OT '96 Saturn DOHC won't start

Wife says car suddenly started running very rough on the way to work Friday. Check the codes, get only "cylinder 3 misfire". OK, plugs are old anyway, so replace all four. Now get "cylinder 3 misfire" *and* "cylinder 4 misfire". Yes, I gapped them correctly, double- checked underhood sticker and factory service manual.
Pull plug wires. #3 wire and coil tower corroded. Replace all four plug wires, and wire-brush #3 coil tower. Now only "cylinder 4 misfire". Engine still runs badly rough.
Coils check OK on ohmmeter, but heck, they're 18 years old, replace them anyway. No change in rough running or in codes.
Hmm. Fuel problem? Pressure seems OK, so why is #4 misfiring? Disconnect #4 fuel injector. Engine runs the same with or without #4 connected. Voltage is present at all four injector connectors.
Removed fuel rail, replaced #4 injector, reinstalled fuel rail.
Now it won't start at all. Cranks OK, and sounds like it's almost catching, but it doesn't run. No codes, either.
Service manual points at crankshaft position sensor or stuck EGR valve as possible causes. Pulled sensor, checks out OK (measures 778 ohms, spec is 700 to 900). Pulled EGR valve, not stuck.
And at this point I am baffled. Any ideas?

Doug Miller wrote:
  Quoted Text Here  
Timing belt ? Have you done a compression check ? If it has a distributor , is the inside clean with no carbon tracks ? Did you bleed all the air from the fuel rail after you reinstalled it ? Did you miss a plug when putting it all back together ? Lots of possibilities to screw up !

  Quoted Text Here  
Chain.
  Quoted Text Here  
All four cylinders in spec.
  Quoted Text Here  
Ignition is totally electronic.
  Quoted Text Here  
No. Factory service manual doesn't mention any need to do so.
  Quoted Text Here  
No. Remember that it ran -- but ran rough -- before I replaced the injector. I didn't touch anything in the ignition system while doing that.
  Quoted Text Here  
I should add that I have checked and rechecked the ignition connections to verify the correct firing order; all is well.

My favorite technique is a squirt of starting fluid.
If it starts and dies, you have fuel issues. If it never starts, it is ignition.

  Quoted Text Here  
Rotor and cap?

On Sun, 2 Feb 2014 17:26:09 -0800, "PrecisionmachinisT"
  Quoted Text Here  
No distributor on that vehicle (like most today, it uses Lost Spark - 2 plugs per coil)
What he needs to do is confirm he has spark, and confirm injector operation and fuel pressure. Pull the plugs. Are they clean and white, wet, or black?? Check fuel pressure - is their pressure in the rail with the engine sitting? If not, is there pressure when cranking?
Stethoscope on injectors - do they click when cranking??

  Quoted Text Here  
replace
  Quoted Text Here  
plug
  Quoted Text Here  
badly
  Quoted Text Here  
them
  Quoted Text Here  
Disconnect
  Quoted Text Here  
is
  Quoted Text Here  
valve as
  Quoted Text Here  
to
  Quoted Text Here  
Okay, the reason I mention this is because I just went through pretty much the same thing with my 98 chev v6, turned out that the rotor had shorted to ground after some 248K miles.
  Quoted Text Here  
Wouldn't hurt to crank it over in the dark while looking for extraneous sparks.
  Quoted Text Here  

  Quoted Text Here  
    [ ... ]
  Quoted Text Here  
    [ ... ]
  Quoted Text Here  
    One possiblity occurs to me.
    Doesn't the fuel rail need to be purged after disconnecting it? I think that there may be a Schrader valve (like tire filling valves) for the purpose. If you have air trapped in the fail, the injectors aren't going to work.
    Good Luck,         DoN.

wrote:
  Quoted Text Here  
They generally self purge as they are a recirculating system (bleed-by regulator)

On 2/4/2014 6:43 AM, snipped-for-privacy@snyder.on.ca wrote: ...
  Quoted Text Here  
Don't I recall hearing of a recirc valve that fails to open on a lot of GM of roughly this vintage? Does the Saturn have same issue, maybe?

wrote:
  Quoted Text Here  
Factory service manual says:
Prime fuel system: a. cycle ignition on for 5 seconds and then off for 10 seconds. b. repeat step a twice. c. crank engine until it starts (maximum cranking time 20 seconds) d. if engine does not start, repeat steps a-c.
Did that. Still doesn't start.
  Quoted Text Here  
Yes, it has a Shrader valve. There's pressure in the rail, and when I open the valve, fuel comes out, not air.

wrote:
  Quoted Text Here  
run.
  Quoted Text Here  
Update: pulled all four plugs after about 30 seconds of cranking. Enging still won't start, but all four plugs are wet with fuel.
So that tells me I have an ignition problem, a timing problem, or an airflow problem.
Don't really see how it can be an ignition problem, since I've replaced darn near the whole ignition system: plugs, wires, and coils.
Also: the coolant temperature sensor is a known cause of crank-but-no-start problems on the Saturn S-series. Replaced that too -- it's only $13 -- with no effect.

On 2/4/2014 3:09 PM, Doug Miller wrote: ...
  Quoted Text Here  
Haven't you checked for spark at each cylinder to prove it?

On 2/4/2014 3:48 PM, dpb wrote:
  Quoted Text Here  
Oh, one other issue I've had w/ particularly Champion plugs--if they ever get truly wet, they never fire reliably again.
Mr Goodwrench at the truck shop in town told me that once't upon a time years ago after was having trouble after a sticky float had filled the crankcase on old 4x4 and was still having dickens of time getting it to run even after fixing carb issues and thoroughly cleaning dry plugs...replaced despite their looking almost like new and away she went...

  Quoted Text Here  
Yes, I have. One by one, pulled a plug wire off and put a spare plug in it, cranked the engine, observed spark. All four are firing.
  Quoted Text Here  

On 2/4/2014 4:24 PM, Doug Miller wrote:
  Quoted Text Here  
Then there's no point in continuing down the path of spark unless the spark is terribly weak but not non-existent.
If there's fuel on the plugs after crank and you've dried them first then that would pretty-much eliminate injectors, delivery system for fuel as well. Assuming the fuel isn't contaminated, that would also seem to be eliminated as the problem.
Compression (think you said it checked, don't recall for sure if this thread or another)???
Timing would be a next logical check...

  Quoted Text Here  
Yes, I did check compression. All four cylinders were within spec, and the difference between high and low was about 8 psi.
  Quoted Text Here  
Agreed. The part that really puzzles me is that after replacing all that stuff in the ignition system, it still ran, just ran rough -- then would not start at all after I removed and reinstalled the injector rail.
I'm starting to wonder if I buggered up a wire somewhere, while doing that?

On 2/4/2014 4:40 PM, Doug Miller wrote:
  Quoted Text Here  
Always a possibility.
The one thing I was coming back to mention just as an outside possiblity--the check of the plugs by using another instead of the ones in the engine leaves one with the fact you've checked the one plug is firing and that there's HV applied to the individual plugs but you haven't actually confirmed the four plugs themselves are firing. That's back to the kind of thing I mentioned above about Champion and not working well at all after having once been drowned. I'd go the extra step and actually check the plugs themselves, not just the spare.
Is this the carb-injected or fully injected arrangement?
Have you tried a squirt of starter fluid while cranking?

  Quoted Text Here  
I'll do that tomorrow morning and post the results.
  Quoted Text Here  
Fully injected.
  Quoted Text Here  
Yes, and got a very impressive backfire through the throttle body.

On Wed, 5 Feb 2014 00:36:57 +0000 (UTC), Doug Miller
  Quoted Text Here  
Impressive backfire through the throttle body implies firing when the intake valve is open. Are you sure the timing is correct? If the timing has been changing the car would run worse and worse. It may be that the timing finally changed enough at the same time as when the fuel rail was opened and maybe that's why it no longer starts. And the starting failure at that time was just coincidental. Eric