Here's an interesting one
08-23-2015, 08:16 PM
#1
Teching In
Thread Starter
Join Date: Mar 2015
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
Here's an interesting one I've been chasing a part-throttle stumble on my car for months.
- It randomly throws pending codes for the knock sensors, P0327 and P0331, both for input low voltage, but never a CEL. If it has a pending code, it clears on the next ignition cycle. I checked this with a cheap code reader.
- Checked fuel pressure with a gauge under load, pressure never changes
- Checked for vac leaks, found none
- Changed fuel filter, plugs, wires, MAF, etc chasing my ***
Tired of guessing and spending money on parts that weren't the problem, I finally decided to buy a better scanner that can read PID information and live data from the sensors. Everything was as it should be, except I noticed the MAP sensor was giving a steady reading of at 2.5psi. Whether the car was off with the key in (it should read atmospheric pressure, 14.5psi) or when the car was running, it always read the same. Sometimes it would jump between 2 and 3psi, but nothing like it's supposed to. According to a tech manual I found for the LS1, the computer will store a code only if the signal voltage is too low or too high. Since mine sits in the middle, it thinks it's fine, even though it's not working like it should.
So the good news is, I found my problem. But here's the interesting part. When I'm driving with the scan tool hooked up and I feel a stumble, I'll look at the graph and the MAP sensor starts working as it should. A few minutes later when I start to feel the car bucking again, I look and the graph goes flat. So the MAP sensor is intermittently reading as it should, and staying at a constant voltage otherwise. The fueling is messed up due to this, and is probably causing the knock sensor circuit issues.
If I park the car while the MAP sensor is working correctly, shut the car off, then restart it, the graph goes back to flatline. It works intermittently, it seems, and no amount of sticking my hand behind the intake to wiggle the wires makes a difference.
The tech manual also says that knock sensor DTCs are only set when the MAP sensor is at a certain voltage, which mine will only hit when the sensor happens to be working correctly.
So the question is, is there a better chance that the sensor is bad, or that the computer itself is intermittently reading the signal incorrectly?
- It randomly throws pending codes for the knock sensors, P0327 and P0331, both for input low voltage, but never a CEL. If it has a pending code, it clears on the next ignition cycle. I checked this with a cheap code reader.
- Checked fuel pressure with a gauge under load, pressure never changes
- Checked for vac leaks, found none
- Changed fuel filter, plugs, wires, MAF, etc chasing my ***
Tired of guessing and spending money on parts that weren't the problem, I finally decided to buy a better scanner that can read PID information and live data from the sensors. Everything was as it should be, except I noticed the MAP sensor was giving a steady reading of at 2.5psi. Whether the car was off with the key in (it should read atmospheric pressure, 14.5psi) or when the car was running, it always read the same. Sometimes it would jump between 2 and 3psi, but nothing like it's supposed to. According to a tech manual I found for the LS1, the computer will store a code only if the signal voltage is too low or too high. Since mine sits in the middle, it thinks it's fine, even though it's not working like it should.
So the good news is, I found my problem. But here's the interesting part. When I'm driving with the scan tool hooked up and I feel a stumble, I'll look at the graph and the MAP sensor starts working as it should. A few minutes later when I start to feel the car bucking again, I look and the graph goes flat. So the MAP sensor is intermittently reading as it should, and staying at a constant voltage otherwise. The fueling is messed up due to this, and is probably causing the knock sensor circuit issues.
If I park the car while the MAP sensor is working correctly, shut the car off, then restart it, the graph goes back to flatline. It works intermittently, it seems, and no amount of sticking my hand behind the intake to wiggle the wires makes a difference.
The tech manual also says that knock sensor DTCs are only set when the MAP sensor is at a certain voltage, which mine will only hit when the sensor happens to be working correctly.
So the question is, is there a better chance that the sensor is bad, or that the computer itself is intermittently reading the signal incorrectly?
08-23-2015, 08:23 PM
#2
Teching In
Thread Starter
Join Date: Mar 2015
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
Here's another question: Would it be detrimental to just unplug the MAP sensor for now until I have the time to pull the intake? If the computer knows the MAP sensor is dead, can it compensate by using the MAF only? Driving around with a bucking car is a pain in the ***.
08-24-2015, 08:38 PM
#4
Teching In
Thread Starter
Join Date: Mar 2015
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
Thanks for the response.
I have an aftermarket aluminum intake, so the sensor is held in with a bracket and a screw. I can't just pull it off the intake because there's no room to unbolt it, I'll have to remove the intake first.
I checked the wiring to the sensor. It has three pins; the outer two are 5v and low reference signals, the middle is the output signal from the sensor. With the key on, I probed the 5v and the ground pins and got 5v. This is good as it means the sensor is getting 5v from the PCM as it should. Then I did a resistance check on the signal wire from the sensor plug to the PCM plug and got .3ohms, which is good as it means the signal wire is intact.
On my scanner, it shows 1.5psi with the sensor unplugged, and 2.5psi as before with the sensor plugged in.
According to an online source, a 0v reading on a 1 bar map sensor signal wire will read as 10kPa, or 1.5psi. So the computer is fine, and the sensor is my issue.
Will report back after I get a new sensor. I just realized that I can just plug it in and check to see if it reads 14.5 like it should, and install it later.
I have an aftermarket aluminum intake, so the sensor is held in with a bracket and a screw. I can't just pull it off the intake because there's no room to unbolt it, I'll have to remove the intake first.
I checked the wiring to the sensor. It has three pins; the outer two are 5v and low reference signals, the middle is the output signal from the sensor. With the key on, I probed the 5v and the ground pins and got 5v. This is good as it means the sensor is getting 5v from the PCM as it should. Then I did a resistance check on the signal wire from the sensor plug to the PCM plug and got .3ohms, which is good as it means the signal wire is intact.
On my scanner, it shows 1.5psi with the sensor unplugged, and 2.5psi as before with the sensor plugged in.
According to an online source, a 0v reading on a 1 bar map sensor signal wire will read as 10kPa, or 1.5psi. So the computer is fine, and the sensor is my issue.
Will report back after I get a new sensor. I just realized that I can just plug it in and check to see if it reads 14.5 like it should, and install it later.
Last edited by mangid; 08-24-2015 at 10:10 PM.
08-25-2015, 01:18 PM
#6
Teching In
Thread Starter
Join Date: Mar 2015
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
Instead of removing the intake and introducing the possibility of a vacuum leak from the intake gaskets, breaking the oil pressure sensor, or spending hours of my time that I don't have, I left the old sensor in place and T'd into a vacuum line to run the new MAP sensor. Cars runs perfect again.
Hopefully this can help someone in the future who has a rough running car with no CEL.
Hopefully this can help someone in the future who has a rough running car with no CEL.
Trending Topics
08-27-2015, 12:40 PM
#8
Teching In
Thread Starter
Join Date: Mar 2015
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
"If it's stupid and it works, it ain't stupid"