1994 B350 5.2L w/85k miles on the clock. Vehicle is new-to-me. Drove home fine. Started going through it since it sat for several years. Plopped in a new battery cause old one was toast. Changed spark plugs and oil/filter yesterday and vehicle started right up but died about one minute later. Now I have a crank, no-start condition. I have fuel but no fire at the ignition coil. Here's what I've done, in order, to diagnose and and I still can't figure this out (bolded items of interest which may serve as clues):

• Check Engine light does illuminate and key dance gives me codes 12 and 55 (no surprise)
• Verified fuel prime at key on and verified pressure at rail is ~38-40 psi (extended cranking produces fuel smell so injectors are firing)
• In-line spark tester at coil terminal to distributor wire verified no spark
• Tested voltage to gray wire at coil electrical connector: 0v KO/EO and during cranking it jumps around wildly giving me random voltages ranging from 3v-10v. This doesn't seem right. However, this might just be my multimeter displaying the signal as it catches it/refreshes.
• Replaced ignition coil and still no spark at coil
• Replaced crank position sensor (CPS); no change
• Checked CPS sensor and cam position sensor/pickup coil for 5v reference; confirmed 5v at harnesses
• Inspected cam position sensor (did not replace), replaced rotor and distributor cap (no abnormal wear noticed on these items) for good measure
• Removed ASD and fuel pump relays and cleaned terminals, reinstalled, and vehicle fired right up but died less than a minute later and won't re-start.
• Replaced relays temporarily with five-terminal relays I had on hand; no change
• Inspected wiring and fusible links for shorts to ground and nothing out of the ordinary

I see lots of threads about failing PCMs and the trick of heating them up with a hair dryer. I'm not ready to condemn the PCM just yet. It's 90*F+ so the PCM is getting plenty warm. What next?

 

Okay, so NOW I'm getting a code 42 after re-checking this AM so there's something going on in that circuit. I have some 40A five-pin connector relays that I swapped in yesterday and surprise that didn't fix it up.

 

Update #2: Ended up being whatever relay that is next to the PCM. I threw in the guts of a new relay on the harness and it started right up and I let it idle for about 15min w/no issues.

 

I'm pretty certain one is the ASD relay, and the other is the fuel pump relay.

 

From all the reading I did to research this issue, my understanding is the Auto Shut Down (ASD) relay is located at approx 4 o'clock in relation to the brake booster on the firewall and is the one on the left (if facing the firewall) and the fuel pump relay is on the right. These two relays share a bracket. On RockAuto, the same part is listed under several sections such as, "ignition relay," "computer control relay," "fuel pump / circuit opening relay," "fuel injection relay," and finally, "auto shutdown relay." This leftmost relay also has a green with orange stripe wire going into it which matches the same scheme as the wire going to the ignition coil and also each injector. I believe the culprit in my case was some sort of ignition relay, best I can tell based on the blue and red wiring colors and gauge thickness of said wires. If I have any more issues down the road, I'm going to spring for a hard copy shop manual so I have all the schematics at the ready.

 

On my 96 all the relays are the same and you can swap them around to find one that isn't working.

 

Normally, that would be my go-to move. However, there are only three relays under the hood (ASD, fuel pump, and ignition/PCM?) and removal of one would cause a no-start condition. There are probably more squirrled away behind the dash, etc but I wasn't about to go just popping out relays not knowing what circuit they belonged too. Learned years ago to always have a few relays bopping around for troubleshooting.

Anyhoo...I saw multiple threads going back years with similar no-start/no spark conditions, with a few of them dropping off and never being updated with the solution. Hopefully this thread helps someone else later on down the road.