynotune or piggyback normally takes a day unless there are issues.

Just not tomorrow or even Friday! 😉

why sqwerty? you got him all booked up then????

lol - i think i might have found the problem. well - i hope so. found a 12v supply wire that goes to coil pack #1 was spliced, then something added (probably immobiliser or something), then that something was removed.

in order to reconnect the wire after it was removed, the fukkas solder bare copperstrand to each end of the splice and twisted them together. that was then taped up.

but it has a home right on top of the tappet covers!!! so it gets hot, the tape's glue melts and it unravels itself, and then the 12v makes intermittent contact with the grounded tappet cover, shorting the circuit out.

i tried shooting timing while the problem was there, and during idling you can see the strobe light flashing, and when you rev up the timing light only flashes momentarily - and only when the engine makes the "whap" sound like its about to rev up.

but the strobe stops flashing completely, until it goes whap again!!!

so its like this "whap (flashing strobe) ... nothing (no flashing strobe)... whap (flashing strobe)... nothing ( no strobe) and so on and so on!!!!

but i will remove the engine harness today, to see what other nasty surprises lie waiting for me.

i used to think it was murphy - now i'm begginning to seriously dislike the previous owners, and previous owner's mechanics, of my car!!!!!!!

found more cockups with my harness!

its obviously tired from the 270 odd thousand kilometers, and from being on top of the cylinder head and near to the turbo.

the insulation on many of the wires is totally cooked. its brittle and tends to break right near the connecting plug. that means there is exposed 12v wire right next to exposed ground wire on coilplug one for instance!!!

and there are a few plugs like that. the water temp sender plug was only connected with one single strand of copper, the rest of the wire had broken off.

all of the exposed wire had corroded too, and were a green rusty type of colour.

basically it offered no more heat or electrical insulation!!!

once i've rewired, i really hope that the problem goes away, otherwise we can have a bring and burn braai. we can cook a couple of steaks on top of my burning ca18det!!!