Balls alive! Car soon back on the way to work.
 

Haven't debugged this yet, but the car is parked down at the park 'n ride lot down the street from my work. It'll be towed home if it doesn't start tonight.

Timeline:

a week or so ago, the car died randomly while driving slowly (25 mph, light throttle) and I was able to coast to the curb and restart it. It restarted with a backfire and I went on my way. the logs show nothing other than RPM going to zero.

Over the last few days, I've felt the "hiccup" during cruise. The car would buck for an instant--likely dying and restarting over the course of a split second on the fwy. it did this a few times on the way to work. only under light throttle. flooring it seemed ok.

then I pull off the freeway today and am waiting in the left turn lane to turn onto the street my office is on... the car feels a little boggy through the turn and dies dead in the far crosswalk. Tried a half dozen times to restart with various states of pedal (none, some, flood-clear) and nothin.

some dude in a bad tie helped me push it to a lot. It's humid today so he's a badass.

Checked a few things on the adaptronic--nothing appears odd. loaded a fresh map: no change. I have sync and whatnot.

I did notice that the injector current stays at zero during cranking in the F11 data screen, but the voltages are reasonable. I popped the hood and nothing looked odd. wiggled some sensor connections and no change.

...

Obviously when I get the bitch home I'll check for fuel and spark but I'm open to other ideas. I suspect it's a broken connection somewhere that finally came completely apart.

...

PS All people who mod cars should have AAA Plus. Having a car towed up to 100 miles for whatver they charge is a bargain. Especially if you use it annually.

Matt

Damn. Keep us posted.

Matt, you have COPS, right?

yes I have COPs. and they were hot. but everything was hot, so... :)

(dwell is 2600 usec FYI)

Check the crank and cam sensor signals and sync status.
Knock on aluminum, the miata has never died on me (except when it threw a rod after I over revved it).
The (unhacked) BMW, OTOH ...

the adaptronic's info window shows crank and synch (cam) status. they appear to be correct and working.

On occasion mine will hiccup during cruise and full throttle. Nothing in the logs to suggest why but it has never completely shut down.

Mine used to do this at random as well. Has not happened one single time since I went back to peak and hold instead of saturated injectors weeks ago.

Lemme know if you suspect if it is your Adaptronic box and I can bring mine over to help.

Good luck!
Cheers,
Matt

+1 on having AAA. I have had it 2 years and used it once on the Miata and once on the BMW for a failed water pump. Hope you get the problem sorted out. I had a MAP sensor connector flake on me about a year ago and caused a similar problem.

He has a hand made harness so its a little harder to trouble shoot. Let me know if I can do anything to help.

My guess is Crank Angle sensor. My Miata did the same thing not too long ago.

+1 on the AAA coverage. I bought it for me and my wife for the reasons stated.

Crank angle sensor would be my guess too. Even though mine is 'working' I suspect it might be on the path to failure since it has the screwed-up trailing edge problem.

if it were, it wouldn't register any RPM in the logs.

the first time it died a few days ago, I happened to be logging. Note that the LC1 doesn't output useable data when it's too lean, rich, or battery voltage is low. You can see in the graph just to the left of the marker where rpm drops and MAP goes to atmospheric. You can see me crank it once and it almost starts and then the second time it catches and runs.

For some reason in this log batt voltage drops to zero (!?) for a split second and it backfired as it started probably due to some random enrichment kicking in. MVSS (veh speed) goes to infinity briefly also.

Second set of graphs is today. basically me cranking a few times and nothing happening. note battery voltage is low-ish but not zero!

btw, it's always struck me as odd that the adaptronic records the injector time even if the injectors aren't firing.

edit is for suckers.

my guess is more likely cam sensor. mine's a little hacked up from being broken and repaired and I have a spare so it's an easy one to test.
http://www.partsgeek.com/images/full...1755737OEA.JPG
See the thin area between the connector and body? snapped off. so I carved out the contacts and wired it back together. been fine for yeeeears which is why I never put in the spare. :)

That sounds exactly like what happened while my tps wires were breaking inside the shrink wrap.

y8s,
Are you running closed or open loop for fuel?
Also the LC1 will read AFR from 8 to 18, ( the engine wont run outside this range) when it reads 20 it’s warming up or in a fault mode (have a look in the LC1’s manual).
I had a faulty OEM O2 sensor which caused no end of intermittent faults, I unplugged both O2 sensors and ran open loop with the fuel this helped isolate the problem for me.
I only have the CAS (NA8A)on mine so I can’t comment on the other sensor.
Hope this helps (I was able to try another ecu in my car and the fault remained), I am sure the problem in not the ecu.
Ted

I'm running autotune (closed loop + corrections to fuel map).

It may be the adaptronic excluding <9 and >22 AFR from the LC1 serial connection.

In either case, the Adaptronic uses a fixed fuel pulse for cranking. I suspect it should at least *try* to turn over even if TPS, MAP, and O2 feedback are not available. Hell, it should do more than it does if I lose cam sync.

I have also lots of start problems with my shitty custom harness. With rich numbers in the cranking table, the car starts when it want to. I tried lot of numbers in the cranking table and different cranking advance timing. Nothing work fine.

Also have lots of hiccups...

y8s, the only thing we have in common is we used BAT85 diodes and we have the same '01 car.

I noticed a curious fact. If I clear the Pcodes the idle goes high until the pcodes come back (1-2 seconds)

Maybe the problem is related with the sharing of the CKP or CAS signals.

Depending on how the TPS fails it will do different things. Mine failed to WOT so it triggered the fuel cut, so no fuel when cranking. If it fails closed it will start just fine.

From your logs though it doesn't look like its the TPS.

The adaptronic will only display from 10-20 AFR.

When mine was dieing at low speed I'd just stick it back in gear and pop the clutch. I was never sure if it would start on its own.

I had a hiccup then no start problem myself. Sounds similar to yours

Turned out to be a ever so slightly loose ribbon cable inside the MS box.

update... the car is starting now.

I pulled a COP out and shoved an old plug into it and grounded the threads to check for spark... didn't appear to jump an arc so I groped all the wires leading to the plugs for the COP triggers and as I got to the ECU, I swear I felt one of the stock ECU connections *click* into its seat.

I tried the key again and the car started... oddly the tach was dead and the idle was high. A few tries later, it started normally and everything worked.

I came in and checked the wiring harness pinout. the middle connector carries IAC, ignition, and tach signals. This makes some sense since the adaptronic was getting a normal tach signal as it is spliced before the stock ECU. The adaptronic is also not handling idle control.

I can only assume the no-spark was a fluke of my vision or grounding.

At least it wasn't something serious. Just something mysterious.


PS who changed my title?

Ha, same thing happened to me. My electrical system got very flaky after I had done an electrical system mod. At first I thought it was the new electronics, but it turns out one of the boomslang connectors that plugs into the stock ECU (parallel install) was not fully seated and making intermittent contact. Pushed it back in, problem gone.

I hope to jeebus that's all it was.

i was right about a connector. I win.

I believe you claimed it was my homemade harness... which is probably fine. but it doesn't work so well when not attached to car.

i said one of your connectors as a possibility, I remember this.

no, i think you said it had something to do with kitty photos. I'm sure of it.

I don't know what you're talking about but the problem was Yoko Ono for sure.

more fucking around yielded more information.

1. my obdii scanner usb connector broke and wasn't making contact so I couldn't check the MIL

2. after fixing #1, the MIL revealed itself to be a P1345 Cam Angle Sensor.

3. I started the car and let it idle for a while. After several minutes of normal idling, it started doing a weird "buh" and the idle would go briefly unsteady. It did this a few times (I was logging, nothing was obvious) and eventually crapped itself out.

4. I couldn't get it restarted, though it was behaving much more like it wasn't synching rather than just dead-dead.

5. Dug out one of my spare cam sensors (yeah, plural) and switched it in.

6. Started the car right up and the MIL was out--obviously it figured out there was a working cam sensor and corrected itself.

7. came inside to evaluate the original sensor:

http://photos-e-0.ak.fbcdn.net/hphot..._5961051_n.jpg

You love my hack job, I know you do. Surprised it lasted as many years as it did.

The reason I never replaced it was because those god damn sensors are SO easy to break I figured it wasn't worth the effort.

So there you have it. zoom.

So, during this down time. Did you happen to swap out a certain ewg spring?

no. why would I do that?

You are so gay.... FIN