MS3x hiccuped and was stranded on shoulder. Help me figure out why pls.
 

This happened Friday on the way home from work and I haven't screwed with it or driven it since. I am hoping you all can help me troueshoot this, but first we need to duplicate the problem.

All week my tune has been evolving, and on Friday I was to the point where I was happy enough with my tune to leave the laptop in the trunk. I left work and started to drive home at interstate highway speeds. About 20 minutes into my trip, the car died for a moment (at approximately 60mph) so I pushed in the clutch and restarted the car and continued driving. My exit came up pretty quickly after that and it seemed like the engine was attempting to maintain 2500rpm or so. When I got to the light I got in neutral and waited for my turn to go. When my light turned green I attempted to accelerate just like I always do but the car immediately died. Luckily I had enough momentum to coast over to the shoulder and get out of traffic.

Once stopped I tried several times to restart the car. No luck. I went to get the laptop out of the trunk do I could see what was going on with the megasquirt. I am having problems with my tuning laptop hard drive so by the time it did the chkdsk sequence and I got into tunerstudio I was able to restart the car on the first turn of the key. I drove home the rest of the way with both the SD card logging as well as the laptop. I didn't have a single problem for better or for worse.

At the time I had my problem:
Coolant was about 190-195 (guessing from experience)
A/C was off
Heater was on speed 4 (top down and it was brisk out)
Radio was on and blasting
Headlights were on
No load on cig lighter plug
Cooling fan MAY have cycled at those temps
Car has about 85k miles
Fuel pump never replaced
MS3x has 6 grounds directly to engine block (not shared with parallel ecu)
MS3x is otherwise wired in parallel with stock ecu
MS3x controls everything except EGR (not hooked up) and alternator for now.


I googled and searched here and MSextra and it looks like it may have been a CPU reset. However, I have a MapDaddy so it shouldn't have to wait for the engine to stop to get a reference map reading, and it seemed like it took some time before I was able to restart the car, but then I was not comfortable cranking the car over more than two or three times. I wanted to have enough juice left to start the car on the stock ecu if need be.

On the way home tomorrow I will try and load up the electronics to see if I can duplicate the problem while logging for more data. Any thoughts in the mean time? If it was a CPU reset, any ideas on how to fix it so it doesn't happen again?

Also, at time of first episode I was at approximately 3k rpm. At stoking I never got above 2500 rpm. (If I did, I really was launching like a dork)

No problems on drive to work today. I stressed the charging system as much as I could but could not duplicate my problem.

The only other variable I can think of was that the case to the MS3 was warm to the touch but then I had my heat on so I can't say one way or another if it was radiating heat itself.

I will post more as I figure stuff out. Ideas welcome.

Any chance it was flakey CAS wiring?

Anything is possible. I know that when I was having my problems getting the car started last year I was so frustrated I made a new patch harness and made sure to shield the crank sensor wire this time.

By flakey do you mean an intermittant open in the line somewhere or are you talking about possible noise confusing the MS? I imagine to check for an intermittant open I could let the car run and shake harnesses while logging - correct? Now that I think of it, RPMs stayed at zero during the failure while ECU seemed to still be powered up while at the stoplight.

Also, I just read in another thread that the NB CAS is known to be flakey on it's own. Perhaps I should order one up and keep it in the car to try next time.

The problem is - it is hard to troubleshoot when the car is running and running fine. I am hoping to find things to check while it's running so I can reduce the chance of me being stranded again.

Does anyone know if the stock ECU goes into limp (batch fire) when it doesn't see the CAS or if it just dies completely.

Intermittent open. You can often spot if there's a problem with the wire starting to lose its insulation at the connector.

Have you traced the vacuum lines to confirm you don't have a loose fitting?

agree with the cps thing. maybe check your CAS as well.

Haven't had that problem since I posted this. If it happens again I will resurrect this topic.

Did it survive autocross with no issues? Or did you not go?

I saw the forecast for rain and thunderstorms all day and thought that the day would be a waste so I slept in. Then we have a beautiful day and I kicked myself. Either way, I got a whole bunch done so I felt it was somewhat of a worthwhile trade off.

Btw, If you want to bring your car by Mike at Red Mist for an estimate let me know. I'll meet you there and maybe we can stop by Cruisin' 72 for a beer or burger after.

Had same/similar problem. Neither trigger on any of the loggers. Had to get car towed home. I will start dissecting it all tonight.

only time this ever happened to me it was a broken cam sensor. It's just sitting up there waiting for something to bash it and make it unuseable.

What can break a cam sensor? What would I see in tunerstudio that would help me identify that as a problem.

you would not see cam sync, just crank. pop the hood and look at it is faster though. it's on the back of the intake cam at the top of the valve cover near the firewall.

I did yesterday and didn't see any obvious damage to the sensor or wiring. I don't recall seeing ANY triggers. Not just cam. Each sensor seems to have multiple wires going to it. Do they share any common tie points that could be knocking out both sensors?

I will have to pull my wiring diagram...

EDIT: It appears that the main relay provides power to both sensors, and they both share a common ground. Possibly fused.

they share 12v.

After looking at the wiring diagram, here is what I will be checking:

Position "A" on both the cam and crank position sensor plug to 12v (White/Red wire)
Position "C" on both the cam and crank position sensor to ground. (Black/Blue wire)

If "A" doesn't have power, check wiring from main relay and "tie point (?) X-16.

If "B" doesn't indicate ground, locate tie point (?) X-19 and check for continuity.

X-19 and X-16 both appear to be under the dash.

Problem solved. Pin to vehicle harness crank signal was broken. Cannibalized one off of my 1.0 harness, re soldered and I am back in business.

I have to say, if I hadn't built this all myself I wouldn't have had te confidence and knowledge to find the problem and fix it so quickly. Turns out there is something to this "experience" thing after all.