CarTalk.com Blogs Car Info Our Show Deals Mechanics Files Vehicle Donation

VW electrical problems

Our 1999 VW Jetta II TDI needs a new instrument cluster every 90,000 miles. It first loses function with a sudden loss of the temperature and fuel guages.



Then the speed/od ometer and tach will function inconsistently.



This will be our third one and I do not want to pay over $600 to VW again. Is there another source than VW?



The VW company knows there are certain years of the Mexican built VW’s that are lemon’s concerning the electrical components.



In fact, the new 2010 Jettas on the door frame plate specifically state that all electrical components are from EU or Cananda.



I do not know how to get VW to take responsiblity for their lemon’s, but we certainly have one.



Course, we get 50 mpg, so what can I say?



On the other hand, I have a long list of problems with the car including repairing much of the plastic with fine wire re-enforced JB weld, an original starter that had no reserve torque, and dropped the systems voltage below 9 volts, and on and on.



If there is a VW employee of high rank, preferably right next to the CEO, reading this please advise me as to what I can do to get VW to compensate me for my wife’s frazzled trust of our lemon!



John H.



It’s going to be impossible to be precise as to what the problem is on your car but I’m having a hard time believing 3 clusters have gone bad. My feeling is that there is an underlying problem affecting the cluster rather than the cluster itself.

To make this post even odder, what sane person in this world would use JB Weld on wiring?

“An ORIGINAL starter with no reserve torque dropping the systems voltage below 9 volts” means a starter problem alright; a worn out one.
How long do you think a TDI starter will last considering the beating (a.k.a. high current draw) it takes?
A worn out starter will drag the voltage down due to that annoying amperage/voltage inversion factor.

At this point, a Lemon it’s not. And I’m real curious as to who is behind this JB Weld business. Jeez.

Dear ok 4450,

I am sorry I did not emphasize a bit more the operative word “plastic” leading into the concept of using JB weld with embedded fine wire. I assure you I do not use it on electrical wire. I have used that combination to fix a glove box hinge, the hinge on the arm rest lid, the clamps holding the window glass of the front doors on the moveable track (which I finally had to buy a new interior door plate in order to get the new clamps to hold the glass, and guess what? The replacement had metal clamps instead of plastic, the cup holder which has been repaired 2 times or 3 times, and I have now given up on it (never used them much and they still broke), shall I go on.

Concerning the starter, I finally paid the VW price because that is the only place I could get the updated part number. All after market suppliers only had the original part number. The superceded number has much more torque with significantly less voltage drop. VW knows the original was not adequate.

How long should it last? I have old (20 and 30 years old) farm equipment with Daf, Detroit Diesels, Cummins, AC, Deutz and other diesel engines. A starter change on any of them is an extremely rare thing on this farm. So that is how I judged the quality of the VW starter.

Fuel tank check valve leaks. Added an inline check valve between the filter and injector pump.

Windshield wipers that drag and barely move. VW dealership said we needed a new wiper motor. I disagreed and disassembled the oscillating shafts on which the wiper arms mount from the mounting system through which they oscillate. Nice design, nice little o rings at the exit and entrance points of the shaft through the mounting, but never lubricated at the factory. I had to press the one shaft out it was so dry and bound. Shall I go on? There actually is a lot more. Like no acceleration at odd times in Chicago traffic. Never diagnosed the root of the problem; changed fuel filters several times, has not done it for a number of years now.

Concerning the instrument cluster. The last time I sent it to a guy in Canada. He did not think anything was wrong with it. I thought too, a connecting plug somewhere was corroded. Most likely the ones under the battery. Could not find a problem. Finally got a new instrument cluster from VW. Has worked fine for 90,000 miles. So I am not figuring out what in the wiring system is affecting the instrument cluster, but replacing the instrument cluster keeps us going another 90,000 miles, so I really ponder if there is something in a connector or the wiring harness causing the problem.

The engine is good in the car. That is a positive. My question for automakers is if VW can make a diesel engine that gets 50 mpg for 35 years or so. Where have the rest of the automotive engineers been hiding? Conservation of energy is a good thing, but the rest of the car has to be dependable, which it has not been.

One cannot compare an automobile starter to a starter used on farm equipment for the simple reason that it’s not a matter of age; it’s a matter of key cycles. In other words, the number of times the starter has been used.
Since the automobile starter gets used (generally) on a daily basis and even repeatedly on the same day the VW starter may see more key cycles in one week than a piece of farm equipment sees in months.

I wished I could be of more assistance on the instrument cluster problem but I do not have a wiring schematic for this model. However, it is my opinion the problem is more likely related to the wire harness (either the connector in the back of the cluster or the connectors in the back of the fuse block) rather than a faulty cluster. (I’m kind of leaning towards the back of the fuse block but that’s just theorizing.)

Without a schematic I’m just guessing (barely) here and pretty much dead in the water.

You’re correct about diesel fuel mileage. Matter of fact, the early VW diesels (Rabbits/Jettas/etc.) would routinely get 55 MPG and we had one guy come in once complaining because his VW “only” got 57 MPG and his neighbor was getting 60. One would think that 55 MPG back then would translate to even far better MPG today.

I really appreciate, ok4450, your interest in my 99 VW problems. This is the most interaction I have had, in some ways, with anyone about this vehicle.

I have the Bentley service manual and have pondered over the circuit diagrams for many hours in the past 4 or 5 years.

What is discouraging is that I think I have each new situation figured out and corrected, then either that or something else happens. So every month, (maybe longer) or so I pull out the service manual to try to figure out the new problem. I have come to carry the service manual in the trunk at all times! My wife drives this car to work and it has let her down too often for her to consider another VW.

I like your hunches at where the problem may lie. So here is another story.

The typical syndrome, in our case, of a failed instrument cluster is the simultaneous loss of function of the fuel guage and temp guage (This happened a few weeks ago at about 5am while my wife was driving to her nursing job on wet/icy/snowy slushy roads before the highway crews were on the job), and the tach and speed/od ometer may or may not start acting up. So if it is connector problems with momentary shorts which affects something permanently in the instrument cluster, the road conditions of this most recent problem sounds likely. However when I have checked connectors they seem dry and clean as they all have quite good sealing properties from the weather.

We have had episodes in the past of a tach that bounces all over the place, may rest at 0 for a while, and so on, then will work fine for months. The story that I want to relate now is the speed/od ometer and the sending unit.

Even when the instrument cluster may be working properly a few years ago the speedometer would go to 0, bounce around, but never went over the actual speed we were driving. This of course jerked the car since the function of the cruise control was affected by interrupted speedometer input. VW said it was a faulty sending unit. I disagreed and bench tested it. Output was continuously proportional with rotational speed. At that point I just about had to assume an interruption on the ground circuit, or the 12 v supply circuit. The surface of the plastic drive gear was unmarred, as good as new. Since the one side of the sender is ground I spliced into that wire at the sender and went directly (bypassing the wiring harness), about 10" or 12" to engine block ground. The new direct ground seemed to help for awhile, but then we had further problems with the speedometer bouncing around. So I about had to assume interrupted power supply. I bypassed the wiring harness and picked up the power supply under the dash right behind the fuse block, went directly to the sender. That solved the problem for about a couple years. Now that the fuel and temp guages do not work, yesterday for a short time, about 15 minutes the speedometer jumped around. My wife stopped and shut it off, upon my recommemdation, thinking something about a computer has to reset itself. After doing that twice or so in about 8 miles the speedometer is working again, but since the guages are not working I will be surprised if we do not have some more of the bouncing speedometer syndrome which will force me to pay VW for another instrument cluster, and then we can get another 90k out of it.

Well, you have heard enough of my story by now, I am sure. In closing I would say that I have never had any piece of equipment or vehicle in my 62 years with the mysteries embedded in the 99 VW Jetta.

I like your thoughts about a faulty power source or ground causing the issue. There may also be a faulty regulator circuit in the dash since you stated that the temperature and fuel gauges are effected. These problems sound like they are really due to a bad power connection somewhere. If you are going to keep this vehicle I recommend you purchase the factory service manual for it so you can investigate the problems yourself. It sounds like you know what you are doing in your testing. Having a service manual is wise investment for problems like this.

The speed sensor could still be a suspect on the speedometer bouncing problem. The fact that it checks ok on the bench does not mean it’s fine all of the time. They can be a hit and miss thing.

Just for the heck of it, have you tried removing the fuse block and removing/installing any wire connectors? Sometimes connectors will develop a scale on them over time and the act of unplugging and plugging them back in can cure a problem.

About all I can suggest is sitting down with the wiring diagram, going over every circuit failure and see if they have a common source of power. Since items like the temp and fuel guages ground through the temp sender and tank sending unit respectively odds are very slim there would be simultaneous failures of all of these things at the same time.

Also try to trace this back to the ignition switch; an item that can fail and is often overlooked. Odds are this is not the problem because it’s likely that other things would also be affected; possible the lighting, etc.

One more thing, I don’t remember if this model uses an instrument cluster regulator or not but could it be you have a voltage problem (erratic alternator or whatever) that is taking out this regulator?

This regulator (called a “pitchfork” in slang) controls the voltage supply for the tmep gauge and gas gauge. It’s a tiny fingernail sized chip with 3 prongs and is usually mounted on the back of the instrument cluster. You might take a look at one of the old clusters and see if it has this item.

The pitchfork is used to stabilize voltage and keep the gauge needles from wandering around. Just a random thought anyway.

(Got another thought but got to go. Will be back tonight and throw this one out there for consideration.)

You might also consider the possibility of an internal wire failure. It is possible to have a length of wire that will show continuity with a VOM or even a test light but will not be capable of passing enough current to operate anything.

You might check the schematic and determine if all of the failed items receive their power from a common point. As a test method a spliced wire could be routed from the power take off point to some junction near the cluster.

I ran into a VW many years ago that would randomly quit running, towed in several times, and it was due to an internal wire fault like this. That car caused several weeks of aggravation before I figured it out.