odd problem

[InsanityInc]

Well, my idle is a little funny like I said before, but that's only part of it. Basically, the idle will slowly come down to 800, and then when it gets there the engine will become pretty shaky. You can feel the shaking from inside, and you can visibly see the engine rocking about. Also, a smog test showed the car running rich. I've replaced just about every sensor on the car. The general problem is that the drivability is a bit rough, such as coming off the acclerator for a while, then getting back on makes the car sputter pretty noticably. Also, the strangest problem is that every time I shut off the car for any length of time, then go to go somewhere, this is what happens: 1st gear feels fine all the way to 3000, I shift to second, and as soon as I get up to just about 2500rpm in second gear the car sputters and hesitates pretty badly, then you give it a bit of gas and it's fine, it never does that again until you shut off the car for a while. About the only thing I have to go off of is that one time the car threw a knock sensor fault code, but I replaced the knock sensor, though it was one from another engine and not a new one. Also, the computer threw an injector leak code once, and the #4 injector doesn't seem to click as loudly as the others when I put a screwdriver to it. Also, if I watch the timing with a timing light when it's idling rough, the timing seems to bounce around a bit which coincides with the roughness (ie, the engine jerks a bit and at the same time the timing gets a little screwy).

EDIT: oh, and I've swapped the MAF for another one, no real change.

[NISTECH]

When you say you failed smog for being rich, which number reading are you reffering to?

[InsanityInc]

CO and HC are high, so I'm guessing I have a rich condition. NOx was also high, strangely enough. Though I think my EGR system was a bit screwey.

[NISTECH]

typically when all 3 readings fail it indicates a cat failure. What the numbers were actually helps asses the failure better for me. But having both CO and HC high does indicate rich condition but generally in that case NOx would be low due to the excess fuel cooling the cyl a bit. If you have your numbers to post as well as you specs it could be much more helpful.

[InsanityInc]

typically when all 3 readings fail it indicates a cat failure. What the numbers were actually helps asses the failure better for me. But having both CO and HC high does indicate rich condition but generally in that case NOx would be low due to the excess fuel cooling the cyl a bit. If you have your numbers to post as well as you specs it could be much more helpful.

I think HC was 90/130 or some such and CO was like 2.5% on both 5mph and 25mph. NOx was like 600/1100. I've got a 240sx with a KA24DE with a header, 3" cat, 3" catback and an intake. Do you think it means anything that the #4 injector isn't clicking as loudly as the others? It's a very noticable difference in volume; you can barely hear it.

[NISTECH]

Are you in california? I know othe states read it differently. But here we get maximum, average, and measured.

You may be hearing the other injectors click through the rail and #4 isnt doin anything. If it doesnt fire at all you would have a lean condition on that cyl. but the others would come up slightly to try to obtain proper fuel mixture. You may have a dead injector causeing SOME of your problem. But if all three readings are in the failing range I think there is more to it then just that.

[InsanityInc]

Are you in california? I know othe states read it differently. But here we get maximum, average, and measured.

You may be hearing the other injectors click through the rail and #4 isnt doin anything. If it doesnt fire at all you would have a lean condition on that cyl. but the others would come up slightly to try to obtain proper fuel mixture. You may have a dead injector causeing SOME of your problem. But if all three readings are in the failing range I think there is more to it then just that.

Well, I know my EGR valve was screwy, I've since replaced it. Also, the #4 injector definitely isn't completely dead. Unplugging the harness makes the engine WAY rougher. The wierdest part is that I've replaced just about everything on the car. I haven't gotten a smog test since I replaced the most recent things (PCV valve, EGR valve, and that new MAF I tried) but I should be doing so soon.

And yeah, I'm in CA. There were max/avg on the readout, but what I posted were the measured numbers, since that was what my car was putting out.

[NISTECH]

How is your O2 sensor? If unplugging your # 4 injector is having a serious effect on your engine I seriously doubt thats your problem.

[InsanityInc]

How is your O2 sensor? If unplugging your # 4 injector is having a serious effect on your engine I seriously doubt thats your problem.

Just replaced the O2 sensor actually. Tested it with a multimeter. Seemed to fluctuate between .735v and .45v or so fairly rapidly. Couldn't it just be a slight problem with the injector though? I guess it could be my knock sensor though. With my old one that was definitely bad I had similar symptoms...

[InsanityInc]

went and checked the voltage on the MAF as per the FSM. Warmed up at idle it's like .88-.92v, and with the ignition in the on position, it's .245v. Says it should be approx .2v at the on position, and approx 1.0-1.4 at idle. Could I have a MAF problem?

[NISTECH]

Knock sensor wont effect your idle. It only effects the advance of timing under accel. If the sensor is functioning it wil pull the timing back when it detects detination. On the other hand if it is not funtioning it will have a slightly less advance under accel [kinda a safe mode]but will not stop detination or preignition. The actions and reactions of the knock sensor can hardly be noticed under normal driving conditions if at all.

On your MAF check the ground circuit with the car runing and the maf plugged in. You should see no more then .02volts on the ground side. If its higher then that splice and ADDITIONAL ground to that wire and run it to your intake manifold where the other ground lugs are. Clean those with a wire brush as well while your there.

[InsanityInc]

Alright, I'll try that. Also, my old knock sensor was so busted that it was detecting "knock" at idle, and the timing would do all kinds of wacky things.

[InsanityInc]

Oh, also, how should I spice it, at the wire or the connector itself, or does it matter?

[InsanityInc]

Yeah, the ground measures .029v, so I'll try splicing another ground.

[InsanityInc]

well, I soldered on another ground and that dropped the voltage on the ground to .002v, but it didn't seem to have any drastic effect on the way the car feels or idles.

[InsanityInc]

Also, the car is very jerky while engine braking with 0 throttle. Don't know if that helps you any.

[NISTECH]

I reread through your problem and a few things stood out [your injector code you had, your very high CO reading during the smog test,and your concern about #4 injector] Here is where I think you should go next.Get a fuel gauge on your car and do a injector leak down test. See if you have a leaking injector causing this engine to run rough.This is done by setting up your fuel gauge between the fuel filter and the fuel rail. You tee it in. Turn the key to the on position[dont start it] let your fuel pressure build, should pop up to around 40psi. After 5 sec's the pump shuts off. watch your pressure. It should not drop more then a few lbs when the pump shuts off. If it drops more then 10 to 15 lbs in only a matter of ~20sec's you have to determine where the pressure loss is coming from. You do this by repeating the test and pinching hoses one at a time till you have isolated different areas where the problem could be. start with the line between your fuel gauge and the fuel filter. If your leak down is no longer present you have a bad check valve on the pump. If the leak continues then you repeat the test and pinch the return line. If the leak down is no longer present then you have a bad pressure regulator. If the leak continues then you have a leaking injector(s).

[InsanityInc]

I reread through your problem and a few things stood out [your injector code you had, your very high CO reading during the smog test,and your concern about #4 injector] Here is where I think you should go next.Get a fuel gauge on your car and do a injector leak down test. See if you have a leaking injector causing this engine to run rough.This is done by setting up your fuel gauge between the fuel filter and the fuel rail. You tee it in. Turn the key to the on position[dont start it] let your fuel pressure build, should pop up to around 40psi. After 5 sec's the pump shuts off. watch your pressure. It should not drop more then a few lbs when the pump shuts off. If it drops more then 10 to 15 lbs in only a matter of ~20sec's you have to determine where the pressure loss is coming from. You do this by repeating the test and pinching hoses one at a time till you have isolated different areas where the problem could be. start with the line between your fuel gauge and the fuel filter. If your leak down is no longer present you have a bad check valve on the pump. If the leak continues then you repeat the test and pinch the return line. If the leak down is no longer present then you have a bad pressure regulator. If the leak continues then you have a leaking injector(s).

Well, I know the fuel pressure is good, as I had to get an adjustable fuel pressure regulator after I installed my walbro 255lph pump, so I doubt there's anything wrong with either of those. I also had the fuel rail off the other day, and it seemed like the #3 injector had a slight leak, so I replaced it with a new injector I had lying around. The #4 injector didn't seem to be outright leaking, but I'm thinking that it might not be operating correctly, or at the worst, the ECU might be sending bad signals :\. I'm going to try putting my MAF back on, as opposed to the new one. The new to me one I put on looks like it's in worse shape than mine anyhow.

[InsanityInc]

I'm going to try throwing in another injector to #4 and see if that changes anything. I've gotten rather proficient at swapping injectors at this point. I'm also going to probe the connector and test the voltage the ECU is sending to the #4 injector as opposed to the other ones, though I really hope that isn't the problem....

[InsanityInc]

also, by the way, are s13 and s14 KA injectors interchangable?

[InsanityInc]

well, it seems like the #4 injector is actually clicking fine, I just couldn't get a good angle on it to my ear before, so I left that alone. I really have no idea what the problem could be. I've replaced literally every sensor on the car. TPS, CTS, MAF, KS, CKPS, O2S, and every part of the fuel system, too.

Also, I tried unplugging my O2 sensor just for kicks and when I drove it around it didn't seem to run any differently at all, or if nothing else it seemed like it maybe ran a little better with it unplugged. Could I have a bad ECU? I really have no idea what the problem could be, it's always done this.

Another thing it does is if you push in the clutch and hold it at pretty much any RPM level for 30 seconds or so, every few seconds the engine will pop and stutter like it's misfiring or something. If I watch the engine vacuum when it's doing this, the vacuum spikes down when it misfires.

Oh, and I checked the pressure again just for kicks and it sits steadily at 34psi for quite a while after the fuel pump shuts off (let it sit for a few minutes and it didn't really drop much at all.)

[InsanityInc]

Another wierd thing: if I try to do that O2 sensor test via the ECU (rotating the screw on the ECU while it's running), it doesn't do anything. I get no readout at all on the dash on the CEL or the ECU, but if I hook up a mutlimeter to the O2 sensor, it gives off a fluctuating voltage.

[NISTECH]

If you have a wiring diagram for your car locate the signal wire for the O2 at the ecm. Back probe that wire and see what voltage your getting there. Does it even resemble what you were getting at the O2.

[InsanityInc]

Ok, I got the O2 check thing to work with the CEL and whatnot... not sure why it worked this time and never did before, but whatever. It blinked, but did so pretty erratically. Not sure if that's normal. The car's operation definitely seems unaffected by the O2 being connected or not though.

I'll check the voltage at the ecu.
Modified by InsanityInc at 4:42 AM 8/19/2005

[InsanityInc]

yeah, the voltage is the same at the ECU. Though, when I was messing around down there trying to get at the damn thing so I could probe it, it seemed like wiggling the mass of wires around would somtimes have an effect on how the car would run. Should I try probing some other devices at the ECU? Also, it doesn't seem to throw the code 33 for the O2 sensor being unplugged.

[NISTECH]

Start playing with that harness down there. You might be onto something with it effecting the cars idle. You could have a bad/corroded connection down there.

[InsanityInc]

Start playing with that harness down there. You might be onto something with it effecting the cars idle. You could have a bad/corroded connection down there.

well, I've had it off and nothing looks obviously corroded. I live in CA, so it's luckily not a big problem where I live, but there could definitely be a funky wire or something. It has to be something at the ECU at this point simply because like I said all the damn sensors are new.

[NISTECH]

What part of california??

[InsanityInc]

Folsom, it's semi-near sacramento

[InsanityInc]

Well, I tried checking a number of the voltages at the ECU, and they all pretty much check out. The only thing that was a bit funny was that the TPS voltage went from .8v-4.5v, when it should go from .3v-5v Not sure if that's causing any major problems. Also, the CTS voltage was reading about 1.1 when it was near operating temp, not sure if that's normal. I also couldn't seem to make it run better by wiggling anything this time.