1978 - 1996 Big Bronco  
Sponsored by:
Sponsored by:

Crank no start when cold

Thread Tools
 
Search this Thread
 
  #1  
Old 03-22-2010, 04:22 PM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
Crank no start when cold

My '96 with 351 is sometimes hard to start when it is cold (<40F?). It will crank just fine but won't fire. Sometimes it takes 5 minutes of cranking in 5 second intervals to get it to run and then it will sputter and run poorly for 15-20 seconds. It never does this when it is warm. I can hear the fuel pump prime and then stop like it should when I turn the key to run. I can also hear the fuel pump shut of right after I stop cranking. The battery is brand new and puts out 12.5V in run and 10.5V at crank. There aren't any relevant codes (EGR not opening but that's a separate issue). The IAC is brand new and it does the same thing even if you hold the throttle half open.

Any ideas?
 
  #2  
Old 03-22-2010, 04:29 PM
greystreak92's Avatar
greystreak92
greystreak92 is offline
Post Fiend
Join Date: Sep 2002
Location: Gateway to the West
Posts: 9,179
Likes: 0
Received 7 Likes on 7 Posts
Are you getting any other fault codes?
 
  #3  
Old 03-22-2010, 04:32 PM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by greystreak92
Are you getting any other fault codes?
No, just the P0401 for the EGR.
 
  #4  
Old 03-22-2010, 04:46 PM
greystreak92's Avatar
greystreak92
greystreak92 is offline
Post Fiend
Join Date: Sep 2002
Location: Gateway to the West
Posts: 9,179
Likes: 0
Received 7 Likes on 7 Posts
Ok... might look into that one sooner rather than later for mileage concerns. However, the more likely scenario with the hard start with a cold engine is a sticky ECT (Engine Coolant Temperature)sensor. This sensor won't necessarily throw a code because the computer will only throw codes if it gets a reading that is "out of range" and if the sensor sticks it can stick ANYWHERE within the proper range yet be sending incorrect information because it is still telling the computer that the engine is warmer than it really is because the sensor has "stuck". If you have a multi-meter and are handy enough to test the sensor for resistance, you can determine if this is indeed the problem.

Take resistance reading with the sensor installed but disconnected from the wiring harness. Measure resistance across the terminals of the sensor that run to the GREY/RED and the LT.GREEN/RED wires. The attached chart should give you an idea of what the resistance reading should be depending upon the ambient temperature outside when you do this. Remember that steel/iron not in direct contact with a heat source is approximately 20 degrees colder though.
 
  #5  
Old 03-22-2010, 04:52 PM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by greystreak92
Ok... might look into that one sooner rather than later for mileage concerns. However, the more likely scenario with the hard start with a cold engine is a sticky ECT (Engine Coolant Temperature)sensor. This sensor won't necessarily throw a code because the computer will only throw codes if it gets a reading that is "out of range" and if the sensor sticks it can stick ANYWHERE within the proper range yet be sending incorrect information because it is still telling the computer that the engine is warmer than it really is because the sensor has "stuck". If you have a multi-meter and are handy enough to test the sensor for resistance, you can determine if this is indeed the problem.

Take resistance reading with the sensor installed but disconnected from the wiring harness. Measure resistance across the terminals of the sensor that run to the GREY/RED and the LT.GREEN/RED wires. The attached chart should give you an idea of what the resistance reading should be depending upon the ambient temperature outside when you do this. Remember that steel/iron not in direct contact with a heat source is approximately 20 degrees colder though.
The ECT readings I get via the ScanGuageII are always what I would expect them to be. It always matches the outside air temp first thing and swings to 185-190 when running.

EDIT: btw, I know there are two coolant sensors but I'm pretty sure since the SG2 gets the data out of the OBD2 port that I'm seeing the same data the EEC is and not the dash sender data.
 
  #6  
Old 03-22-2010, 05:46 PM
greystreak92's Avatar
greystreak92
greystreak92 is offline
Post Fiend
Join Date: Sep 2002
Location: Gateway to the West
Posts: 9,179
Likes: 0
Received 7 Likes on 7 Posts
The concern here is that the resistance with the sensor "dead cold" is not high enough to "tell" the computer that the engine is indeed as cold as it is. The ECT doesn't take temperature readings it changes resistance as thermal conditions change. The colder the engine, the higher the resistance. The same is true for the ACT sensor. This won't necessarily tell a fault scanner there is a problem unless the scanner is actually taking resistance readings from the sensor. The problem is not that it gets from cold to hot its that its not telling the computer the truth about the temperature of the engine because it has lost some resistance value with age or some damage. All electronic components change resistance values with age and exposure to the elements.
 
  #7  
Old 03-22-2010, 05:52 PM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by greystreak92
The concern here is that the resistance with the sensor "dead cold" is not high enough to "tell" the computer that the engine is indeed as cold as it is. The ECT doesn't take temperature readings it changes resistance as thermal conditions change. The colder the engine, the higher the resistance. The same is true for the ACT sensor. This won't necessarily tell a fault scanner there is a problem unless the scanner is actually taking resistance readings from the sensor. The problem is not that it gets from cold to hot its that its not telling the computer the truth about the temperature of the engine because it has lost some resistance value with age or some damage. All electronic components change resistance values with age and exposure to the elements.
Sunday morning it was 33F when it gave me trouble. The ScanGauge2 was showing a coolant temp of 33F that it got from the ECU. If the ECT was bad then the ECU would have some bogus reading to share rather than exactly what it should have. The ScanGauge2 doesn't do it's own readings on the sensors. All it knows is what the ECU tells it it thinks it knows, which in this case is exactly right on.
 
  #8  
Old 03-22-2010, 05:58 PM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
I'll add this as well. When it does finally start there's nothing out of the ordinary from the exhaust like it was flooded.
 
  #9  
Old 03-22-2010, 11:08 PM
greystreak92's Avatar
greystreak92
greystreak92 is offline
Post Fiend
Join Date: Sep 2002
Location: Gateway to the West
Posts: 9,179
Likes: 0
Received 7 Likes on 7 Posts
As I said, check the ACT sensor reading too... its the only other thermal sensor on the engine and can have the same effect on things. The ECT and ACT send data to different pins of the ECM. Keep in mind too that with the EGR issue, the ECM is expecting that air flow and not having it is causing the ECM to compensate in some way. It may be opening the IAC a bit more since it knows the EGR isn't there causing a lean condition which in very cold conditions makes starting difficult. Just speculation on my part at this point.

The next step is to check fuel pressure. Although it seems very odd that such problem would only rear its ugly head under cold conditions. Fuel system dryer might help in the event you have a condensation issue causing frost in fuel lines.

If that doesn't get you anywhere, you could have weak coil connections. Cold causes slightly loose connections to weaken further by virtue of metals contracting. It would have to be a coil issue though since everything in front of the coil is monitored by the ECM and ICM and SHOULD throw a fault code.
 
  #10  
Old 03-23-2010, 12:45 AM
Encho's Avatar
Encho
Encho is offline
The Southernmost Mod
Join Date: Dec 2004
Location: Caracas, Venezuela
Posts: 6,902
Received 17 Likes on 14 Posts
Be careful with those 5 minutes / 5 secs. crankings, those might lead to a fire in the wiring on the steering column. If you can't avoid it, don't drive the truck until you have figured out the cause.
 
  #11  
Old 03-23-2010, 05:59 AM
b4hntn's Avatar
b4hntn
b4hntn is offline
Posting Guru

Join Date: Apr 2005
Location: Longview, TX
Posts: 2,018
Likes: 0
Received 0 Likes on 0 Posts
I agree with Greystreak. On my '94 I pulled both the ACT and ECT and put them in the oven beside a good thermometer. Then increased the temp a little at a time. Resistance readings on both were way off compared to the chart so I bought new ones and the engine ran much better.
 
  #12  
Old 04-13-2010, 05:26 PM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
UPDATE -

1 - All the temp sensors check out.
2 - I found out why I kept getting the EGR codes. I had it down to either a vac leak somewhere or the solenoid was bad. I borrowed a mityvac last night and was going to go through and test every line, fitting and solenoid for leaks. The first part I took off was the reservoir. When I turned it over I found that there was a quarter sized chunk missing from the bottom where someone had leaned on it and broken off one of the mounting tabs. I used JB Weld to glue the mounting tab back on and it gave an OBD2 pass today. I think I picked up 1-2 MPG today but I can't tell from just one day.
3 - I put a fuel pressure tester on it yesterday and I get good readings while it's running. It did take longer than I expected to prime up, though. It only had 10PSI in the rail when I got the gauge on - perfectly normal from having the shraeder open. However, it took 6 times turning the key from OFF to RUN to get it up over 30 PSI. Is that normal? I have a new fuel filter to put on when I have time.
 
  #13  
Old 04-13-2010, 06:31 PM
greystreak92's Avatar
greystreak92
greystreak92 is offline
Post Fiend
Join Date: Sep 2002
Location: Gateway to the West
Posts: 9,179
Likes: 0
Received 7 Likes on 7 Posts
I'd be suspicious of the filter with the pump taking that long to get up to normal pressure range. The filter begins to act like a check valve as it clogs up.
 
  #14  
Old 04-15-2010, 09:06 PM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
I pulled the fuel pump relay to depressurize the fuel system before changing the filter. The truck wouldn't start at all without the relay in. My '95 would run for a good 10 seconds like that. After all that it still blasted at least 6oz of fuel at me when I pulled the tank side line off. The gas from the tank side had a yellow tint and lots of little black specks like pepper in it. It seemed to prime up a lot faster after I got it back together. Even with the lines and filter full of air it only took cycling the key to RUN once before it would start. We're supposed to get back to cold weather so hopefully this takes care of it. I'll cut the filter open tomorrow once its had time to evaporate a little.
 
  #15  
Old 04-19-2010, 09:26 AM
mprice's Avatar
mprice
mprice is offline
Elder User
Thread Starter
Join Date: Feb 2007
Location: Chicagoland
Posts: 601
Likes: 0
Received 0 Likes on 0 Posts
45* here this morning and it took about 5 minutes to get started. What finally got it started was turning it to run and back off five times before cranking. That finally got it to catch.

Any ideas?
 


Quick Reply: Crank no start when cold



All times are GMT -5. The time now is 03:29 PM.