hard start...long crank SOLVED
03-22-2013, 01:50 PM
#1
Safety Car
Thread Starter
hard start...long crank SOLVED
as a FYI to everyone that has the hard to start syndrome in the mornings...do the key on then off then on again and crank thing...
or the just crank it for 5 seconds till it gives and fires...
I've dealt with that crap for several yrs and done a TON of research and study on the problem and THOUGHT that I had learned something. I think I was WRONG
The last couple days I've been working on some mods to the cooling system since I have to pull the goofey water pump off since its got a case of slippery impeller disease...so I was doing some other system maint and decided to reroute some heater lines. I got rid of most rubber and use metal tubing in most places and so I can have a flush/vent valve inline.
Anyway, I was mocking up a manifold for the relocation of the temp sensors from the front of the intake to place them in the hot line from the intake to the heater core. This gets me more room in front and access to the sensors anytime. They were hidden before...not good.
anyway, I did move them to the new manifold about where the heater hose used to live behind the a/c compressor, so they sit in there with the flush valve and the usual heater lines with the rear 1/4" egr water line at the end so everything flows nice and in a line where it should be...
After a brief start up to look for leaks, it hit me that the car started wayyyy too easy. Did it again. Fired right up. Went back later after it got cold and did it again. Started right away. No grinding, no key on/off BS...it just started...New starter BTW thats great ! supplied by a member that sold it to me...great mini that spins the motor easily...Sounds like a dodge minivan starter but thats ok...
anyway, it seems that somehow, someway,. the temp sensors have been at fault. It may have been the CS switch, or plug...I dunno. I moved both and cleaned both sensor probes and connections so I can;t say which works better. Regardless, the thing starts like it should and I'm now a whole lot happier about getting in it in the morning.
So, for all you guys that are still having that hard to start first thing problem...go back and mess with the CS switch and the eng temp sensor. One or the other might be behind your 1st start, hard start issues...
worked for me !
or the just crank it for 5 seconds till it gives and fires...
I've dealt with that crap for several yrs and done a TON of research and study on the problem and THOUGHT that I had learned something. I think I was WRONG
The last couple days I've been working on some mods to the cooling system since I have to pull the goofey water pump off since its got a case of slippery impeller disease...so I was doing some other system maint and decided to reroute some heater lines. I got rid of most rubber and use metal tubing in most places and so I can have a flush/vent valve inline.
Anyway, I was mocking up a manifold for the relocation of the temp sensors from the front of the intake to place them in the hot line from the intake to the heater core. This gets me more room in front and access to the sensors anytime. They were hidden before...not good.
anyway, I did move them to the new manifold about where the heater hose used to live behind the a/c compressor, so they sit in there with the flush valve and the usual heater lines with the rear 1/4" egr water line at the end so everything flows nice and in a line where it should be...
After a brief start up to look for leaks, it hit me that the car started wayyyy too easy. Did it again. Fired right up. Went back later after it got cold and did it again. Started right away. No grinding, no key on/off BS...it just started...New starter BTW thats great ! supplied by a member that sold it to me...great mini that spins the motor easily...Sounds like a dodge minivan starter but thats ok...
anyway, it seems that somehow, someway,. the temp sensors have been at fault. It may have been the CS switch, or plug...I dunno. I moved both and cleaned both sensor probes and connections so I can;t say which works better. Regardless, the thing starts like it should and I'm now a whole lot happier about getting in it in the morning.
So, for all you guys that are still having that hard to start first thing problem...go back and mess with the CS switch and the eng temp sensor. One or the other might be behind your 1st start, hard start issues...
worked for me !
The following 3 users liked this post by leesvet:
03-22-2013, 08:09 PM
#2
Melting Slicks
All too often we see trouble posts where the poster never returns with a fix. Great to see a post of this nature.
The following users liked this post:
ggabesz (07-19-2022)
The following users liked this post:
ggabesz (07-19-2022)
03-24-2013, 06:59 AM
#5
Pro
Thanks for posting the fix
.
Thats one of those hard to find, keep scratching your head problems that can drive you nuts. With the age of the cars the connections get fouled up just enough to cause a drivability problem but not enough to cause a big problem, turn on the check engine light and so on.
Good find
Gary
. Thats one of those hard to find, keep scratching your head problems that can drive you nuts. With the age of the cars the connections get fouled up just enough to cause a drivability problem but not enough to cause a big problem, turn on the check engine light and so on.
Good find
Gary
The following users liked this post:
ggabesz (07-19-2022)
03-25-2013, 05:48 PM
#6
Burning Brakes
The coolant temp sensor reports the engine temp to the ECM/PCM. This is critical for the fueling and spark advance. I recently (yesterday) went through this same thing.
The issue that I have found is that even with a slightly failing or incorrect CTS, the car wont throw a code. Instead, it will use what information it has and try to make things work. It doesnt know that the coolant temp sensor is wrong. It will truly think that the car is at say 180*, when in fact it is around ambient temp.
A bit of an explanation...
The coolant temperature sensor on the water pump (LT-1/LT-4) is the PCM and digital gauge temp sensor. It uses a 5v signal from the PCM. This is sent out on the yellow wire (ckt 410). The Coolant temp sensor is a thermistor. It functions on a non-linear scale. As the temp raises the resistance of the sensor is lowered. The 5v signal from the PCM is sent to the sensor, the other wire of the sensor is a ground (which is shared by your oil temp and TPS and terminated at the ECM/PCM). The computer measures the voltage drop, or the level that the 5v is pulled down. It knows what voltage corresponds to what temperature.
As the coolant temp is provided to the ECM/PCM, the computer can decide how much fuel to give the engine and how to advance the ignition. This also has several other inputs for the fuel and spark, but the coolant temp is critical.
If the sensor reads too low of a resistance, it will show that the engine is warmer than it actually is. If an engine is cold, it requires more fuel and spark advance to start. So if it is warm, it requires less fuel and spark.
A little bit of electronics info... a resistor will always fail OPEN. Meaning the resistance will increase as it fails. Rarely if ever will a resistor fail as a SHORT.
So as a coolant temp sensor fails, its resistance to ground is increased, and the subsequent result is an increased temperature sent to the PCM. This will cause an increase in cranking time, because there is not enough fuel being injected to start the cold engine. The plug will be cold, as will the whole combustion chamber and the ignition will be advanced too far to ignite the mixture. Only after several no-fire events on several cylinders will there be enough fuel to ignite. Even after it ignites, the engine will be running too lean to properly warm up. During this warm up event, the PCM may be in closed loop due to the data from the high reading from the coolant temp sensor. In this case the O2 sensors will see the lean condition and will do what they can to get the right amount of fuel into the engine. Vice, a temperature that is too cold to set closed loop, the PCM will not be using the O2 sensors and you will be attempting to warm an engine with the incorrect amount of fuel and spark advance.
As far as the diagnostics for this problem, they can be found in the FSM, using the flow charts for high/low voltage on the coolant temp sensor. (because it measures the voltage drop, a high voltage would indicate little to no drop or little to infinite resistance to ground, and a low voltage would indicate a direct short to ground pulling it directly down to 0v)
To add to the complexity of diagnosing this problem, I find that the cluster will display a value that could be taken as correct even if the coolant temp sensor is unplugged. This value will even climb and reach a temp that looks correct. Yet this is data sent from the CCM to the cluster which comes from the PCM at a request by the CCM. Only if the ignition is keyed on with the sensor unplugged will you get a diagnostic trouble code from the PCM. If you unplug the sensor while the ignition is already on, you will get the last stored value as your display. (And strangely enough this value will still raise with the sensor unplugged. yeah, wtf???)
Hope this helps. Also, here are the values for resistance vs. temp. This should apply to most GM temp sensors (IAT, OTS and CTS)
The issue that I have found is that even with a slightly failing or incorrect CTS, the car wont throw a code. Instead, it will use what information it has and try to make things work. It doesnt know that the coolant temp sensor is wrong. It will truly think that the car is at say 180*, when in fact it is around ambient temp.
A bit of an explanation...
The coolant temperature sensor on the water pump (LT-1/LT-4) is the PCM and digital gauge temp sensor. It uses a 5v signal from the PCM. This is sent out on the yellow wire (ckt 410). The Coolant temp sensor is a thermistor. It functions on a non-linear scale. As the temp raises the resistance of the sensor is lowered. The 5v signal from the PCM is sent to the sensor, the other wire of the sensor is a ground (which is shared by your oil temp and TPS and terminated at the ECM/PCM). The computer measures the voltage drop, or the level that the 5v is pulled down. It knows what voltage corresponds to what temperature.
As the coolant temp is provided to the ECM/PCM, the computer can decide how much fuel to give the engine and how to advance the ignition. This also has several other inputs for the fuel and spark, but the coolant temp is critical.
If the sensor reads too low of a resistance, it will show that the engine is warmer than it actually is. If an engine is cold, it requires more fuel and spark advance to start. So if it is warm, it requires less fuel and spark.
A little bit of electronics info... a resistor will always fail OPEN. Meaning the resistance will increase as it fails. Rarely if ever will a resistor fail as a SHORT.
So as a coolant temp sensor fails, its resistance to ground is increased, and the subsequent result is an increased temperature sent to the PCM. This will cause an increase in cranking time, because there is not enough fuel being injected to start the cold engine. The plug will be cold, as will the whole combustion chamber and the ignition will be advanced too far to ignite the mixture. Only after several no-fire events on several cylinders will there be enough fuel to ignite. Even after it ignites, the engine will be running too lean to properly warm up. During this warm up event, the PCM may be in closed loop due to the data from the high reading from the coolant temp sensor. In this case the O2 sensors will see the lean condition and will do what they can to get the right amount of fuel into the engine. Vice, a temperature that is too cold to set closed loop, the PCM will not be using the O2 sensors and you will be attempting to warm an engine with the incorrect amount of fuel and spark advance.
As far as the diagnostics for this problem, they can be found in the FSM, using the flow charts for high/low voltage on the coolant temp sensor. (because it measures the voltage drop, a high voltage would indicate little to no drop or little to infinite resistance to ground, and a low voltage would indicate a direct short to ground pulling it directly down to 0v)
To add to the complexity of diagnosing this problem, I find that the cluster will display a value that could be taken as correct even if the coolant temp sensor is unplugged. This value will even climb and reach a temp that looks correct. Yet this is data sent from the CCM to the cluster which comes from the PCM at a request by the CCM. Only if the ignition is keyed on with the sensor unplugged will you get a diagnostic trouble code from the PCM. If you unplug the sensor while the ignition is already on, you will get the last stored value as your display. (And strangely enough this value will still raise with the sensor unplugged. yeah, wtf???)
Hope this helps. Also, here are the values for resistance vs. temp. This should apply to most GM temp sensors (IAT, OTS and CTS)
The following 3 users liked this post by merlot566jka:
03-25-2013, 06:38 PM
#7
The coolant temp sensor reports the engine temp to the ECM/PCM. This is critical for the fueling and spark advance. I recently (yesterday) went through this same thing.
The issue that I have found is that even with a slightly failing or incorrect CTS, the car wont throw a code. Instead, it will use what information it has and try to make things work. It doesnt know that the coolant temp sensor is wrong. It will truly think that the car is at say 180*, when in fact it is around ambient temp.
A bit of an explanation...
The coolant temperature sensor on the water pump (LT-1/LT-4) is the PCM and digital gauge temp sensor. It uses a 5v signal from the PCM. This is sent out on the yellow wire (ckt 410). The Coolant temp sensor is a thermistor. It functions on a non-linear scale. As the temp raises the resistance of the sensor is lowered. The 5v signal from the PCM is sent to the sensor, the other wire of the sensor is a ground (which is shared by your oil temp and TPS and terminated at the ECM/PCM). The computer measures the voltage drop, or the level that the 5v is pulled down. It knows what voltage corresponds to what temperature.
As the coolant temp is provided to the ECM/PCM, the computer can decide how much fuel to give the engine and how to advance the ignition. This also has several other inputs for the fuel and spark, but the coolant temp is critical.
If the sensor reads too low of a resistance, it will show that the engine is warmer than it actually is. If an engine is cold, it requires more fuel and spark advance to start. So if it is warm, it requires less fuel and spark.
A little bit of electronics info... a resistor will always fail OPEN. Meaning the resistance will increase as it fails. Rarely if ever will a resistor fail as a SHORT.
So as a coolant temp sensor fails, its resistance to ground is increased, and the subsequent result is an increased temperature sent to the PCM. This will cause an increase in cranking time, because there is not enough fuel being injected to start the cold engine. The plug will be cold, as will the whole combustion chamber and the ignition will be advanced too far to ignite the mixture. Only after several no-fire events on several cylinders will there be enough fuel to ignite. Even after it ignites, the engine will be running too lean to properly warm up. During this warm up event, the PCM may be in closed loop due to the data from the high reading from the coolant temp sensor. In this case the O2 sensors will see the lean condition and will do what they can to get the right amount of fuel into the engine. Vice, a temperature that is too cold to set closed loop, the PCM will not be using the O2 sensors and you will be attempting to warm an engine with the incorrect amount of fuel and spark advance.
As far as the diagnostics for this problem, they can be found in the FSM, using the flow charts for high/low voltage on the coolant temp sensor. (because it measures the voltage drop, a high voltage would indicate little to no drop or little to infinite resistance to ground, and a low voltage would indicate a direct short to ground pulling it directly down to 0v)
To add to the complexity of diagnosing this problem, I find that the cluster will display a value that could be taken as correct even if the coolant temp sensor is unplugged. This value will even climb and reach a temp that looks correct. Yet this is data sent from the CCM to the cluster which comes from the PCM at a request by the CCM. Only if the ignition is keyed on with the sensor unplugged will you get a diagnostic trouble code from the PCM. If you unplug the sensor while the ignition is already on, you will get the last stored value as your display. (And strangely enough this value will still raise with the sensor unplugged. yeah, wtf???)
Hope this helps. Also, here are the values for resistance vs. temp. This should apply to most GM temp sensors (IAT, OTS and CTS)
The issue that I have found is that even with a slightly failing or incorrect CTS, the car wont throw a code. Instead, it will use what information it has and try to make things work. It doesnt know that the coolant temp sensor is wrong. It will truly think that the car is at say 180*, when in fact it is around ambient temp.
A bit of an explanation...
The coolant temperature sensor on the water pump (LT-1/LT-4) is the PCM and digital gauge temp sensor. It uses a 5v signal from the PCM. This is sent out on the yellow wire (ckt 410). The Coolant temp sensor is a thermistor. It functions on a non-linear scale. As the temp raises the resistance of the sensor is lowered. The 5v signal from the PCM is sent to the sensor, the other wire of the sensor is a ground (which is shared by your oil temp and TPS and terminated at the ECM/PCM). The computer measures the voltage drop, or the level that the 5v is pulled down. It knows what voltage corresponds to what temperature.
As the coolant temp is provided to the ECM/PCM, the computer can decide how much fuel to give the engine and how to advance the ignition. This also has several other inputs for the fuel and spark, but the coolant temp is critical.
If the sensor reads too low of a resistance, it will show that the engine is warmer than it actually is. If an engine is cold, it requires more fuel and spark advance to start. So if it is warm, it requires less fuel and spark.
A little bit of electronics info... a resistor will always fail OPEN. Meaning the resistance will increase as it fails. Rarely if ever will a resistor fail as a SHORT.
So as a coolant temp sensor fails, its resistance to ground is increased, and the subsequent result is an increased temperature sent to the PCM. This will cause an increase in cranking time, because there is not enough fuel being injected to start the cold engine. The plug will be cold, as will the whole combustion chamber and the ignition will be advanced too far to ignite the mixture. Only after several no-fire events on several cylinders will there be enough fuel to ignite. Even after it ignites, the engine will be running too lean to properly warm up. During this warm up event, the PCM may be in closed loop due to the data from the high reading from the coolant temp sensor. In this case the O2 sensors will see the lean condition and will do what they can to get the right amount of fuel into the engine. Vice, a temperature that is too cold to set closed loop, the PCM will not be using the O2 sensors and you will be attempting to warm an engine with the incorrect amount of fuel and spark advance.
As far as the diagnostics for this problem, they can be found in the FSM, using the flow charts for high/low voltage on the coolant temp sensor. (because it measures the voltage drop, a high voltage would indicate little to no drop or little to infinite resistance to ground, and a low voltage would indicate a direct short to ground pulling it directly down to 0v)
To add to the complexity of diagnosing this problem, I find that the cluster will display a value that could be taken as correct even if the coolant temp sensor is unplugged. This value will even climb and reach a temp that looks correct. Yet this is data sent from the CCM to the cluster which comes from the PCM at a request by the CCM. Only if the ignition is keyed on with the sensor unplugged will you get a diagnostic trouble code from the PCM. If you unplug the sensor while the ignition is already on, you will get the last stored value as your display. (And strangely enough this value will still raise with the sensor unplugged. yeah, wtf???)
Hope this helps. Also, here are the values for resistance vs. temp. This should apply to most GM temp sensors (IAT, OTS and CTS)
The following users liked this post:
mrm8480 (08-02-2017)
03-25-2013, 06:49 PM
#8
Burning Brakes
205 is within correct value, but if you dont see some fluctuation, I would suspect the sensors true reading.
The following users liked this post:
mrm8480 (08-02-2017)
03-25-2013, 10:42 PM
#10
Safety Car
Thread Starter
lol...
yeah, BUT..
there are lots of spring fix-it projects ahead. I drive it a little..but its a matter of funds available and time to get all the things done that I want done this year. Regardless, its still as much fun today (wrenching or driving)as it was in 1987 when I got my first.
Thanks !
yeah, BUT..
there are lots of spring fix-it projects ahead. I drive it a little..but its a matter of funds available and time to get all the things done that I want done this year. Regardless, its still as much fun today (wrenching or driving)as it was in 1987 when I got my first.
Thanks !
04-22-2013, 04:38 AM
#11
Race Director
That graph looks suspect to me. I graphed the charted temperatures many years ago and the curve is EXTREMELY non-linear. That's why the temperature gauge shows "LO" below 150°. The curve hits zero ohms at 304°F?
The ECM doesn't use the same temperature sensor as the dashboard coolant temperature gauge.
Here is my version of the chart and graph. I modified the chart to show the degrees Fahrenheit in 10° increments (temperatures on the graph are °C). The temperatures are "backwards" because Excel graphs them in the order they appear in the table. The gap in the curve is caused by the missing data in the chart:
Last edited by Cliff Harris; 09-24-2013 at 01:44 AM. Reason: Corrected zero ohm temperature.
04-22-2013, 08:21 PM
#14
Team Owner
Member Since: Dec 2000
Location: SE NY
Posts: 90,675
Likes: 0
Received 300 Likes
on
274 Posts
Cruise-In II Veteran
Some interesting info here.
At least on a L98 you can check the ECM CTS against the gauge sensor with a scan too. Since I'm now having longer cranks on my "new" '88, this will be one of my check points
At least on a L98 you can check the ECM CTS against the gauge sensor with a scan too. Since I'm now having longer cranks on my "new" '88, this will be one of my check points
The following users liked this post:
mrm8480 (08-02-2017)
09-22-2013, 06:03 PM
#15
Howdy Folks!!.. 
OK... but what if your problem is the opposite. With my "88" L98 it snaps right off when it's cold... But a bear when it's warm...
Thanks for any replies...
Fast Freddie
OK... but what if your problem is the opposite. With my "88" L98 it snaps right off when it's cold... But a bear when it's warm...
Thanks for any replies...
Fast Freddie
The following users liked this post:
mrm8480 (08-02-2017)
09-23-2013, 12:28 AM
#16
Race Director
If the connection to the CTS is open (or the pins are corroded), you will get this symptom. It could be a bad CTS as well. The ECM thinks the engine is cold and richens the mixture, which makes it hard to start.
You could try putting the gas pedal to the floor, which puts the ECM in "clear flood" mode. That turns off the injectors.
You could try putting the gas pedal to the floor, which puts the ECM in "clear flood" mode. That turns off the injectors.
The following users liked this post:
mrm8480 (08-02-2017)
09-23-2013, 12:03 PM
#17
If the connection to the CTS is open (or the pins are corroded), you will get this symptom. It could be a bad CTS as well. The ECM thinks the engine is cold and richens the mixture, which makes it hard to start.
You could try putting the gas pedal to the floor, which puts the ECM in "clear flood" mode. That turns off the injectors.
You could try putting the gas pedal to the floor, which puts the ECM in "clear flood" mode. That turns off the injectors.
But I'm an Old retired (retarded) Locomotive Engineer Cliff, without a clue as to what the CTS is...
... sorry... I will try that "pedal to the floor" trick.Fast Freddie
09-24-2013, 01:49 AM
#18
Race Director
CTS is Coolant Temperature Sensor. There are two.
One is in one of the heads (side varied with year). That one is used by the instrument panel.
The other one is in either the front of the intake manifold or the water pump (depends on year). That one is used by the ECM (Electronic Control Module -- the computer).
If you have an auxiliary fan in front of the radiator then it is controlled by a temperature switch in one of the heads (side varied by year). If you have two fans behind the radiator then they are both controlled by the ECM.
One is in one of the heads (side varied with year). That one is used by the instrument panel.
The other one is in either the front of the intake manifold or the water pump (depends on year). That one is used by the ECM (Electronic Control Module -- the computer).
If you have an auxiliary fan in front of the radiator then it is controlled by a temperature switch in one of the heads (side varied by year). If you have two fans behind the radiator then they are both controlled by the ECM.
The following users liked this post:
mrm8480 (08-02-2017)
09-24-2013, 05:26 AM
#19
Instructor
Member Since: Sep 2003
Location: Las VEgas NV
Posts: 179
Likes: 0
Received 0 Likes
on
0 Posts
09-24-2013, 06:44 AM
#20
Instructor
Member Since: Sep 2003
Location: Las VEgas NV
Posts: 179
Likes: 0
Received 0 Likes
on
0 Posts
Can someone help me with a little electronics 101.
With an ohm meter and for example the CTS
1. If I jumper the two connections on the CTS I can determine the resistance
How do I check the 5v coming in from the ECM. One jumper on the 5v wire, the other jumper where?
2. The other wire going back to the ECM is the ground, what exactly does that mean. Is that the wire that the ECM receives the resistance from the sensor?
Please be simple. Very little brain in this area.
With an ohm meter and for example the CTS
1. If I jumper the two connections on the CTS I can determine the resistance
How do I check the 5v coming in from the ECM. One jumper on the 5v wire, the other jumper where?
2. The other wire going back to the ECM is the ground, what exactly does that mean. Is that the wire that the ECM receives the resistance from the sensor?
Please be simple. Very little brain in this area.
Last edited by rhinoevans; 09-24-2013 at 11:03 AM.
