*Running a 165 ecm and a special ARAP based bin file that displays my DIY WB-o2 output in the datastream (although it's not currently connected).*

A few days ago my eng started trying to die while cruising down the road steady state. Hard to tell if it's loosing spark or fuel or both (didn't have my wb-02 going at the moment). It will shut off for maybe a second, then come back to life for a sec, then die for a sec, then back to life again. It does this over and over for 10-20 secs, then it's fine for the next 100 miles. The check eng light flickers while this is happening. Afterwards, all I get is a code 12.
Over the last week, this episode has happened once every 1 or 2 days. Every time it acts very similiar and the duration is roughly the same for these episodes (10-20 secs).

I replaced the Wells brand ign module with a Borg Warner unit thinking that might be the problem. Worked ok for a day, then this morn it returned. This time I was datalogging when it happened. Hopefully after I zip the CSV file I can upload it. Basically what it shows is all normal readings for cruising down the highway at 60 mph, Then at datapoint 2213 all columns go to zero except; % of throttle goes to -13% (yes, negative!), coolant temp goes to -40 (yep, negative), air-fuel ratio goes to 99.8, power calculated is -57hp.

The engine run time resets itself back to zero.
The eng goes into open loop for abt 40 secs, then back to closed loop. Abt 5-10 secs into this episode I noticed my laptop had become unresponsive (All readings appeared frozen and I was also unable to stop datalogging). After a few secs, the laptop started working normally again and the datalogging stopped. I was able to restart the datalog. The remainder of the drive was uneventful. I've had the same problem with the laptop powered off, so I don't think it's related to the laptop.

I normally get a datapoint every .2 to .3 secs, but during this episode, there's a 18 sec gap from the last valid sampling to the next dataset (that shows mainly zeros). The next 3 readings have a 2-5 sec gaps between then (still all zeros in data field). Then there's another 75 sec gap before the next dataset (all good data), then a blank row, then everything returns to normal.

Tried wiggling all the wiring under the hood with eng idling, no issues found.

The dash volt gauge stays at it's normal position throughout the episode. Didn't noticed what the tach was doing.The ECM is usually the last thing I would replace, but it's kinda looking like it may the problem. Has anyone datalogged a similiar event?? Does it sound like a bad bin, bad burn, bad ECM??

 

They are noted for having mystery electrical problems. Find a good spare, and always keep it handy for doing R+R ecm testing.

The only ignition modules, I ever use are A/C Delco, period.......

Might take a close look at the battery cables, and connections, real close.

 

This sounds a lot like the problems I was having when my 4.9 Caddy's ECM started acting up. My ECM would start reporting like 151C AIT temp, 151C engine temp, 100KPA map, 100% TPS, and the engine would start bucking like mad and just die while just crusing down the street. It would always do it once the ECM got nice and warm. I actually got the car for free because of this! (very nice 92 Eldo) I swapped the ECM out and the problems never returned.

Love the car, but gave it to my mom because she was in need of a car and I missed driving my TPI truck.

 

Sounds like you a describing a ZIF socket problem. Flashing SES, processor reset, no codes, timer reset to zero. Dead ringer for a bad PROM pin connection.

You wouldn't happen to be using a ZIF socket? Losing ECM power/ground may do it. I am not sure if the SES would light up in that case.

 

Batt cables look ok. Grounds look good and are tight. I never lost sys voltage during the events according to my dash voltmeter.

I am using a ZIF socket on the memcal. It has been installed for 3-4 years and hasn't been disturbed for a couple of months. I'll drop the ecm down and check it as soon as I can.
I'm currently nursing a broken coller bone and shoulder blade, so I'm one-handed and have a limited ability to maneuver under the dash.

So far it has only happened after I've been driving abt 10-15 miles from a cold start-up. So a bad connection on the ecm board does seem kinda likely since it would still be warming up.

I had a spare ecm that came with my parts doner car, but I let a friend borrow it a couple of years ago for his test bench. I'll see if I can get it back from him. I believe that ecm is from a SD vehicle since it has the knock sensor gizmo on the memcal. I think it will work if I swap my memcal into it,, right?

 

It seems to me that just the ECM lost power. Just because you didn't lose power to the volt meter doesn't mean you didn't lose power to the ECM. Fuses are designed to be the weak link in the power circuit and can become flaky. IIWY I would replace the ECM fuse(s) just to be sure. BTDT

 

It's possible the ecm could be losing power. Guess I could wire up a led to the ecm power wire and see if it flashes when the ecm goofs up.

It hasn't happen for a few days now. Every time it has happened, I've been on smooth roads. Seems if it was a hardware problem it would act up more on bumpy roads (which I drive on daily).

The bin file I'm using came with a disclaimer that it hadn't been tested (it was hacked to have the wb-o2 in the datastream). Wonder if that might be the problem??

 

Who did your WB hack, and what base BIN was used?

 

Got the bin from here. Next to the last link on the page.
DIY-EFI WB datalogging Project

It's based off the ARAP bin.

 

Just an update: Problem hasn't re-occured in over a week now and I've been driving it every day. My laptop did die pretty soon after the last occurance.

Maybe the laptop was doing it (it was connected to the aldl, but not powered up). Using a different laptop now.. So far, so good.