Rb20Det Lag/bogging

[teteous]

Issue with my r32 rb20det skyline is that under acceleration it will accelerate between 2-3k RPM, bog down severely between 3-4k RPM and boost hard between 4-5k RPM and bog down again after that.

I have changed plugs/gapped, replaced and cleaned AFM, changed fuel pump/filter, confirmed correct timing.

Using consult cable I get a code 34 (Knock sensor - Not reading correct resistance) and no other codes.

I cannot find any vac leaks and everything is tight and connected..

I have a feeling because of code 34 it might be going into some type of limp mode but am unsure.

Also, when it bogs down the timing drops to 0-5 degrees..

Any feedback would be appreciated.

Edited June 19, 2013 by teteous

[GTSBoy]

It's probably switching back to the knock fuel and ignition maps (the standard ECU has a proper set and a set it switches to when knock is detected). And the R32 ECU also then has the ability to pull even more timing out if it continues to have knock while on the knock map. This will make it as doughy as all get out. So, if your knock sensor is faulty (and let's face it, you can be pretty sure that it is!!) then it may be registering false knock and making the ECU spaz out.

Replace knock sensor with new one, or with ~500Ohm resistor, and see how you go.

[mungy]

need more info on the car, mods, history of the car and what happened just before the problem.

nothing "just happens" something would have been noticably different about the car before hand

[teteous]

Bought the car in not 100% running state, been slowly fixing it as listed above and this is the point I have gotten to.

So no idea on previous history. Is running a r34 turbo/gtr radiator, turbo back exhaust, constant power feed to fuel pump, air con/heater removed, everything else engine wise is standard.

Heading to Jaycar today to get the resistors, will putting the resistors directly into the ecu pin outs for both sensors work?

Edited June 19, 2013 by teteous

[GTSBoy]

Only if you disconnect the wiring out to the sensors.

[teteous]

Only if you disconnect the wiring out to the sensors.

Ok, so i cut both 23/24, added the 470 ohm resistors to both and grounded the other end to a ground on the ECU.

It made a difference in the revs as soon as I did this as it was on a cold map (1700-2000rpm) and instantly dropped to hot map once the resistors were in.

But it still gives the knock sensor code 34 even after clearing, as soon as I start the car and the code re-appears.. Any ideas would be great.

[GTSBoy]

I don't know if 470 is close enough to the original sensor's impedence. Have a google around for what the best resistor value is. You may need to series or parallel another resistor on there to get closer.

[david69]

^{when I got my skyline it wouldn't rev past 4500.the previous owner changed the usual crap that everybody changes when they don't know whats going on.ended up the fuel filter sock was full of brown crap.Old tank with a lot crap in it. I cleaned the fuel pump sock then I changed the fuel filter,problem solved,its worth a try.cheers}

[teteous]

I don't know if 470 is close enough to the original sensor's impedence. Have a google around for what the best resistor value is. You may need to series or parallel another resistor on there to get closer.

Yea I was thinking that too so I read all the threads i could find and they either said 470, 500 (which isnt a actual resistor impedance) or 1M ohm.

I ran two of the 470's in parallel and it still has the code & hesitation (but reduced the idle weirdly), I guess I can try in series if that may make a difference. I guess next step would be to try another known working ECU..

^{when I got my skyline it wouldn't rev past 4500.the previous owner changed the usual crap that everybody changes when they don't know whats going on.ended up the fuel filter sock was full of brown crap.Old tank with a lot crap in it. I cleaned the fuel pump sock then I changed the fuel filter,problem solved,its worth a try.cheers}

I shall try that, I have another Bosch 040 laying around so I guess I can just swap that in and see if it makes a difference, it could be contributing to it

Edited June 23, 2013 by teteous

[GTSBoy]

Not. Not 2x 470s. A 470 and a something else to get you closer to 500. Like a 47, or preferably something around 30 ohms.

And there is always the possibility that there is a damaged track inside the ECU to consider also. If that were the case, you could hook up whatever you liked and get the same result. I don't know if the knock sensor circuitry/tracks in a Nissan ECU have EVER been found to be faulty, but there's always a first time. Have you tried another ECU, just in case?

[teteous]

Thanks GTSBoy,

I have limited understanding of resistors, but wouldn't running x2 470's actually reduce the value by half (ie 470/2 = 235k ohm).. So really I should get x2 1M ohm resistors & run those in parallel to make exactly 500k ohm (but twice the wattage?).

Or alternatively run the 470 with say a 47 as mentioned in series?

Edited June 24, 2013 by teteous

[GTSBoy]

Any two resistors in series simply add the resistance. So a 470 and a 47 = 517. That might be too high, so you could use a 30 instead of a 47 and get 500 exactly.

Two 470s in parallel gives you half the value of one of them (so you end up with 235 like you say), but this is useless to you. Yeah, you could get two 1M resistors in parallel, but I was assuming that given you already had the 470s it was worth starting there.

Jaycar's 510 ohm jobbies might do.

What would actually be smarter would be to get a variable resistor that works in your desired range and try adjusting it around the 500k mark and see if the fault code goes away. Could be tedious. But worth a try.

But one thing I think you have to be prepared for is that the problem may not be that the knock sensor resistance is actually out of range. There may be some other problem.

More swaptronics is also an option. Try to obtain another ECU and some knock sensors and see if swapping them out helps.

[teteous]

Thanks for all the info,

Got the 30 ohm & 1m ohm resistors (only $0.46/pack of 8 so why not..) and i'll test these out tonight. Hopefully being closer to 500 ohms will get it running again.

My mate is looking for a spare ecu, so if this fails i'll go down the swapping ecu's route.

[teteous]

Well running the 470 + 30 ohm resistors in series still has the Code 34 coming up, as well as running two 500 ohm resistors in parallel has the code 34..

This is making me believe it must be something to do with the ECU now as there is nothing to do with the knock sensors.

[bananaweesle]

how to you have the resistors connected u can use 1m ohm and connect both knock sensor wires together connect to one side of resistor then connect the other end of the resistor to a good ground. code should definitely go away after that.

[teteous]

how to you have the resistors connected u can use 1m ohm and connect both knock sensor wires together connect to one side of resistor then connect the other end of the resistor to a good ground. code should definitely go away after that.

Ill give it a go, I have pin 23/24 cut at the ECU loom which I soldered separately to each wire x1 470 & x1 30 ohm resistors in series. These both have the same ground of pin 50 (Ground control unit) - This is a known good ground as I have my SAFC connected to this ground also.

Edited June 26, 2013 by teteous

[DVS JEZ]

470k ohm resistor?

Just double checking. Also try 2 new resistors as i have seen them not work the first time

[teteous]

470k ohm resistor?

Just double checking. Also try 2 new resistors as i have seen them not work the first time

Ok ill try a couple more, I first tried 470's on each knock sensor, then tried 470 + 30 in series. After those failed I tried connecting both of the knock sensors wiring from the ecu together, attaching a 1M resistor and grounding that to the chassis but still have the same issue.

I've tried another ECU and same issue to rule that out.

[teteous]

Couldn't get any of the resistor mods to work in the end.

Just ended up replacing the knock loom and fixed the knock issue.

[GTSBoy]

Update for later readers. Please note that I have mentioned 470 and 500 Ohm resistors in previous posts a number of times when in fact they should be kOhm. 470000 Ohm and 500000 Ohm.