Allpar Forums banner

98 TJ electrical issue

28K views 13 replies 5 participants last post by  David Eidell 
#1 ·
i posted this in a different forum, but i think i should've posted it here. i cant see how to delete the original, but i'm desperate for help so i'm going to just post it here too.....

Hi all. My hero and uncle sent me here for help and hopefully someone can, as i am ready to tear my hair out.

i have a 98 jeep wrangler tj 4.0 liter straight 6.

The alternator is not charging the battery. I replaced alternator, didn't fix it. had the old alternator belt tested and it still works. changed the battery since the old one died from the unknown (at that point) issue.

grabbed the old volt meter and checked every wire, ground, terminal, i could find, include the fusible link and everything tested fine.

this morning i replaced the pcm. still no luck. checked all fuses, rechecked all wires terminals and grounds..... still all fine.

this is where i am at..... i do know that when the new pcm was flashed, the mileage wasn't put it properly and doesn't match my mileage in the cluster gauges. but wouldn't that make the pcm just not work and i would lose all the other components (fuel injectors... etc)?

am i somehow frying the pcm's? if i got a pcm that was bad to begin with, what are the chances the "new" one would have the exact same problem.

only other symptom i have is when i run it (on battery alone) after about a half mile, the check gauges light comes on and the gauge needle buries to "9", even though the battery is still putting out 12 volts (or there abouts)

what in the world could i be missing here?

thanks in advance folks.
 
See less See more
#2 ·
Welcome to Allpar. It will need diagnosis. Hold off replacing anything more until some clues can be gathered. It sounds like you have replaced all the major charging system components and all that is left as a possibility is the wiring harness (unless a part is bad).
You also mentioned not being able to get fault codes. The key-dance does not work with all vehicles and you may need to look for fault codes with an OBDII engine code reader. There may or may not be a P0622:
P0622 Generator Field Control Circuit Malfunction (at https://www.autocodes.com/p0622.html )
If there is a voltage regulator or field circuit problem, it should set a fault code and turn on the red 'Battery' light and/or the yellow 'ck eng' light. Are either of these lights coming on?
You can follow the charging system diagnostics with the help of a simple multimeter here:
http://oskin.ru/pub/chrysler-dodge/...ee_Diesel_Versions_Include/english/ezg_8c.pdf

Use the charging system wiring diagrams for wire colors shown in chapter 8W-20 here:
http://gershon.ucoz.com/JEEP/08-W-Wiring_diagrams.pdf

Connector pinouts are shown in chapter 8W-80. Connector and ground locations are shown in chapter 8W-90. Splice locations are shown in chapter 8W-95. Happy hunting. :)
 
#3 ·
Thank you so much for these resources.

i started thinking about sensors and unhooked the battery temperature sensor and now it's jumping between 12 & 13 volts on my gauge. i can't check the sensor it'self till tomorrow morning (my neighbor has the volt meter at the moment).

But no the battery light nor the check engine light neither came on until i disconnected the bat temp sensor.

i very much appreciate the wiring diagrams as i've been blindly grabbing any and all wire & grounds to check.

and thanks for the warm welcome. i hope i can be some value here. My brain gets the logic of my beautiful TJ (Judy) and hope to offer some wisdom to others.

on a funnier note, when i called the service department of one of our local jeep dealerships he said, "98 tj? sure i can help. i can transfer you to sales." yeah, right!!!!!

i will be sure to update on here what i learn from this experience.
 
#4 ·
I did like the TJ as it seems like a sensible size, nimble and simple Jeep. The new ones are large, even when parked next to a Grand Cherokee for comparison.
If the battery temp sensor is disconnected, it should light the red Battery symbol on the dash and possibly set a P1492 code. Is that or the yellow 'ck eng' lamp lit?
P1492 Battery Temperature Sensor Voltage Too high OBDII Engine Light Trouble Code | Engine-Codes.com (at https://www.engine-codes.com/p1492.html )
Many PCMs use that temperature reading as the 'ambient' intake air temperature reading. It also helps to regulate charging voltage as those requirements will change with temperature.
"Jumping between 12 & 13 volts" isn't right. It should 'usually' be a fairly steady 13-14 volts. It sounds like a poor (loose/corroded) connection somewhere in the wiring?
Poor connections (a high resistance) will create heat if it is carrying any larger amounts of current. You may see evidence of burning or melted plastic at the point of poor contact.
Start with clean battery terminals and move to the connections in and under the black plastic fuse/relay box. On the ground side, clean and secure connections are required from the battery to both the engine and body. Sometimes a connection can 'look' clean and tight from the outside, but by loosening and retightening a fastener can restore a previous poor connection. Unseen corrosion can form under the bolt head and washer.
An alternate method of reading codes that is different than the key-dance that may work for you: With the key off, press and hold in the odometer reset button. Roll the key to 'on' (the video version below does the 3 'clicks') and then release the reset button. The gauges should sweep and the warning lights and odometer digits should light for a self-check.
Read the odometer display for any 'P' codes and be prepared to write them down.

View: https://www.youtube.com/watch?v=HOQZRXZeK7k
 
#5 ·
Only code im getting is the 1492 which makes sense cuz i remove the battery temp sensor. I tested it and its working fine. (I used an obdII) now there are no codes.

Couple of new questions....
Should the alternator still put out voltage if it has no hot lead?

And are there any other sensors or something in the gauge cluster (?) that the pcm could be reading to cause it to shut the alternator off? The whole garbage in garbage out thing. And my battery gauge drops to 9 (buried) & check gauges light comes on. If i restart the jeep it shows 12 volts (seems to read fine) but after a minute it drops again and the check gauges light comes back on even though the battery is still putting out 12volts on my meter.

I drained the battery down to 11.25 and when i started the jeep the alternator kicked on for a short period of time. Could hear it whir but the volts never really came up. Couldve been hearing something else but not sure what.
 
#6 ·
Only code im getting is the 1492 which makes sense cuz i remove the battery temp sensor. I tested it and its working fine. (I used an obdII) now there are no codes.

Couple of new questions....
Should the alternator still put out voltage if it has no hot lead?

And are there any other sensors or something in the gauge cluster (?) that the pcm could be reading to cause it to shut the alternator off? The whole garbage in garbage out thing. And my battery gauge drops to 9 (buried) & check gauges light comes on. If i restart the jeep it shows 12 volts (seems to read fine) but after a minute it drops again and the check gauges light comes back on even though the battery is still putting out 12volts on my meter.

I drained the battery down to 11.25 and when i started the jeep the alternator kicked on for a short period of time. Could hear it whir but the volts never really came up. Couldve been hearing something else but not sure what.
 
#7 ·
If the alternator field isn't energized, it won't charge. One side of the field connector should be 'hot' (about battery voltage). The return side of the field to the PCM is a variable resistance to ground determined by the PCM (charging rate).
Both sides of the field circuit must have good continuity.
The large (+) stud and wire on the back of the alternator should be directly to the battery (+) terminal.
You can briefly 'full field' an alternator by grounding the PCM return side to vehicle ground. Voltage should soar to 15-16 volts (only do this briefly) as a test to make sure that the alternator can reach full output. See Test 2 on p. 8C-5.
 
#8 ·
Before I read this i changed the alternator again and got just under 14 volts and was so happy. 3 miles down the road though, same thing. battery gauge buries to 9 and check gauges light came back on.

With that and now your post, i feel like i'm back to a pcm issue. what are the chances that a "new" pcm would have the exact same problem though? i can believe that i would get a bad remanufactured pcm, but i feel like the odds that its the same problem. if my fuel injection wasn't working, or if the obdII reader couldn't communicate with it, it would make a lot more sense to me.

i had to walk away at this point. tomorrow's plan is to tear it all out again and, piece by piece, examine and clean all connectors whether they test fine or look fine. something is getting in the way of charging the battery, i just cannot for the life of me find it.

thank you so much for your willingness to share you knowledge and for putting brainpower to my problem. its extremely nice of you. and i am extremely grateful.
 
#9 ·
I would diagnose what you have for now and not replace any more parts (some for a second time). Let's cross that bridge later, if we have to. Sure, I have seen defective used and remanufactured merchandise out of the box. They can't possibly bench test them for every contingency or function. These aren't brand-new PCMs and are going on 20 years old with unknown histories.
If the battery is going dead, then I have to believe that what the instrument cluster gauge is telling you about 'low voltage' is true.
When the voltage falls and it isn't charging, is the time to open the hood and begin point-to-point measurements with your voltmeter in order to find the culprit.
The alternator may 'whirr' when it is charging (normal) and fall quiet when it stops charging. This is important to listen, see and feel in diagnosis. Sometimes I have listened for the alternator while gently wiggling or poking connectors and wiring. Listen for the singing or not-singing alternator. Sometimes it helps to turn the headlamps on for added load and to watch for changes in headlamp brightness at the same time as the singing. This has helped me find broken wires and connections in the past.
When there aren't fault codes present to make life easier, sometimes we have to go into 'Sherlock Holmes' mode and investigate to see what we have and don't have. It would really help to find and prove a defect rather than guessing at a problem and replacing a part.
Good diagnostic skills should be the beginning of any repair. Learning them may be more difficult than simply unscrewing an old part and trying a new part. Wrong guesses can and do get expensive and frustrating when they don't pan out, even when they are educated guesses. Having non-OEM new parts or still-defective remanufactured parts can set us backwards and upside down if they turn out to still be the problem in the end.
Diagnostic skills are learned. We can figure out what we want to know and learn how to find out the answer. I have seen Wizards walk up to a car and after a couple of tests, point at the problem (I swear that the guy could see the electricity running through the wires). I wanted to learn that skill and it does come with pre-planning (in your head) how to test and what to test for.
Review the wiring diagram for the charging system. The 3 main players are the alternator (generator), PCM and battery with the wiring running between all three. Write down or print out the wire colors that you will be looking for. Know what voltage you want to see at each wire and what might mean a problem.
Sometimes a 12 volt test light is a better tool than a voltmeter as it takes some current to light a bulb and hardly any to show a voltage on a digital meter. This can be deceptive in troubleshooting as a meter may still show a good 12 volts through a corroded connection, but that same connection may not light or dimly light a 12 volt bulb as it can't pass that needed current. That connection might be missed by a meter test where a test light would catch a definite problem. Best of luck to you. :)
 
  • Like
Reactions: AllanC
#11 ·
. . . The alternator is not charging the battery. I replaced alternator, didn't fix it. had the old alternator belt tested and it still works. changed the battery since the old one died from the unknown (at that point) issue. . . . ..
Go back and look at the excellent reference ImperialCrown gave in post #2, wiring diagrams. Look at page 8W - 20 - 2. It shows the schematic layout of the entire charging system and wiring between components. Pay particular attention to the splice between battery and alternator terminal (S136).You may have a poor connection there. Look at the wiring connector at the PCM for the 2 field wires. Make sure there is positive contact between the pins and receptacles and no corrosion or shorted wires. I have separated that particular diagram with additional notes for items to check. There is an additional image that identifies the electrical connections at the alternator.

Text Diagram Technical drawing Line Floor plan
Auto part Clutch part Automotive engine part Font Automotive alternator
 
#14 ·
The old-fashioned simple ECU diagnostics via turning the key off then on three times then counting check engine light flashes

Can be utterly cheated. Without having the system go into check engine open loop.

It means replacing the voltage regulator with one that has a voltage set point compatible with the climate the car operates in.

Warm hot climate like Phoenix 13.9 volts is fine
Medium climate like Dallas Ft Worth 14.0 is fine
Cold climate like Anchorage 14.3 is fine

TRANSPO ELECTRONICS OFFERS AN ADJUSTABLE VOLTAGE REGUATOR

The two existing field wires on the rear of the alternator must pass through a FIFTY WATT TEN OHM power resistor. This means dis-connecting the two wires a the field terminals of the alternator then reconnecting the wires one at each end of the resistor. This satisfies the check engine light issue.

Careful attention must be made to wiring the F540 type ford regulator correctly and this is not a job for amateurs. The regulator HOUSING needs a connecting ground wire from one of the 1/4" mounting holes. The A terminal needs a 14 gauge always hot battery connection, the S terminal must have an ignition HOT but nor accessory hot switched wire to it. The S terminal needs a 14 gauge wire run to one fielf terminal and the other field terminal needs a new wire that is connected to cn excellent chassis negative.

Benefits are it totally bypasses a sensitive ECU connected system. Replacement ECU's are worth more than these old cars.
The voltage setting stays rock solid steady but sags ideally with high temps due to a regulator thermistor

The negatives are obvious. Buying a power resistor and voltage regulator off eBay. Only the Ford RED regulator connector with clips is suitable.
A substantial amount of labor involved in making a new wire harness. Making a quality wire harness is no place for sloppiness.
BUT! It beats the heck out of junking an other wise good run around car because of ECU problems.

And yes, the above has been performed on dozens of mid nineties K-cars with no errors or long term issues.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top