2002 Jeep Grand Cherokee transmission not shifting stalling hesitation

Tiny
UBERDOG
  • MEMBER
  • 2002 JEEP GRAND CHEROKEE
  • 4.7L
  • V8
  • 4WD
  • AUTOMATIC
  • 140,000 MILES
Hi I have 2002 jeep grand Cherokee 4.7l I had shift issues. I changed the shift solenoid pack I did a filter change and put the atf4 fluid in old fluid looked fine not much of anything in the pan. It seemed to help the shifting issues. I still had engine lite on saying bad throttle position sensor. I replaced that I also replaced the camshaft position sensor and tested the battery voltage. I still had the engine light on and the TPS fault. I bought A remanufactured PCM and installed it. I test drove it and it seemed great light was out. I pulled into my drive way and the engine light came back on but it still seemed to be running well. The next day I drove it and it started hesitating but still shifting then it started doing intermittent hard downshifting. I ran the codes and it showed p601 PCM internal failure and the code for TPS failure. I cleared the codes out and test drove it again. It would not shift the rpms just kept going up the idle was extremely high causing it to slam when I put it into drive or reverse. I got back parked it went to work came home. I disconnected the battery and held the brake pedal down for 60 seconds to hopefully clear it out. After this the engine light was still on but the p601 code was gone and it said p120 code again. I did not attempt to clear the codes out because when I do it will not shift until you disconnect the battery and reconnect it and the p0120code is still there. I ordered a remanufactured TCM hoping maybe this will correct the issue. I hope I am on the right track and would really appreciate your input Thanks Uberdog
Do you
have the same problem?
Yes
No
Friday, October 10th, 2014 AT 11:00 PM

14 Replies

Tiny
CARADIODOC
  • EXPERT
Hold the phone! Blaming the computers for everything is a throwback to the days when GM had a big pile of problems, mostly in the '80s and early '90s, and replacing them solved a lot of intermittent problems. Since then that has been burned into a lot of mechanics' memories.

You could potentially install more problems by replacing the Transmission Computer. It learns the volume of fluid it takes to apply each clutch pack, and it continually updates the shift points to maintain a nice crisp, like-new shift quality. A new computer can cause mushy or sluggish shifts or it can cause harsh dragster-style shifts until that "clutch volume index", (CVI) is relearned. That rarely takes more than a couple of miles or a half dozen shift cycles, but until that happens, there can be excessive slippage in the clutch packs, especially at the mileage you listed, and that can put the system into "limp mode" where it stays in second gear.

All of what you described could be the result of the TPS code. You'll notice that code description did not say to replace the sensor or that it was bad. They never do. They only indicate the circuit or system that needs further diagnosis or the unacceptable operating condition. Throttle position sensors have a very low failure rate. The vast majority of problems related to them are wiring and connector problems. Intermittent problems can act up so briefly that we can't see them with a digital voltmeter, but the Engine Computer can respond fast enough to catch it and set a code.

Where I would start is by erasing the code, leaving the ignition switch on, then wiggling the connectors and wires to see if the Check Engine light comes back on and that same code sets. The typical things to look for are stretched or corroded terminals in a connector, and wires rubbing through on a sharp metal bracket or where a harness slides back and forth on the body as the engine rocks.
Was this
answer
helpful?
Yes
No
+1
Saturday, October 11th, 2014 AT 12:28 AM
Tiny
UBERDOG
  • MEMBER
Ok I hear you. I have checked a lot of the engine compartment connections and wiggled them around. I repaired one broken ground I found underneath the vehicle.I could not get any noticeable difference to take place. I have not replaced the battery but the battery seems to work just fine and shows like 14 volts across the top of it. Should I get a new battery also would putting a ground Wire form the body of the pcm to the frame be helpful or harmful?
Was this
answer
helpful?
Yes
No
-1
Saturday, October 11th, 2014 AT 7:30 AM
Tiny
UBERDOG
  • MEMBER
I replaced the battery checked my connections cleared the codes out. The engine light came back on right away. I drove it around came back scanned it again now I have a p0121 code as well as the p601 code but it is shifting fine. Doesn't the 601 indicate TCM failure.I also failed to mention that my cruise control is not working it lights up but does not work I checked vacum lines they appear fine. I pulled up on the brake pedal but the cruise does not work fuses are all good. Help!
Was this
answer
helpful?
Yes
No
-1
Saturday, October 11th, 2014 AT 10:11 AM
Tiny
CARADIODOC
  • EXPERT
Measure the three voltages on the TPS by back-probing the terminals through the rubber seals. These voltages are only correct when the sensor is plugged in. You should find 5.0 volts on the feed wire, 0.2 volts on the ground wire, and around 0.5 to 0.7 volts on the signal wire at idle. That signal voltage will go to as much as 4.2 to 4.5 volts at wide-open-throttle, but it should do that smoothly. If there's a bad spot in the sensor, the signal voltage will pop up to 5.0 volts for a moment and that will set a code. If you see that with a digital voltmeter, you really have a bad spot. Most glitches occur way too rapidly to be detected by digital meters, but the Engine Computer will catch it and set a code. If the Check Engine light turns on again after you erase the codes, and then work the throttle, it almost has to be caused by the sensor because nothing else is being disturbed or is vibrating. That would be weird because those sensors don't fail that often, and two bad ones would really be uncommon.

For the cruise control you really need Chrysler's DRB3 scanner. It will list all the operating parameters including vehicle speed, which switches are pressed, and which outputs are being activated. Better yet, if all the other inputs and outputs appear to be correct, this is the only scanner I'm aware of that displays "reason for last cutout". That will be "vehicle speed too low" when you're just sitting there, so you have to observe that during a test drive above 35 mph. Things you might find include, "brake pedal pressed", and "ignition switch off". When those are indicated incorrectly, it's usually a wiring or connector problem or an arced contact on the brake light switch. There's up to three different switches in the brake light switch, and two of them are involved with the cruise control.

I've seen code 601 before but I can't remember what caused it. I'm pretty sure that one won't turn on the Check Engine light, so unless there's some additional code, you'll never know that one set. Only detected problems that could adversely affect emissions must turn the Check Engine light on. There's hundreds of potential codes that don't turn the light on.
Was this
answer
helpful?
Yes
No
-1
Saturday, October 11th, 2014 AT 9:04 PM
Tiny
UBERDOG
  • MEMBER
Thanks I will check the voltages tomorrow. I will let you know what I came up with. It seems like everytime I check it I get new codes I just got a 354 code and a101 code Thanks Uberdog.
Was this
answer
helpful?
Yes
No
-1
Saturday, October 11th, 2014 AT 9:33 PM
Tiny
UBERDOG
  • MEMBER
Ok I am back I checked the voltage like you said. I 4.56 on the middle wire 5.23 on the end one other one is.07 when I increased the RPM the center one went to 4.1 but it went down gradually I did not see any rapid drop. Do you think it is odd that when I clear out the codes that the idle goes up and then it wont start shifting again til I drive around the block 5 or 6 times to reset the engine lite now its shifting and it shows a code p0121 again I checked the rest of the wires and the fuse block wires as well as removed the battery tray to check the wires and vacuam lines under there but did not find anything. I also checked in the cab up under the dash and the wires that run along the edge of the vehicle floor but found nothing. What do you think?
Was this
answer
helpful?
Yes
No
-1
Sunday, October 12th, 2014 AT 4:51 PM
Tiny
CARADIODOC
  • EXPERT
You found the problem in your second sentence. 4.56 volts represents wide-open-throttle. As I mentioned in my previous reply, you must find around half a volt on the signal wire at idle.

The throttle position sensor is a "potentiometer" just like those used for volume controls in older tvs. The volume could be adjusted anywhere between "0" and maximum. The TPS works the same way but the range is 0.0 volts to 5.0 volts. The difference here is there are mechanical stops in the sensor that prevent it from going below 0.5 volts or above 4.5 volts. (Those two values are the standard values used when discussing electrical theory of operation. In actual practice no two sensors will be exactly alike and they'll reach different minimum and maximum voltages, but they'll be close). The point is the signal voltage will never reach 0.0 or 5.0 volts due to throttle position. It CAN reach those values due to a wiring problem, and that is what sets a fault code.

The easier one to understand is if there's a break in the 5.0 volt supply. That can be a cut wire, a stretched terminal in the connector, or a break inside the sensor. With no 5.0 volts, you'll only see 0.0 volts everywhere along the carbon strip inside the sensor, so you'll always see 0.0 volts on the signal wire, regardless of its position. Less obvious is when you're missing the ground wire due to the same three potential causes. You'll have 5.0 volts at every point on the carbon strip so you'll have 5.0 volts on the signal wire, regardless of throttle position. 0.0 volts will set the code "TPS voltage too low". 5.0 volts will set the code "TPS voltage too high".

Not to confuse the issue, but even harder to understand is a break in the signal wire. The same three things can happen, but for the sake of this sad story, lets say the wire is cut. You'll still read the correct signal voltage at the sensor and on that wire all the way up to the break, but after the break, on the wire leading to the computer, you'll see 5.0 volts. If you'd connect a scanner to view live data, it would also show 5.0 volts for the TPS, regardless of how much you pressed the accelerator pedal. This will throw most people, even those who understand electrical theory very well.

Inside the computer that signal wire feeds a lot of circuitry before the computer sees it and acts on it. Due to all that circuitry, the voltage coming back out on that cut signal wire could "float" to some random value and it would bounce around all over. As long as those random voltages stayed between 0.5 and 4.5 volts, the computer would accept it as normal, and it would interpret it as how much the throttle is opened. If the throttle was at idle, with 0.5 volts on the sensor's signal wire, and the voltage in the computer floated to, ... Lets say 3.2 volts, the computer would think more fuel is needed.

Besides throttle position, the computer also knows direction of throttle change and rate of change. Snapping the throttle open causes an instant increase in air flow, but it takes a bigger change in momentum to get gas flow to speed up, so older cars used an accelerator pump on the carburetors. Here, based on the rate of signal voltage change, the computer simply requests more fuel by holding the injectors open for longer pulses.

The computer is also firing ignition coils and injectors, and those currents can cause voltage spikes. All of these circuits are tied together, so all those pulses, along with turning on a radiator fan relay, looking at cruise control switch settings, and dozens of other things, all of those changing voltages can influence what the TPS voltage floats to.

This is a case of when someone says their engine runs better when they disconnect a sensor. If the TPS voltage was allowed to float to some random value and the computer tried to run the engine on those values, the amount of fuel commanded would constantly be wrong. By unplugging the sensor, or when the ground or 5.0 volt feed wire are broken, the voltage WILL go to 0.0 or 5.0 volts which are unacceptable conditions. The computer will set the appropriate fault code, and it will know it can't rely on that sensor's readings. Instead, it will inject an approximate value that it has calculated by looking at all the other operating conditions and sensor values. For example, it knows engine speed, and at 2000 rpm, it expects the throttle to be open about one third of the way so it might inject 1.5 volts.

All of this is background for this next item. To prevent the signal voltage from floating to some random value, there is a "pull-up" resistor in the computer. That resistor is so big electrically that it has no effect on the circuit whatsoever, that is until the signal wire is cut. The pull-up resistor is connected to the 5.0 volt supply in the computer. When the signal wire is broken, (or the TPS is disconnected), voltage appears on the computer's TPS terminal through that resistor. The computer sees 5.0 volts and recognizes that as an unacceptable condition so it sets the code "TPS voltage too high", but it doesn't know why it's too high. That's why fault codes never say to replace parts. They only indicate the circuit that needs further diagnosis.

Also, as a point of interest, a "pull-down" resistor can be used the same way. It is connected to ground, so if the signal wire is broken, the computer will see 0.0 volts which is also an unacceptable condition and will trigger a fault code.

All of this brings me to the next problem. I'm not sure I followed exactly what you found on the signal wire as far as it going down to 4.1 volts. You never found it near 0.5 volts which it should have been at idle. A break in the ground circuit would send 5.0 volts down the signal wire. A break in the signal wire would result in 5.0 volts at the computer but you would see the correct 0.5 to 4.5 volts at the sensor. You found excessive voltage but not the full 5.0 volts. I can see how that could happen inside a TPS but that would not be a common failure.

Instead, what I'd like you to do is turn on the ignition switch, remeasure the voltage on the signal wire, then unplug the sensor and measure that signal wire again. With it unplugged, if you find anything other than 5.0 volts, we have to think on it. Next, look at those three terminals to see if they're corroded or stretched. I have a feeling the ground terminal or the signal terminal is not making good contact with the pin in the sensor.

If you look at code 601, it doesn't say the TPS voltage is too high or low, it just says "malfunction". That might be because the signal voltage is still below 4.5 volts but the computer knows it can't be right. Even 4.1 volts represents close to wide-open-throttle. If the engine really was running that fast, the rpm signals would be a lot higher and the MAP sensor, which measures intake manifold vacuum, would be a lot lower. Those don't agree with throttle position signal voltage, and that is likely why this code set.
Was this
answer
helpful?
Yes
No
Sunday, October 12th, 2014 AT 10:47 PM
Tiny
UBERDOG
  • MEMBER
I checked it like you said and here are the numbers I got. Ignition on plugged in the center power wire had 4.56 unplugged it had 4.62 the signal wire had 5.14 plugged in and 5.14 unplugged I started the vehicle and slowly ran the idle up while checking the signal wire 4.19 was what it read.
Was this
answer
helpful?
Yes
No
Monday, October 13th, 2014 AT 11:52 AM
Tiny
UBERDOG
  • MEMBER
I went back and tested the signal wire voltage with the ignition on and I manually operated the throttle when I had it allthe way open I got a reading of 1.2 at the signal wire
Was this
answer
helpful?
Yes
No
-1
Monday, October 13th, 2014 AT 12:19 PM
Tiny
CARADIODOC
  • EXPERT
That's backward. Where do you have the voltmeter's negative probe?
Was this
answer
helpful?
Yes
No
-1
Monday, October 13th, 2014 AT 4:55 PM
Tiny
UBERDOG
  • MEMBER
They were connected right. After you explained all of that. I went out there and tested it a second time. I saw that the voltage was dropping as the rpms went up. I knew that was wrong because you said the voltage would increase. I had installed a brand new throttle position sensor 3 days ago. I thought it could be bad, but it was brand new.
I was pretty sure when I saw the voltage drop it had to be bad. I installed a brand new one the one that cost three dollars more and it was fixed. I really felt it was highly unlikely that it would be bad but it was. I learned a valuable lesson spend a few more bucks and save yourself a lot of grief. I really appreciate your help and sharing all your knowledge and getting me into the right mind set and redirecting me to solve this problem. I learned a ton of new troubleshooting skills with your help. So once again thanks for the help. Uberdog
Was this
answer
helpful?
Yes
No
-1
Monday, October 13th, 2014 AT 6:46 PM
Tiny
CARADIODOC
  • EXPERT
You're welcome. To be fair, I don't think you got a defective sensor. I think you had one for a different application. Thanks to my tv repair background, I took a bunch of these apart to install "bugs" for my students to diagnose. Chrysler products are my specialty and we had a bunch of them that were donated for student learning experiences, and I can share that there isn't much inside them and very little that can go wrong. Every TPS I've played with was spring-loaded to go back to idle position, so you just bolt 'em on and you're done. The issue is though some manufacturers might use the same design but on the other side of the throttle body. That means it has to rotate the other way. The plugs could be the same, or they might put the locking tab on the other side of the plug. That's where you might be able to connect your plug to a sensor that's in effect, backward.

Now, to address the other concerns, are you still getting other fault codes? There is an issue you should be made aware of resulting from disconnecting the battery, (or letting it run dead). You said you're doing that to erase the fault codes. When you do that the Engine Computer will lose its memory, and on Chrysler products one of the things that has to be relearned is "minimum throttle". Until it learns that, it won't know when it has to be in control of idle speed. You may need to hold the accelerator pedal down 1/4" to get the engine to start or stay running, (more is not better), you won't get the nice idle "flare-up" to 1500 rpm at start up, and it will tend to stall at stop signs. To meet the conditions for the relearn to take place, drive at highway speed with the engine warmed up, then coast for at least seven seconds without touching the pedals.
Was this
answer
helpful?
Yes
No
Monday, October 13th, 2014 AT 8:48 PM
Tiny
UBERDOG
  • MEMBER
You are right on with the incorrect part I got to thinking about it and took a good look at the two TPSs I knew it was operating backwards because of the voltage drop but it being defective did not seem right. So upon looking at them I realized 1 the p/n was different so I looked it up. The engine is a 4.7 and its a 02 Cherokee the part I was given was for a 02 Wrangler with a 4.0 so I was sold the wrong part. It looks identical as you said and fit right on there but when I installed it the same as the one I removed. So I put it on upside down. Like you said it was for a different engine and mounted on the other side I feel pretty stupid but it looked exactly the same. I do blame the parts store for part of it. I would return it but I don't have the receipt. I learned a lot through this though. Thanks.
Was this
answer
helpful?
Yes
No
Monday, October 13th, 2014 AT 9:08 PM
Tiny
CARADIODOC
  • EXPERT
Welcome to my world! It's easy to feel stupid AFTER you figure out the problem, but remember that most of the other people you asked didn't know either. This is how I learned best when fixing tvs, and unfortunately I learned a pile, ... The hard way.
Was this
answer
helpful?
Yes
No
Monday, October 13th, 2014 AT 10:26 PM

Please login or register to post a reply.

Recommended Guides