My Z is a sick puppy

[Tere]

Just got in from trying to troubleshoot, and it's the most puzzling thing I've ever seen.

Yesterday, I was having a dyno tune done (authorized ATI dealer/installer/tuner) and when I started it up to back it off the dyno, the whole instrument panel lit up like a Christmas tree and stayed lit. Eventually, all the lights except the check engine light (flashing on and off) went out, and it sounded like it was only running on five cylinders. The guy thought it was a bad plug on #3 because the header pipe was cool. We checked it for oil in the raditor and water in the oil -- nothing. The oil and water pressure are both good.

So I drove it home (carefully) last night, and then swapped out the plug on #3 with a brand new one today. During the swap out, I checked both compression and spark -- everything was good. Started her up and it was still running rough. So I checked all the pipes and #5 pipe felt cool. So I repeated the procedure on #5 and used the plug from #3 (plug was in nice shape). Compression and spark were good. Fired her up and it still ran rough. Felt the pipes again and #4 pipe was now the cool one. I went through the procedure again and used the #5 plug (slight carboning, but good shape). Compression and spark were good. You guessed it, it still ran rough.

Geez, I hate these all electronic cars. They sure can drive you to drink (I did) rapidly. It almost sounds like maybe the ECU is doing some strange science to the timing -- almost sounds like a car with a cam in it (cam lope at idle). And why does it appear like random cylinders aren't firing or are firing weakly?

At no time during the dyno was it ever leaned out too much, and the dyno A/F graphs and data sheets all looked good. We did 6 dyno pulls. On pull 3 and 5 the guy hit the rev limiter. Pulls 1-4 were done in 4th gear, and pulls 5 and 6 were in 5th gear. After pull 3 we put a big fan up front to keep the heat down, and we let her cool down a bit before doing the last three pulls.

My wideband O2 read richer than his the whole time -- mine is measured at the cat to header flange on the driver's side, and his measures at the tailpipe with a long tube that goes to the dyno machine. He did all the FMU adjustment using his wideband, so even if mine is more accurate, it would be running richer than his, so there's a pad there. On pull 5, the A/F went right to 12 on his data sheet at 3,600 rpm and held steady to 4,500, then tapered to 11.6 and held steady to 6,000 rpm. Then it did a slight rise to 12 which coincided with the typical 350 dyno hp drop at 6,100 rpm. It's a really nice A/F curve. I suspect something happened on pull 6 to cause the problem, but he didn't mention seeing anything out of the ordinary -- just said it was a bad pull.

Even stranger is the check engine light going on and off. ECU senses things peachy and turns the light out, then senses a bad condition and turns the light on???

So, anyone have any ideas? If I don't figure it out tomorrow, I guess I'll have to take it in on Monday. Sure wish I had a Consult II to read the codes!

[Boomer]

I think you may have boggled the ECU. Some of the real early posts had owners having to return to the dealer after the engine light came on and wouldn't go off. They went into a limp-home sort of tizzy and their dealer had to reboot the ECU or one of the other devices. I would bet the ECU, but the dealer has a special sequencing they have to do to get normal operation again, I think. Whatever it is, your best bet is the dealer, IMO. Good luck, Tere. Boomer

[Tere]

I tend to agree with you Boomer. I can't find anything physically wrong with plugs, heads, pistons, etc. And the fact that every light on the panel stayed on for a long time tend to point to the ECU -- for instance, it took about 2 minutes for the ABS and SLIP light to go out, and the car was standing still.

Oh yeah, here's another potential pointer to the ECU or O2 sensing. Normally at idle, the O2 should read about 14.7, which my wideband always did (actually varied from 14.3-14.7). Now my wideband jumps up to about 15.6 and the light goes on, then when it drops back to 14.7, the light goes out. At idle, the Aeromotive DFMU is doing nothing to the fuel, so it could be that one of the stock O2 sensors got toasted during the dyno runs and is giving the ECU some bad data which the ECU is trying to compensate for. There's a "chicken and egg" argument in there somewhere... LOL... I love this online brainstorming.

I think I'll run the ECU through the reset procedure today and see what happens. For sure it can't hurt anything to do the reset. It's the wildest thing I've seen and it really had me going nuts. The only cure for the insanity of course was a few Shiner Blondes (beer). Geez, I have to deal with whacked out computers at work all day, and now I have to deal with a whacked out computer on my Z -- gimme another beer... LOL... :D

[Dr Bonz]

Iwas going to suggest the same thing Tere. Did you do the ECU reset steps listed here:

1. Confirm that accelerator pedal is fully released, turn ignition switch “ON” and wait 3 seconds.
2. Repeat the following steps (2a and 2b) procedures quickly five times within 5 seconds.
2a. Fully depress the accelerator pedal (HARD).
2b. Fully release the accelerator pedal.
3. Wait 7 seconds, fully depress the accelerator pedal and keep it for approx. 10 seconds until the CEL starts blinking.
4. Fully release the accelerator pedal (while the CEL is still blinking)
5. Wait about 10 seconds.
6. Fully depress the accelerator pedal and keep it for more than 10 seconds.
7. Fully release the accelerator pedal (The CEL light will continue to blink).
8. Turn ignition switch to “OFF” position and now you can start the car. The CEL light should be gone.

If the CEL light continues to remain ON, repeat the above steps. Timing is EXTREMELY critical to resetting the ECU.

This is from the Technosquare site on resetting the ECU BYW.

Good Luck!

[NSANY]

Can't the ECU also be reset by just disconnecting the battery terminal for a bit? :confused:

[Tere]

Yep, disconnecting the battery is supposed to do it, but I'll just use the "Atari Method" which should ensure that everything is completely cleared. Wish I had an extra O2 sensor that I could play swap out games with -- as I recall though the left and right sensors have different connector patterns, so it's not that easy.

I'll let y'all know the results of the big ECU reset, and I'm going to look more closely at the A/F and check engine light patterns if the reset doesn't do it. A Consult II would make my life a whole lot easier at the moment.

[Dr Bonz]

I think you have to disconnect for like 12-24 hours though don't you?

[NSANY]

Overnight should be MORE than enough. If it's long enough to lose your radio presets, I think it's long enough to reset the ECU to factory default.

[Boomer]

Ah, the real experts show up. Maybe you won't have to make that trip to the dealer. Way to go Bonz/NSANY.

[NSANY]

Teri, you should think about getting an OBD II scanner for your garage. I have the AutoTap:

http://www.autotap.com/products.html

It will read any code, but it can't tweak all of them. At least you can see what the story is for yourself and refer to the shop manual. On a domestic vehicle, the AutoTap can muck with practically any setting in the entire ECU. Nissan's proprietary stuff hasn't been opened up for the aftermarket OBD II readers yet.

[Tere]

Yep, I can see that the AutoTap would be helpful. I sure wish Nissan would release the code so we could have a real tuner kit like the LS-1 folks have. I thought there was a federal law that went into effect about a year ago that manufacturers had to release the codes -- our luck, there's probably a loophole in regard to foreign manufacturers. I know SEMA has been pushing hard for the legislation.

[kbirdsong]

Wonder if this recall has anything to do with it.


Recall Alert: INFINITI G35, INFINITI I35, INFINITI M45, INFINITI Q45, NISSAN 350Z, NISSAN ALTIMA, NISSAN MAXIMA, NISSAN SENTRA
Owner Notification Date: Dec 02, 2003
Number of Units Potentially Affected: 630000
Component Description: ENGINE AND ENGINE COOLING
NHTSA Campaign Number: 03V455000 Schedule an Appointment


--------------------------------------------------------------------------------

Description of Recall Campaign: ON CERTAIN PASSENGER VEHICLES, THE CIRCUIT BOARD FOR THE CRANK POSITION SENSOR MAY HAVE AN IMPROPER SOLDER JOINT DUE TO SOLDER DEFORMATION CAUSED BY HEAT STRESS ACCELERATED BY THE EXISTENCE OF FLUX RESIDUE DURING THE SOLDERING PROCESS.
THIS COULD CAUSE THE "SERVICE ENGINE SOON" WARNING LIGHT TO COME ON, CREATE A NO START CONDITION, CAUSE REDUCED ENGINE POWER, OR CAUSE THE ENGINE TO STOP RUNNING WITHOUT WARNING DURING VEHICLE OPERATION, WHICH COULD RESULT IN A CRASH.

DEALERS WILL REPLACE THE CRANK POSITION SENSORS. VEHICLES INVOLVED IN A PREVIOUS RECALL CAMPAIGN, 01V357, ARE ALSO INCLUDED IN THIS NEW CAMPAIGN. THE REPLACEMENT SENSORS USED IN THAT CAMPAIGN ARE ALSO AFFECTED. OWNER NOTIFICATION IS EXPECTED TO BEGIN DURING DECEMBER 2003. OWNERS SHOULD CONTACT NISSAN AT 1-800-647-7261 OR INFINITI AT 1-800-662-6200.

[Tere]

As I recall, that one is only supposed to affect a small number of the early 350s -- mine was manufactured in Dec '02, so it should have the "fix" for the crank sensor.

The dyno does cause heat stress though, so you never know what could get fried...

[kbirdsong]

Ok, i did not know that. I am glad to here it however because my Z should have been made after yours.

[Steve350Z]

hope everything turns out good for you Tere

[Tere]

Thanks Steve! Ah, the trials and tribulations of tinkering with cars -- never a dull moment... :D

[steach]

Originally posted by NSANY
Can't the ECU also be reset by just disconnecting the battery terminal for a bit? :confused:

NSANY, Tere's HMO would never cover such a simple solution.

ps. Can a Z be added to a family health plan?

[Tere]

LMAO... That's what we need! Health insurance for our Zs! :D

Well, I did the ECU reset trick. Then started her up and the light stayed out -- the A/F was running a nice 14.7. Idle was a little rough and a tad lower than normal (about 500 rpm instead of my usual 800). Ran the rpm up to a steady 1,500 and she smoothed out a bit. The A/F took an occasional excursion to 15.3-15.5 but would drop back to 14.7 fairly quickly, and the light stayed out. Took her up to 2,500 rpm, and the A/F would bounce up as high as 17.5; the check engine light would come on and stay on until the A/F dropped back down to the 15 area. So I've definitely got a big A/F problem. Looks like the ECU is trying to lean out the A/F way too much. During my little test, I saw the A/F get as high as 18 -- not good, way too lean.

I'm going to double-check the operation of the boost bypass valve, but I don't think it's stuck closed, and I'm going to check the DFMU wiring and aux fuel pump just to eliminate all that (I don't think that's the issue though). When I gave it a couple quick revs, the A/F dropped down to the 11-12 range like it should. It's at the slow throttle onset or steady state that the A/F gets whacked, so that's more like something the ECU controls.

Where's Jesse, the god of air/fuel, whene you need him! LOL... :D

[Tere]

Well, the most likely culprit is a bad O2 sensor based on what I'm learning about misfires, and coupled with my wideband O2 readings. Interesting that a bad O2 sensor can indeed cause a misfire on random cylinders. While a vaccuum leak is the next most likely culprit, I couldn't spot any leaks in the obvious spots.

What I found: I had run some plastic wire loom around a couple wires routed down the firewall. Near as I can tell the heat generated during the dyno runs caused the plastic wire loom material to melt (normal driving air flow would have provided enough cooling to prevent the meltdown). The melted material flowed down around the wires and outside metal portion of the left O2 sensor mounted in the header collector. My guess is that the melted plastic caused the sensor to fail. When a sensor fails any number of things can happen, and I think I've just about seen 'em all on this one.

Here's a photo of the big meltdown:

[Dr Bonz]

Nice job Tere. So how much does one of those cost?

[Boomer]

Originally posted by Tere
Well, the most likely culprit is a bad O2 sensor based on what I'm learning about misfires, and coupled with my wideband O2 readings. Interesting that a bad O2 sensor can indeed cause a misfire on random cylinders. While a vaccuum leak is the next most likely culprit, I couldn't spot any leaks in the obvious spots.

What I found: I had run some plastic wire loom around a couple wires routed down the firewall. Near as I can tell the heat generated during the dyno runs caused the plastic wire loom material to melt (normal driving air flow would have provided enough cooling to prevent the meltdown). The melted material flowed down around the wires and outside metal portion of the left O2 sensor mounted in the header collector. My guess is that the melted plastic caused the sensor to fail. When a sensor fails any number of things can happen, and I think I've just about seen 'em all on this one.

Here's a photo of the big meltdown:

Eureka! You have found it! What are you replacing the melted mess with, more plastic or an exotic fix?

[peptidbond]

Sounds like the culprit to me.

Bonz,

O2 sensors are usually $80-100 for most Nissans. While the Z maybe different, I doubt that it is. An O2 sensor is such a universal item that I doubt Nissan would put a unique one on the Z. My guess is that Tere will have to go to Auto Zone and have one ordered. They normally do not keep the ones for Nissan in stock. The good news is that the order is a 24 hour turn-around.

[peptidbond]

Where is the update Tere?

[Tere]

Sorry about the delayed response... the 15-hour work days are killing me, and then Z Chickz forum got a facelift...

Wouldn't ya know, the one plug I didn't check (the #6 pipe was hot every time I checked it) -- somehow the gap got closed. I'm guessing it was a clipped piece of zip tie that managed to pop into an open intake pipe unnoticed (when we were doing the headers and cats). I had the piston and cylinder scoped and there wasn't a scratch anywhere. One of these days when the spirit moves me, I'll drop the cat and see if there's anything in there (provided it's not vaporized).

Anyway, the 350 is now cooking right along, none the worse for wear. Moral to the story is always check all the plugs! :D