I have owned a 1999 Jeep Grand Cherokee for about four years now. I love the thing but have been experiencing a lot of problems with it over the past few months. At first, it would just shut down while driving down the road. It didn't spit and sputter and then quit, just a clean and quick shut down - almost as if the key had been turned off. Sometimes it would fire right back up, other times it would only crank but not fire. But it always started back up after a few (15-30) minutes.
About a month ago, I paid almost $1,000 to have my computer replaced. It was running really rough, missing and sputtering, and the check engine light was on. Every time they hooked it up to read the code it would show a different code but all were related to a cylinder misfire. The number of the cylinder would continually change though. After checking the coil pack, spark plugs, etc, the mechanic said it must be a bad computer and replaced it.
Since then I have had major problems getting the thing to start at times. I can leave the house in the morning and it will start fine, stop and start the car several times throughout the day with no problems, but then - BAM - it will refuse to start. Again, it will crank normally but never actually turns over and fires. After waiting a few (30-45) minutes, it will start right up as if nothing has happened.
I've taken it to several mechanics, but each tell me they have no idea what the problem is. Any suggestions would be a major answer to prayers and greatly appreciated.
P.S. Momma of two kids who have about a billion things to do in a week so a reliable vehicle every day is really important. Anything you have to offer would be greatly appreciated. : )
May, 26, 2011 AT 7:47 PM
Most likely there was nothing wrong with the Engine Computer. That's what a lot of mechanics fall back on when they can't figure out the real cause. That's because General Motors DOES have a real lot of expensive computer trouble and that's what they're used to finding. Chrysler products, at least up to that time period had very little computer trouble.
There are two sensors that commonly cause the problem you described. They are the camshaft position sensor and crankshaft position sensor. Normally a diagnostic fault code wll be set in memory to direct the mechanic to the right circuit, but if no code is being set, the Jeep dealer has the Chrysler DRB3 scanner that does the best job of displaying live data while driving the vehicle. Once the engine stalls, it will display "no" or "present" for the signals from each sensor while you crank the engine. Many independent shops also have that DRB3 because with a plug-in card it can be used on any brand of car sold in the U.S. Starting with the '96 models.
It is very common for those sensors to become heat-sensitive when they fail. Once they cool down, typically in a half hour to an hour, they will work again. This is so common, your mechanics should be aware of this as it happens a lot on all brands of cars and trucks.
I can tell you how to test it but that is the only way you're going to find your problem.
All "crank, no start" conditions are approached in the same way. Every engine requires certain functions to be able to run. Some of these functions rely on specific components to work and some components are part of more than one function so it is important to see the whole picture to be able to conclude anything about what may have failed. Also, these functions can ONLY be tested during the failure. Any other time and they will simply test good because the problem isn't present at the moment.
If you approach this in any other way, you are merely guessing and that only serves to replace unnecessary parts and wastes money.
Every engine requires spark, fuel and compression to run. That's what we have to look for.
These are the basics that need to be tested and will give us the info required to isolate a cause.
1) Test for spark at the plug end of the wire using a spark tester. If none found, check for power supply on the + terminal of the coil with the key on.
2) Test for injector pulse using a small bulb called a noid light. If none found, check for power supply at one side of the injector with the key on.
3) Use a fuel pressure gauge to test for correct fuel pressure, also noticing if the pressure holds when key is shut off.
4) If all of these things check good, then you would need to do a complete compression test.
Once you have determined which of these functions has dropped out,
you will know which system is having the problem.
May, 26, 2011 AT 8:06 PM
Thank you SO much for your reply. It's a very frustrating thing to be a chick who knows diddly about vehicles and is, therefore, at the mercy of a mechanic. It sucks to know that I might have forked out all that money for something that wasn't needed, but I can't tell you how EXCITED I am to finally make contact with someone who can offer some sort of remedy. Two more question if you don't mind. Are these two sensors expensive (like the computer), or are they fairly reasonable? Also, how difficult/lengthy are they to make? (The last fellow to replace the computer had my car for over three weeks! I was a bit frazzled by the time I got my jeep back.) Again, thanks SO much for the info. : )
May, 27, 2011 AT 1:55 AM
I do not know why Doc ain't come back yet
I did some investigating for you
Some people believe in only "Genuine" expensive parts
I'm not one of them! I have gotten a bad parts before (non working from the start genuine or not!). But I have not had any problems out of "Cloned Parts" that I have used.
As far as changing these puppies, the cam sensor really looks EZ--I do not know about the crankshaft sensor. Its sorta hid, and looks like it may be a bit difficult to get to (see pic) Both may be EZ and you could do this yourself! A repair manual is a real +, I use 'em constantly.
As far as cheating the mechanic, there are other inexpensive routes. My favorite is the feller who works at the self service, Ma and Pop, Small Auto Salvage Yard. He'll slap 'em on for $20 and a cold soft drink!Another is the Technical School Auto Technology Department. If you can schedule it, around here there's a $15-$20 charge on various things (visit the instructor and ask)
As far as me chiming in---I'm a novice with this newer computerized stuff. Caradiodoc is the "PRO", Whatever he says, GOES!
I'm just attempting a little help here, many are lost with fixing their rig or finding resources, many times the answer you get is too short and sweet, and TOO Technical
I may go TOO OVERBOARD. But that's just me!
Here's the locations of both sensors. My 2nd Jeep
Hope this does help. Somehow
May, 27, 2011 AT 3:13 AM
Doc and Medic-
A HUGE thanks to you guys for the helpful info. It's great to finally feel like their might be an end to the sitting in various parking lots for half an hour every other day waiting for my jeep to start. (Ha, ha - Kind of funny when it's not actually happening.) While I'm in no way even close to being mechanically minded enough to attempt this repair, I do have a few good fellows around me that can probably get the job done with the info I got from you guys. Hopefully they will attempt the fix this weekend, and I'll be sure to post the results. I really can't tell you how relieved I am to have a possible answer to such an annoying issue. God bless you guys, and I hope y'all have an awesome weekend! : )
May, 27, 2011 AT 3:17 AM
Oh, and one more thing.
I was able to get a code after reading the various posts on here. Even though my check engine light is not on, I tried the little trick with turning the key three time and seeing if a code came up. I got "P0320" which I understand to mean the PCM is not getting a signal from the crankshaft sensor. (I know that's probably not the technically-correct way of saying it, but hopefully you'll get the idea.) So, I'm going to try replacing that sensor before the camshaft sensor. Again, a big thanks!
May, 27, 2011 AT 4:11 AM
Could be that maybe it came unplugged or the contacts are corroded.
Maybe no need to spend $$$!
You can get little condiment packets of "dielectric grease" at the auto parts store. Smear some on your contacts before you plug 'em back in.
Good luck, come back to this post and update us.
This is an on-going soap opera for us. We like to hear the rest of the story!
Thanks for coming to 2carpros
May, 27, 2011 AT 6:36 AM
Sorry to take so long to reply. This site has stopped working on my main computer.
Happy to hear you have a code related to the crankshaft position sensor. Usually that will cause the Check Engine light to turn on, but that only MUST turn on when the detected problem could have an adverse effect on tail pipe emissions. A non-running engine won't cause excessive emissions.
This sensor sits on the top driver's side of the transmission, right behind the engine block. It is very important to install it with the proper air gap. If that gap is too large, the engine will hesitate, stall intermittently, or not start at all. If the gap is too small, it is possible for it to be broken from being hit by the spinning flex plate.
New sensors from the dealer come with a thick paper spacer glued to the end to set that air gap. Some aftermarket sensors have a thin plastic rib molded onto the end to set the gap. As soon as the engine starts, the paper spacer will slide off. The plastic rib, if used, will wear down. In either case, it's job is done. Anytime a used sensor is going to be reinstalled, a new paper spacer must be used. If there is a partial plastic rib remaining, it must be cut off, then a paper spacer is used to set the gap.
The reason for bringing up that story about the spacer is every once in a while I read where a sensor got broken because someone didn't know how important that air gap is.
Here's a photo from rockauto. Com showing what the sensor looks like, and a drawing from the Chrysler web site that shows where it is located.
May, 27, 2011 AT 1:19 PM
Sorry, thought of a couple more questions.
Would the bad sensor also cause my jeep to run crappy? Ever since the computer was replaced, it has run really rough and even sounds rough when it idles. I've also noticed that sometimes when starting up a hill, it will miss really bad (almost like it's going to die completely) and then all of a sudden it will get a surge of power and shoot forward? Could this be related to the sensor as well? Again, it doesn't do it all the time.
Also, should I go ahead and have both the crankshaft and camshaft sensors replaced since they are both relatively unexpensive? I remember having to have an oxygen sensor replaced a couple years ago. They only replaced one, and not even a week later they had to go back and replace the other also. And, do you think the replacement of the sensors is something your average person with a limited knowledge of mechanical things should attempt? Thanks again for your guidance and wisdom. : )
May, 27, 2011 AT 10:56 PM
First of all, if you are close to a larger city, you will likely have a salvage yard where you pay your buck, then throw your tool box into one of their wheel barrows, and you can spend the whole day there looking for parts. That would be a dandy place to find a similar vehicle to practice on. If you are near Indianapolis or between Ohio and southern Georgia / Alabama, there is a real nice chain of yards called Pull-A-Part. Their yards are very clean and well-organized, and the people and customers are very friendly. You can search their inventory on the internet too. I've been to at least a dozen of their yards in Nashville, Memphis, Knoxville, Indianapolis, Montgomery, four in Georgia, etc. I think they have about two dozen yards altogether. I think all you'll need to remove a crankshaft position sensor is a 10mm wrench but there might be other stuff that's in the way and has to be removed as well. You might even consider buying the used sensor. Parts are REAL inexpensive there. I bought a rust-free sliding door and lift gate for my '88 Grand Caravan about two years ago for less than a hundred bucks for the pair. A similar type of yard in St. Louis wanted $250.00 just for a rusty lift gate.
Don't start replacing more things than necessary. When you consider that new parts can be defective, that air gap is critical, and anything else that can go wrong, you are inserting another variable into the equation each time you change something. If the camshaft position sensor was intermittent, there would be a different fault code stored in the computer.
As for your oxygen sensors, there is a different explanation for that. The Engine Computer is constantly running tests on all of the sensors. Some of those tests involve comparing their readings to those of other sensors. Here's just a few examples: 1. After the engine has been off for more than six hours, the engine coolant temperature sensor and the outside air temperature sensor had better be reading the same temperature.
2. If the coolant temperature sensor says the engine is at 40 degrees when you start it, it had better be reading considerably higher after the engine has been running for five minutes.
3. There will be no signal coming from the vehicle speed sensor when it's standing still. That's a normal condition. There can also be no signal coming from it if it is defective. That's not a normal condition. If the computer sees that the transmission shifted to a higher gear or there is a load on the engine, it knows the vehicle is moving and there had better be a signal coming from the speed sensor.
In all of those examples the computer knows what to expect from each sensor for a given set of conditions, but it has to also know that "given set of conditions". If there is a problem detected with one sensor and a code is set in memory, the computer knows it can not rely on that sensor's readings to compare to other sensors. With nothing reliable to compare to, it will not set a code for the second item. That is likely what happened with your oxygen sensors. You actually have four of them, two on each half of the engine. The front ones measure if the correct amount of fuel is entering the engine. When everything is working correctly, both sensors will report nearly the same readings. As they age, one of them can begin responding slowly. The computer watches that but as long as the readings stay within acceptable limits, no code will be set in memory. If both sensors start to respond slowly, the computer will make fuel delivery adjustments that are not the best but you might not notice that fuel mileage has dropped or tail pipe emissions have increased. Eventually one of the sensor's readings will be bad enough, or unbelievable enough to the computer, that it finally sets that first fault code and turns on the Check Engine light. The second sensor is just as old and just as tired but it hasn't quite failed yet. Also, since there is a problem that was detected for the first sensor, the computer has nothing reliable to compare to the second sensor so it will never set a code for that second one.
Eventually you get the sensor replaced to take care of the fault code in memory. Now the computer has a reliable signal to compare to the second sensor, and it realizes that second one is not performing properly. That's when it sets the second code a week later. Unfortunately that means two trips to the shop, but there was no way to know during the first visit that another problem was going to show up. That's one reason mechanics get the blame very often for causing new or additional problems. Even if they are sharp enough to catch the second problem after they make the first repair, they get accused of "finding more things wrong as long as you keep approving the additional repairs".
General Motors is real bad about that with their anti-lock brakes. It is real common for wheel speed sensors to give problems within 15,000 miles, but only one fault code will be set in memory. Fix that sensor, then a different code pops up. Fix that one, then a different code shows up again. That is more likely to happen when the first code was ignored for many months giving those other things time to fail. Smart mechanics will ask you how long that warning light has been on. If only a few days, you likely have just that one problem. If it's been on for months, they know to prepare you for multiple items and lots of diagnostic time.
To address the poor running and whether it could be related to the crankshaft position sensor, the answer is yes, but it s usually intermittent. The signal from that sensor is either 0 volts or 5 volts and it switches instantly between those two states. Those are the only two conditions the computer recognizes. If the failing sensor starts producing 1 volt or 4 volts, the computer MIGHT understand that it meant 0 and 5 volts, but at some point it s going to get confused.
Think of a Don t Walk sign. The work Walk is always lit up, but the word Don t turns on and off. Those are the only two acceptable conditions. If the word Don t glows very dimly when it should be off, you would understand it s okay to walk. Same is true if it doesn t quite get full brightness when it s on. You know to not walk or you ll get mowed over by a bus! But what do you do when Don t is lit up half brightness? Some people will walk, some won t, and some will step off the curb, then change their minds. That s the confusion the computer sees when the signal doesn t turn fully on or fully off from a failing sensor. It can't decide whether it should fire a spark plug or an injector, or if it is supposed to be waiting for a better pulse.
It s rare for the sensor to produce those confusing signals for very long. Typically they go from working properly to total failure very quickly, with perhaps a few minutes of producing confusing signals. That said, it is also possible for a different type of failure to take place where the signal never reaches the full 5 volts. That is usually aggravated by the slow engine speed during cranking so it s more common to cause a no-start condition, but it can cause a hard start / hesitation / sputtering problem too. There s two ways to verify it s the sensor causing the running problem. The fastest way is to install a new sensor, then see if the problem is gone. Professionals are trained to avoid that method because it involves buying parts that might not be needed. The other way is to connect an oscilloscope that lets you view the waveform and see if it is correct. Most people don t have access to a scope and they can take a while to set up and connect. In this case, since this is a known commonly failed sensor, there s a good chance it will solve the problem so my recommendation is to stick it in. If it doesn t solve the problem, hang onto the old one for a spare.