Pontiac Solstice Forum banner

Replacing crankshaft sensor

7.8K views 84 replies 8 participants last post by  Old School  
you are going to need a Tech2 or clone to program the new sensor to the ECM. It may or may not need to be coded. it's a crap shoot on that one as it just depends on the sensor that is put in. I recommend doing it any time the sensor is replaced just to make sure that there is not going to be any issue with it.
 
  • Helpful
Reactions: wmf246
save yourself a lot of headache. Get an etorx socket and pull the studs out for the intake manifold. Then you can lift and roll the manifold to be able to see and unclip that harness.

You will end up scratching the #4 intake runner in most cases if you do not pull the studs. The studs only have something around 22ft/lbs of torque on them and you can use a pair of pliers or small vice grips to take them out because of the low torque.
 
  • Helpful
Reactions: wmf246
you don't always have to relearn the sensor. There is a possibility of having to. I suggesting having the tool on hand in the event that happens. It may not happen. but it might. it depends on the sensor and how fast it responds to the position change. Just because a sensor is made for a vehicle doesn't mean there isn't differences that should be calculated for. If they didn't need to be accounted for then you would not have the ability to learn the crank position sensor.

The connector that is plugged into the throttle body has a lock on it. You need to use one of you fingernails to catch the lock and push it in the direction the clip comes off in. The lock just slides. once you have the lock undone there is going to be a button that you need to press. This is the hard part because the button is recessed into the clip on the same side as the lock. Pushing it with your finger is not that easy to do. It usually requires using some kind of a pick tool or screwdriver to help in pushing the button.

actually it could be one of these 2 types.

Now I am second guessing myself.

Image




I believe it is this type tho.

Image
 
There are 2 bolts that are hidden.. did you get both of them? If standing on the driver side looking at the engine from the side one is just south of where the hose connects that goes to the brake booster. The other one is near the front of the engine. It is holding the dip stick tube in place.

If you don't pull the studs that hold the intake to the engine you will NOT be able to lift up on the intake unless you undo the harness that goes to the throttle body. You will have to undo it by feel. If you take the studs out then you can lift the intake up on the engine side and get it so that it is standing up almost vertical. You will then be able to see the clip at the throttle body and be able to undo it.


Look at the MAF sensor. That one is going to be just like the one that plugs into the throttle body. You will see a red thing on it and that red thing pulls out to unlock it. Then you will be able to push the small button on it to release the catch on the clip. Even when doing that it might be difficult to get the thing off. It is in a location where dirt and dust end up getting into the clip and that makes it hard to remove. You have to wiggle it slightly to get the thing to come off..

For some crazy reason I think you can use either a 7mm or an 8mm socket to get those studs out, You will have to get the socket lined up and tap it with a hammer to push it on but it will go on. once you get the socket onto a stud it's pretty easy from there to loosen it and wiggle the ratchet to get the socket to come off while the stud is still threaded in a little bit. You can unscrew it the rest of the way by hand.
 
that nipple sticking out is where the brake booster hose attaches to the intake. Just south of that "hidden" is a 13mm bolt that you have to remove. There is a bracket that runs from the intake manifold down to the engine. It is simply a support bracket. It's that bracket that doesn't allow you to easily pull the intake manifold out if the studs are in place in the engine.

If you remove the intake manifold without taking the studs out one of the studs ends up scratching the inside of the #4 intake runner. It also makes it easier to unplug the throttle body when you remove them because it allows you to roll the intake so the head mounting surface is pointing up. at that time you will be able to see the throttle body harness so you can undo the lock on it.

Use a large male allen socket or allen key to unscrew the PCV from the intake manifold. It is located between the #2 and #3 intake runners. It looks like a hole between those runners but there is a plastic piece down in there that does come out. I believe when you shake it you should hear a rattling sound.@HGadget knows if that is the case.\ I recommend replacing that PCV because it is a lot of work to get to it and the part is not expensive. It just makes sense to do it now while you have it apart. The PCV valve is considered to be a consumable item, like an air cleaner.
 
As a note.. I forgot to mention this earlier. There are these round black bushings where the studs and bolts go through the intake manifold. do not loose those as they are to keep the intake manifold positioned properly on the head. To get them out stick a bolt on the hole and move the bolt so it is angled and then pull. they will come right out. They do fall out as well.... Since I forgot to tell you about them grab a flashlight and shine it down the intake ports. I want you to actually do this for 2 years. One is to see if any bushings fell out and dropped into the intake ports and the second reason is to look at the valve stems and see how much carbon buildup is on the stems. If there is an excessive amount of carbon you might consider having the valves cleaned by either dry ice blasting or walnut blasting. The latter you can do at home if you have a compressor. That carbon buildup will cause issues later on and it should be addressed if there is a lot of buildup.
 
chances are where you problem is will be at the ECM connectors. I recommend taking the clips on and off several times. There have been users that have reported issues with corrosion on the pins. taking the connector off and on a few times will usually solve any issues like that. It's an easy check to do.

I believe the CKP sensor is grounded on the head between the firewall and the engine. You might want to take the bolt out and clean the rings and also where the bolt goes into the head. A small wire brush will do the trick there. Also clean the threads on the bolt as well as the underside of the head to ensure you get the best possible connection.


It is highly unlikely that you have a wiring issue due to the proximity of the sensor to the ECM. I have not seen anyone report any issues with the wiring at all. I have seen bad sensors, corroded pins and bad grounds.
 
  • Like
Reactions: Rixfix
we want to check resistance. That's the symbol in the gray area on the left side of the dial. The bottom one.

Terminal 2, set the meter to resistance and touch one of the test leads to ground and the other to the pin. Let me know what the resistance is.

Next 2 tests you need to have KOEO (Key On Engine Off)

Terminal 1 is the 5V. Change the meter to DC volts and move the test lead that is on terminal 2 over to terminal 1. Tell me what the voltage is.

Continuity is the setting above the resistance.

terminal 3 we can test for both continuity and also voltage. Tell me what both of those readings are.
 
  • Like
Reactions: wmf246
This is what I have for the crank sensor as far as DTC's go. So for P0335 There is a circuit check and a rationality check. The circuit check runs as soon as you put the key in and turn it to on. If there is a problem the check engine light would be lit immediately after starting. It would not pop up after some time. I do not think that you have any kind of a shorted circuit because of that behavior. If you read the criteria for code column you can see it says the condition must exist for 4 seconds continuous or a total of 50 seconds if it is a problem that is not constant. So if you car is having an issue and the DTC doesn't light up within 4 seconds of starting the car it is not going to be a shorted condition.... If it takes a minute or 2 for the condition to happen I am going to lean towards a dirty signal the ECM is receiving. That dirty signal can be caused by a bad sensor (which can happen right out of the box) or corrosion/bad connection on the terminal pins either at the ECM or the sensor.

turn the key on and clear the codes with the code scanner. Then start the car and see how long it takes before the check engine lights up. Actually time in seconds how long it takes. It is OK to increase the RPM on the engine, in fact I encourage you to do so.

I really do not think your issue is a shorted wire or even a bad ECM. If you open the hood and look at the wiper cowl on the driver side near the driver side wiper arm attachment, you will see a pretty large hole in the cowl near the seal. I am willing to bet you that hole is clogged with tree dirt and when it rains the water ends up not draining through that hole. It ends up pouring right onto the ECM and clips. This is not uncommon to have happen. The clips are water resistant, they are not designed to keep water out if in a submerging type condition. While the ECM is water tight the pins in the clips are not.



Component/SystemFault CodeMonitor Strategy DescriptionPrimary Malfunction Signal and CriteriaThreshold ValueSpecific UnitsSecondary ParametersEnable ConditionEnable ValueUnitsTime RequiredFrequency of ChecksCriteria for CodeMIL Illuminatioon
Crankshaft Position SensorP0335circuit continuityno engine signal
but phase signals available
0RPMcamshaft revolutions detected>12countsapprox 5 sec0.01 sec continuous4 sec continuous or 50 sec cumulativeimmediate
rationality checkreference gap missing ( sensor signal but no reference ) >6GAPScamshaft revolutions detected>12countsapprox 5 sec0.01 sec continuous4 sec continuous or 50 sec cumulativeimmediate
P0336rationality checkunexpected re-synchronization ( loss of reference mark ) >2600countcamshaft revolutions detected>12countsapprox 5 sec0.01 sec continuous4 sec continuous or 50 sec cumulativeimmediate
rationality checkintermittent loss of engine speed signal >28countcamshaft revolutions detected>12countsapprox 5 sec0.01 sec continuous4 sec continuous or 50 sec cumulativeimmediate
P0338rationality checkdifference in counted teeth between reference gap position events >250crankshaft teethcamshaft revolutions detected>12countsapprox 2 sec1 per rev continuous4 sec continuous or 50 sec cumulativeimmediate
 
  • Helpful
Reactions: wmf246
are you clearing the code with your scan tool? The code won't go away on it's own, or if it does it will take a while before it does. It's not an immediate thing. You need to reset the codes using your scan tool and see if it comes back.

The 4.2 volts is much better. I am betting you have some **** on some of the terminal pins at the ECM. Use the scan tool and clear the codes and then start it up. time how long it takes for the engine light to come on. report back with that time. Easy to do and you don't have to take anything apart to do it.
 
  • Helpful
Reactions: wmf246
The car has to start in order to do the relearn of the CKP. so we need to get the car running for ya if it's not.
 
  • Helpful
Reactions: wmf246
The reason why a sensor might need to be calibrated is because of the signal voltage. The ECM already knows what the pulses are supposed to be. that is not why the sensor gets learned to the vehicle. What the ECM is actually learning is what the ON and OFF voltages are when the pulses occur. and also the switching time of the sensor This is where sensors will differ and if there is enough of a difference it can cause issues when the sensor is being read. The ECM allows for an error margin or deviation from the on and off pulses but if a sensor is outside of that range then a DTC gets set. Not all sensors are made equally as we all know. Cheaper components and variations in the manufacturing process can easily place a sensor outside of that acceptable error amount. This is when the ECM would need to relearn the sensor.
 
  • Helpful
Reactions: wmf246
It's in the ECM programming. Look under ECM replacement and it will tell you about how to learn the CKP sensor.

Doing a relearn is not always a necessity when replacing the sensor and that is why it is not mentioned. The service manual is written with the idea that all parts are going to be OE parts and not aftermarket parts. Using an OE part it might be a 1 in a million possibility of needing to do a relearn. Where as with an aftermarket sensor, who knows...
 
  • Like
Reactions: wmf246
check relay # 26 in the under hood fuse box. That relay is what powers both O2 sensor heaters, the Purge valve solenoid, the MAF and also the VAC pump relay if you have the vacuum pump when GM added mid year in 2008.

If that relay is not working properly or the trigger to that relay is not making it from the ECM to the relay then those sensors would show as a ground fault on the signal wires. The fast that all of the sensors are reading the exact same kind of an error and the only common point between all of those sensors being that relay that is where I would start my investigation.

Pull the relay # 14 and plug it into the #26 location. Orientation doesn't matter. Just make sure that the relays both have the same number on them. The #14 relay is your rear defroster.
Drive the car and see if that solves the issue. If it does then you have a bad relay. If not then follow the directions below.


In order to test you need to have the relay plugged in when performing the test. Take a piece of stranded scrap wire, stranded wire is best, something thin. Strip one end so there is a decent amount of exposed wire, it needs to be long enough to wrap around one of the pins a coupe of times. Strip the other end only a little bit, this is the end you are going to use for testing. For this next bit if you don't have a needle style set of test leads you can use that wire. We need to test the 4 locations in the fuse box the relay pushes into. Use the short stripped end and push it onto one of the slots or use your test lead and put it into the slot. DC volt test so ground the other test lead. You are looking for 12 volts. There is going to be only one pin that has 12 volts on it (key off for this test). Once you find the location of the 12 volts remember it because we don't need to do anything else with it. We need to test the other 3 locations and that needs to be done with the relay inserted. So wrap the long end of the wire around one of the pins that goes into one of those other 3 locations. Insert the relay and then test the wire for 12 volts. You need to locate the pin that reads 12volts, once you have located that leave the relay in and go and turn the key to the ON position. retest that wire, it should no longer be 12 volts. If it is then we know that the signal is not getting to the relay from the ECM. If you have a GXP inspect pin F12 in connector X1 for any corossion or the terminal pin not being fully seated into the clip. You can also test that F12 pin to see if you have 12V on it. If you don't then you know it is a wire issue. If you do then the problem is going to be either the ECM or the connection between that pin and the coorsponding pin on the ECM.