Starts, Fires Up, Dies...

[bar10dah]

Started up (without any issues) and drove my '99 Box w/ 69K miles out about 10 miles to go shopping. Car was parked for about an hour. Came back to the car, started her up, car reved up, reved down, stalled. Started again, same. Started a third time and opened the throttle a bit, revs stayed up. Let off on throttle, idle settled down to the normal ~800 RPM. Hope this isn't a sign of problems to come!

Drove the car about 10 miles to get home and shut it down in the driveway. Cranked the car over again, reved up and settled back down to around 800. Okay, so problem gone. But just to check, shut car off several times and restarted, all normal. Parked for about two hours.

Started the car, same stalling issue. Car just doesn't seem to want to idle after being shut down for a bit. But acts normally after driving.

Initial thoughts, from other posts I've read, possibly the MAF. From my service records, a previous owner replaced it about two years ago. No CEL, but I'll try hooking up the Durametric anyways. I'll try cleaning the MAF and the throttle body this weekend and see if the problem keeps occuring.

[bar10dah]

Okay, so Durametric says:

Bosch Digital Motor Electronics Motronic 5.2.2
Module Identification


Part Number 986.618.621.01
Supplier Part Number 004155073601


Current Fault Codes

P1126:
Oxygen sensing range 1 Cylinder (4-6)

Does that mean replace O2 sensor? And if so, best (but not required) to replace both sides?

Durametric says the oxygen sensing (can't remember what it says here) for both sensors usually floats around 0.97 and 1. Sometimes it dips real low, but always comes back up to around 1. And both sides show about the same numbers. Sometimes they do diverge, where one side reads high while the other side reads low, but they always both return to around 1.

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

Also to note, but not related to the above problem: Whenever I hook up the Durametric, I check for camshaft position deviation to see if I can detect an iminent IMS bearing failure. I only have camshaft position deviation 1. Aren't I supposed to have two sensors? And if so, why can't I see the second one as an option to view? If nobody knows, I guess I can always email Durametric. If I recall correctly, mine always reads -3.0 degrees. I let it idle and I rev the motor a few times, it stays at a constant -3.0 degrees. Tonight, it was reading a steady -2.0 degrees. Not sure why.

[randy_k]

My saga was. Wouldn't crank when it sat for 15 minutes or so. Replaced the camshaft sensor, same year as yours. There is only one.
Then I found that it would idle rough as yours is. Durametric tool indicated the same O2 sensor as yours. But I replaced left and right banks. Problem solved.
I just replaced both because. That is the way I look at things. One is gone. The other is probably not far behind. Kinda like front tires. If one is shot. The the other one is in the near future.
Just my thoughts.
Good luck, Randy

[bar10dah]

A couple other things that came to mind. For the past two days, there was a distinct, albeit slight, "engine" smell after parking the car. Appeared to be coming from the passenger side intake, or abouts that area. I'm not sure how else to describe it, other than it's probably a cross between motor oil and exhaust. Possibly from burning too rich.

Also, for a couple weeks now, I would get a slight puff of smoke on cold start-up. Usually not much, as I've read about the common "smoke bombs" that is the Boxster. This problem may not be related, but the "engine" smell probably is.

From reading other posts, here and on other forums, I could have a vacuum leak, AOS failure, O2 failure, MAF failure, oil tube crack, or oil filler cap leakage.

I pulled the oil filler cap off and the car's RPMs increases by maybe 50. Put the cap back on and RPMs drop back down to around 800. I don't notice any high pressure being sucked into the filler tube. Seemed like normal pressure.

Given the smell, P1126 code, and idling problem, what order should I start tackling this problem? Start with looking for a vacuum leak, then replace the O2 sensors?

[bar10dah]

Originally Posted by randy_k

Durametric tool indicated the same O2 sensor as yours. But I replaced left and right banks. Problem solved.
I just replaced both because. That is the way I look at things. One is gone. The other is probably not far behind.

I'm thinking the same thing. If I do replace, I'll do it in pairs. Pelican lists them for $141 each.

[Meirschwartz]

The camshaft deviation will show under engine module > Actual values > camshaft position 1 deviation and camshaft position 2 deviation.
The value it will show is not important as long as the graph stay steady.

Now, I also believe in the concept of replacing parts in pairs, in a car that has clutch cable and acceleration cable, most of the chances that if one of them breaks, the other one is on it's way (not in a boxster obvcours just i.e). The tires is a good example as well.
How ever, this concept is relevant for mechanical or semi mechanical parts.
Like the clutch switch that broke in my friends car. When I replaced the switch I noticed the the brake light switch is on is way out, so I replaced them both.

The O2 sensors how ever, are electronic (or electric if you will) components.
When the light bulb in my left tail light burned, I didn't replaced all the light bulbs in the car.
If you want to save your $$, just replace the bad sensor. If you don't, replace all 4 of them.
This whey or the other, the sensors are relatively accessible, so it's not an issue to replace the other down the road, if needed.
Good luck.

[bar10dah]

Originally Posted by Meirschwartz

The camshaft deviation will show under engine module > Actual values > camshaft position 1 deviation and camshaft position 2 deviation.
The value it will show is not important as long as the graph stay steady.

I was running version 6.1.1.1. Now that I'm running 6.1.1.3. I can see a Camshaft Position 2. Looks like this wasn't an available feature for the ME522 DME until version 6.1.1.2. Both are rock steady at -3.0 degrees. At idle, under acceleration/decceleration, and high RPMs.

[bar10dah]

So, yesterday and today (days 3 and 4) the car has been running like normal. No more stalling right after starting up. The only thing that's changed is I cleared the fault codes. However, the day of and the day after (days 1 and 2) I reset the codes it still stalled out. What gives?!

[bar10dah]

Took the MAF out and sprayed it down with MAF cleaner. Looked clean coming out, so doesn't look any better after cleaning it. A side note, Craftsman 7-pc Tamper-Proof Torx Bit Socket Set from Sears costs $39.99. For anyone else looking to clean your MAF, the tamper-proof T-20 is required to pull the MAF. $40 is awfully expensive for a 7-pc socket set. They're much cheaper at auto parts stores and even cheaper at Harbor Freight. However, items like these I always buy quality as they're less prone to strip screws/bolts.

Took the throttle body off and saw signs of a light oil coating in the opening. Reading online, this is normal.

Took the AOS tube off. Both ends had a little oil in them. But it wasn't pouring out. So again, from what I've read online, appears normal. The AOS body itself didn't seem like there were any external cracks in it, and no oil build-up around it or any component next to it. Also, took the oil cap off. Reading online, if the AOS is bad, the cap should be hard to pull off due to the high vacuum in the crankcase. But, it came off just fine. So with all this, I'm guessing the AOS is probably fine.

Checked for fault codes with the Durametric again. Nothing.

Started the car. Started fine. Let it idle for about 2-3 mins, then took it for a 20 mile county road ride. Had Durametric hooked up the whole time, monitoring the pre-catalyst O2 sensors (both banks). Signals oscillated from 0.08 to 0.76, which seems fine. Well, online I've read the signals should oscillate between 0.2 and 0.8. 0.2 being very lean and 0.8 being very rich. And a failing O2 sensor should read rich constantly. So, with my readings, the O2 sensors appear to be operating correctly.

Still have to try looking for a vacuum leak. I wonder if those halloween fog machines would work in diagnosing a vacuum leak?

The oil tube and oil filler cap seems fine. Nothing visually wrong that I can see.

[bar10dah]

Three days ago, car started fine the first half of the day. The second half, it stalled on two of the multiple starts.

Two days ago, car sat without begin started.

Yesterday, the first time I started the car, it stalled. Subsequent times throughout the day it was fine.

Today, it started every time.

This intermittent problem baffles me...

[Dale_K]

My bet is the fuel pump.

[bar10dah]

Thanks Dale_K, I hadn't even considered a fuel pump causing my intermittent stall upon starting. However, how does a failing fuel pump set code P1126 (post #2, above) which says, "Oxygen sensing range 1 Cylinder (4-6)?" Or, more than likely, they're two separate issues?

[Dale_K]

My 1126 code issues were fixed with a new MAF. The cleaning I did with genuine MAF sensor cleaner didn't do any good. There were no MAF codes and the data readings looked OK but a new MAF totally cleared up my car. Low fuel pressure might fool the computer into thinking the car is running lean - it tries to give it more fuel, by increasing the duration of the injection pulse up to it's limit value and when that limit is exceeded it sets the CEL.

You could check the fuel pressure at the test port on the fuel rail. I don't know a way of testing the MAF except for trying a known good one. The only feasible test is to disconnect it.

I'm not really an expert on the Boxster so take all this with a grain of salt. I'm only going on my own experience with my 2000 S.

[bar10dah]

No worries, thanks Dale_K. I'm still in the intel gathering stage, so any information (especially personal experience) is useful.

At this point, I'm leaning towards the MAF. I've troubleshot the most I can, without buying/borrowing any special tools. I'm going to drive it for a week or two and see if the P1126 code reappears. I'll move forward from there.

[bar10dah]

So, about a 100 miles later, P1126 has reappeared and P0446 [shutoff valve activated charcoal filter (function) below limit value] has accompanied it. I did refill the fuel tank last night and drove 1.5 miles home. So it was pretty full. Today when I turned the key to on, but not to start, I checked the computer with Durametric, and found P0446. Not sure if the full tank could be at fault or if it's related to my P1126 code but I would guess it's because of the full tank.

I believe my next step will be to order a new MAF and see if it changes anything.

[bar10dah]

Reset faults, drove 15 miles out, parked for five hours, drove 15 miles back. Last mile home, got a CEL. This time got codes P1124 and P1126. From what I've been reading, both of these codes together often mean an air leakage? Either in the intake system or the AOS? I didn't "see" anything that would indicate a leak. Nothing that I could see looked worn or cracked. The oil filler cap's o-ring looked good. And there wasn't any noticeable difficulty in removing the cap, to indicate an AOS problem. At first, with just P1126, I was thinking it can't be an intake air system leak, because the fault would be noticeable in both banks. But now...

Not sure if replacing the MAF is the way to go anymore...

[bar10dah]

Okay, so here's some contradicting info.

Durametric (Bosche DME Motronic 5.2.2) says:
P1124 Oxygen sensing range 1 Cylinder (1-3)
P1126 Oxygen sensing range 1 Cylinder (4-6)

Bentley (Table b. DME M 5.2.2 DTCs) says:
P1124 Fuel pump relay output stage
P1126 Oxygen sensing adaptation, lower load range, cyl. 1-3

What gives?

[bala]

Drive the car with the MAF and push it beyond 70. Note the response. Disconnect the MAF and do the same. If the response is night and day (day being better without MAF) then you might need a good cleaning of the MAF or replace. Don't drive too long though.

I had both bank 1 and bank 2 o2 sensors giving errors - pre cat. Heard parts train had some good prices and got from them for 27 each, used the old connectors and the codes went away immediately. Still have to do a drive test and waiting for my seats so I don't have to sit on the floor and drive!

Seriously, I did one at a time. Swapped the new left one with the old right one and confirmed it was the o2. You dont have to replace the ones that aren't showing errors. Waste of money. But, I did change my fuel cap (got error of that and think it was due to cracked seal ring), spark plugs (old), SP tubes and coil tubes (oil leak), and throttle body clean. I have the Durametric and cleared the codes, did a short test and confirmed all is ok. BTW, there is Camshaft Position 1 deviation and 2 on the latest version. It's right below Ignition Timing.

[bar10dah]

MAF came in today.

While waiting for it to arrive, opened up the engine cover and poked around looking for a leak in the oil filler tubes. Opened up the oil cap and noticed engine runs a little rough, which is normal. Exhaust puffed out a little bit of smoke when I first pulled the oil cap off, but lasted only a couple seconds. Suction at the oil cap appeared normal. Recapped oil. Started moving parts of the oil filler tube around to see if the idle changes. Nothing. However, noticed the tube downstream from the flex tube (#19 being the flex tube and #15 being the part in question, in this parts diagram) had oil/dirt coating the bottom-side. Adjacent parts also had some oil/dirt. So, there's a leak somewhere around that area. But, since there's a vacuum in the crankcase, shouldn't air enter, and not exit (taking oil with it) that area on top of the engine?

I've got to try to find a shop that can do smoke testing...

Installed the MAF. Car stumbled just a bit on start, but smoothed out. Only drove a mile (car already warmed up from idling while looking for vacuum leaks). Had Durametric hooked up. Can't recall the numbers off-hand, but they appeared normal. I'll post the actual values later tonight.

[bar10dah]

I can't figure out how Durametric stores info... so I can't find my logfiles...

Went out for about a 20 mile drive out on county/city roads. Oxygen sensing for both banks hovered around 1. Would flow between the high 0.9s and low 1.0s. Not sure if this is normal, couldn't find info online. However, Durametric's website says if both banks oxygen sensing is faulting, it's generally a MAF or vacuum leak.

Mass air usually around 14, but climbed to a greater number that I can't remember when under load. Not sure the normal value for this either.

Pre-cat O2 sensors were oscillating between 0.2 and 0.8, usually. Which from what I've been reading, is normal.

Codes P1124 and P1126 are back, even after replacing the MAF, which is no surprise. At first, with just the P1126 code, I thought it could be the O2 sensor. But then when I got the combo with P1124, thought it could be the MAF. Afterall, it's unusual for both side's O2 sensors to go out at the same time. So the fault had to be further up the chain. But, replaced the MAF and still get the two codes. So, back to tracing the air leaks. I'll look at replacing that oil filler tube that I described in the previous post and see if anything changes.