By accessing this website, you acknowledge that Edmunds and its third party business partners may use cookies, pixels, and similar technologies to collect information about you and your interactions with the website as described in our
Privacy Statement, and you agree that your use of the website is subject to our
Visitor Agreement.
Comments
Tuesday: "Hi Steve, Just leaving you a message to let you know that our robotic parts replacer has now determined that its previous diagnosis was based on false data, so please come back in for another look at the brakes
Friday: "Steve, our computerized scheduler says that we haven't heard from you in 4 days, and our Electronic Concern Amplifier has been raised to Level Two. Please check in ...."
Meanwhile you never hear of anyone blaming a shop for brake failure, and never see a weird brake failure accident or have any worries about initial build quality either.
Best stick with the "safe" status quo.
about a year ago, the lights on the whole left side of my wife's 2014 Town & Country went out. Took it to the dealer (since its under warranty). They decided it was the BCM. Ordered a replacement and I went back at a later date to get the work done. They did it, I went out to the van to leave, but checked and found the RIGHT side was now out. They took it back in and then decided it was my aftermarket trailer harness causing it, so I gave them permission to remove the harness.
A little time goes by and one of the headlights goes out again. Go back and now they say its a bulb (HID) and replace that. That doesn't last very long, so next they decide its the whole fuse box. This was now October 2015.
since then, I have come to realize that we would lose one headlight whevever it would rain. Finally, come this past Tues, BOTH headlights went out in the rain we had. It has been with them since Wednesday morning and they are at a loss and "waiting for Chrysler to tell us what to do next."
'11 GMC Sierra 1500; '98 Alfa 156 2.0TS; '08 Maser QP; '67 Coronet R/T; '13 Fiat 500c; '20 S90 T6; '22 MB Sprinter 2500 4x4 diesel; '97 Suzuki R Wagon; '96 Opel Astra; '11 Mini Cooper S
Really.
Mine: 1995 318ti Club Sport-2020 C43-1996 Speed Triple Challenge Cup Replica
Wife's: 2021 Sahara 4xe
Son's: 2018 330i xDrive
Service information for HID's has plenty of warnings about the risks of working with the high voltage required to make the lamps work, but nothing about how to test beyond static opens, shorts, and grounds. That's what you see in the majority of "trouble trees" or "flow charts". Instead of having the techs trained to and capable of thinking on their own and finding their own way they send them on a series of static checks that take way longer to perform than they techs get paid to do. Meanwhile, if they were taught the kinds of routines that top techs around the world use, they would make easy work of your headlights being inoperative.
Here is a zoom of the MAP, Baro, TPS and injector command PIDS.
Now the fun starts. Just what does that 31 psi reading mean, and what would it mean if it was at 28.8 in the scan data? Nothing. The answer is the PCM is lying to the scan tool with either of those data pids. The software isn't written to display implausible signals, besides the MAP sensor in this car can't even measure the boost pressure correctly to give the PCM even half a chance. BARO is learned at key on, and then relearned under acceleration via the MAP sensor signal. For example, if the BARO pid was stored at about 4 psi boost and just using rounded off numbers to get to 18 psi that would equal a barometric pressure of 36.64" hg.
At 5psi boost to get 19 psi it is 38.68" hg.
At 6psi boost to get 20 psi it is 40.72" hg.
The BARO calculation is a critical input for the PCM to calculate the engine load data pid. The engine load data, combined with rpm is the first information used in the base fuel injector pulse width calculation. When you look at the data you see the calculated engine load doesn't get above 61%.
As stated previously, this not only has an impact on the base fuel injector pulse, it impacts the spark timing map. (Spark Advance)
Nah, I don't have to wonder. Without that crutch you had no chance and you know it. That's why you stopped where you did. If you had gone to a shop with that information pretending you arrived at that diagnosis the logical response would be, so why are you here and not fixing this yourself?
You repeatedly push the idea that the techs should just do what "the machine" tells them to do. If you knew so much, why didn't you point out that the BARO pid listed was false (or substituted) data? Your comment was " I suppose the signals could be changed - i.e., what happens if the barometric pressure drops to 28.8?" That statement can be interpreted in any number of ways except for the one that it really needed to be. There isn't anywhere to go beyond the point of proving what is wrong, this "cannot be repaired" within the scope of what a typical shop is supposed to do or even be able to do. If you understood that, then you would have known that going to a shop wasn't a plausible outcome of your "diagnosis".
BTW. Engineers cannot write software that can take every possible failure into account because it isn't even possible for them to anticipate every possible failure. That means they cannot write a program that could then analyze for every possible failure. Heck they can't even write a trouble tree that can guide someone through a loss of system communication, let alone write a program or build a tool that can do it. There is no trouble tree written to assist a technician (or anyone else) in analyzing this Jeep's failure. The best that anyone could do would be to write a program that would have identified the MAP sensor signal as faulty. It would not have been able to discern why.
As an example, anything you see on the Internet has already been filtered by a human mind. You cannot see more than the software is designed to see. How interesting! In the Internet we have the "Dark Internet" and in the universe we have "Dark Matter". Perhaps in cars we have a "Dark Diagnostic".
My car won't start. Instead of towing it to a tech, I ask the car why it won't start. Or the OnStar tech or whoever is at the other end of the signal asks the car why it won't start. The software won't need more descriptive info (cranks, no crank, click, no click, gas, no gas, spark, no spark...). It'll be able to tell what parameters are no longer "within range" simply by observation of previous operation of the car. More likely, it'll identify and perhaps even fix the issues whenever those parameters are less than optimal so that you never actually have a no-start issue.
We're getting there already with readiness monitors that track emissions.
Without drill-down testing, a computer won't know diddley about how to fix a car. And no computer can do point to point wiring testing for instance, at least not with the way cars are built now.
Ah, wait, here's where your robots come in!
You'd have to build cars in an entirely different manner to allow a computer to do what you'd want it to do.
Code readers are pitiful - they only tell you that an 02 sensor is bad. Poll a few thousand (million?) parameters and you'll get to the pain point a lot faster.
SD does not operate by the book, but it is not without logic or patience.
Basically, Step 1 is Eliminating Wrong Answers
And like I say - pitiful tool.
Being able to bypass the damaged portion of a chip is nice, but that's a far cry from a self healing circuit as one would have to be in the context of a wire, wire harness, or component failure on a car.
It's one thing for someone to say it, but did you ever see someone explain how to do it? Step #1 isn't just eliminating wrong answers. To do that one would have to know what all of the potential answers are right from the start, and that just isn't possible to do. To try and eliminate "wrong answers" you could spend hours doing random pinpoint tests, which is essentially what a lot of trouble trees make you do.
'11 GMC Sierra 1500; '98 Alfa 156 2.0TS; '08 Maser QP; '67 Coronet R/T; '13 Fiat 500c; '20 S90 T6; '22 MB Sprinter 2500 4x4 diesel; '97 Suzuki R Wagon; '96 Opel Astra; '11 Mini Cooper S
Making progress - at least people realize that such things aren't just science fiction.
The "normal" diagnostics for inoperative headlights is pretty straight forward. Using the scan tool the tech has to pull trouble codes from all of the modules and see if any are generated that are associated to the reported symptom. Then depending on what the tech found, one plausible course would then have been to access the CCN ( instrument cluster) data to see what the switch commands were. Then to the TIPM to see if it is receiving the commands from the cluster, and then finally the outputs from the TIPM to the light assemblies. This is all done with the scan tool, there is no pinpoint testing is performed at this step. Remember they don't get paid enough to do that.
Using this routine the majority of failures would be confirmed and testing like that probably did indicate that the TIPM was faulty, so it was ordered and replaced. It is not a stretch to believe that after the replacement and programming of the TIPM all of the exterior lights probably worked correctly, so there would have been nothing else to search for. Well, until of course another issue was then observed.
Question? How much did they charge to do the rest of the diagnostics and to remove your aftermarket towing harness? FWIW. The tech was only paid for that based on a percentage of what it cost you. Wait. Why "Again"? There is a difference between all of the lights going out on one side of the vehicle and a headlight going out. That's a very important detail and it goes hand in hand with something shifty was asking about yesterday. It's a mistake to associate any new failures to previous repairs when doing diagnostics. Think of this as an Algebra problem. Solving for one variable is much easier than it is to solve for multiple variables. You mean the TIPM again, right? But apparently replacing the lamp fixed the symptom when it was first replaced, is that correct? While he doesn't put it in these words, this is exactly where Steve's ideas lead to. How do you like a shop that is run by Steve? It's cheaper, and yet more profitable for the owner at the consumer's expense for talent when talent is what is really needed.
At this point any diagnostics should be done on this issues own merits. That means the tech has to not know, or forget that he/she knows anything that was done in the past until the exact failure of the affected circuit is pinpointed. Then and only then does the previous history come back into play.
Why is there a hard wire going from the headlights to a black box under the dash in the first place? Why doesn't each lamp have its own integrated box? Now you have two discrete points of communication failure plus 4 feet of wiring subject to environment failure. Got rats?
Try something like this instead of wiki. http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&ved=0ahUKEwjqppTUp53LAhWGKB4KHUt4DfAQFgg2MAI&url=http://sottosgroup.beckman.illinois.edu/nrs131.pdf&usg=AFQjCNFlewVsW24rh-_bGK164rg8KYAN3Q&sig2=M6hYPzxU8WAhvMb_rqBWdw Look at cell phone chargers that you don't have to plug in. They charge the phone by the charger having half of a transformer inside it that is creating a pulsing a magnetic field. That induces power inside the phone (the other half of the transformer) producing an electrical current. . The phone has to be very close to the charger in order for this to work because of the properties of how magnetic fields induce a voltage into a strand of wire. These properties are described by Lenz's Law. https://en.wikipedia.org/wiki/Lenz's_law and Faraday's Law https://en.wikipedia.org/wiki/Faraday's_law_of_induction
Some people like to take both laws into play at the same time. https://en.wikipedia.org/wiki/Electromagnetic_induction
The short answer is in order to transmit enough power to any remote units (such as the headlights) you would need a very powerful magnetic field and it would impact anything else that wasn't sufficiently shielded. What's more if you had other cars around yours doing the same thing, these fields would intermingle and the results of that would be dependent on the exact phasing of the pulsing fields and only predictable on a case by case basis.
BTW, the power to generate such a magnetic field would exceed the power that a car presently requires.
Cars are going to be talking to each other and to the road like crazy "soon". No reason why the discrete headlamp can't communicate with other systems without wires. Wires would generate their own magnetic fields so sticking with copper has its own issues.
What do have now - 14,000+ moving parts in a car? And a bunch of other parts glued together with four or five thousand welds? It's a wonder you can drive a car off the production line, much less expect a tech to be able to fix them when they break.
I don't think self-diagnostics is the future of the automobile for any number of reasons. One, the automobile is not a closed system--there are too many variables from outside the organism to predict the resulting malfunction; Two, the automobile is cldearly evolving towards non-serviceability.
Assuming anyone will bother to own a car when that happens.
We already have self-diagnostics though. OBD stands for on-board diagnostics. It's just in its infancy.
Could a circuit really self heal some day if it fails? What would that look like as it was happening?
Answer, Remember the Tin Whisker phenomenon? https://en.wikipedia.org/wiki/Whisker_(metallurgy)
That leads to a problem that would have to be overcome.
To create self healing circuits one would essentially have to be able to make tin whiskers (or something similar)occur on demand, and only on demand when and where it was needed. It would also have to occur fast enough to restore the circuit without the initial growth being overloaded by the circuit current or intentionally leave the circuit disabled for the period of time that the repair growth would require.
Next question would be. What would that look like to the operator.
Ans. A random failure that is not repeatable.
Operators next question, will it happen again? If so when?
Ans. Good luck trying to answer that.
What kind of circuit failures would this repair?
Ans. Only opens or high resistance.
Are there other failures that could still occur that it would not repair?
Ans. Grounds and shorts.
Could self healing circuits potentially cause a circuit failure if something goes wrong?
Ans. Tin Whiskers are known to cause random circuit failures aka random grounds or shorts. There is reason to expect that an incorrect circuit could be created by the technology which could cause an incorrect ground or short circuit to be created.
So why do we want to try and do this?
Ans. "We" don't. VBG
I went by there and got a printout of the history. You are correct, it was the TIPM. The BCM was my word since I had forgotten exactly what they said way back when.
Here is the history in brief:
12/27/14: Customer states drivers headlamp is out. Tech inspected and found TIPM not powering left side. Part ordered.
(now, to give context, the light came back on before they called to tell me the part was in, and were were very busy for a long time, so I did not go back for a while)
4/25/15: Customer states one headlight not working. TIPM broken internal no signal to left side of vehicle. Installed module; all working correct.
(this is where I go to leave and find both front marker and tailights on one side not working)
4/25/15: Customer states tail and side marker not lighting up. Found fuse to be blown. Installed new fuse and found smoke coming from trailer wiring. Disconnected trailer wiring per customer. All lights working.
(So I think we can chalk this up to an isolated incident unrelated to the headlights, would you agree?)
8/8/15: Customer states passenger headlight is out. Head lamp ballast is broken internally. ordered part.
8/14/15: HID ballast broken internally. Removed bumper, passenger side head lamp assembly, ballast. Upon removal, found ballast and connector filled with corrosion. Used terminal cleaner and electronic cleaner to clean the ballast terminals. Also applied dielectric grease to the terminals. Installed new ballast, reinstalled headlight and front bumper. Verified repair.
10/3/15: Customer states both front headlights out. See history. Both front headlights out---TIPM broken internal. Removed and replaced TIPM.
(This is interesting to find out. She had told me on the phone they were replacing the "whole fuse box")
And that leads us to where we are today with the van being in the shop since Wednesday.
'11 GMC Sierra 1500; '98 Alfa 156 2.0TS; '08 Maser QP; '67 Coronet R/T; '13 Fiat 500c; '20 S90 T6; '22 MB Sprinter 2500 4x4 diesel; '97 Suzuki R Wagon; '96 Opel Astra; '11 Mini Cooper S
How long will it be until the front and rear lights are an LED panel and lens that reconfigures itself as needed for brights, brake, reverse, fog, turn signals, cornering lights, parking lights, flashers? Seems a bit silly to cram in 20 different bulbs on the corners for those functions.
This underscores one of the reasons that the tech has to test and prove what is going on right now, and forget about the reported history while doing the diagnostics. What you wrote here tells a much different story than your first post did.
What did you pay for the diagnosis and removal of the trailer harness? Nothing? Consider that not charging you means that the tech got ripped off. Everybody has a limit to what they can tolerate and that kind of thing eventually chases the talented people right out of the trade and justifiably so. Now you need the kind of talent that would make solving this just part of a normal days work, and it is obviously absent there now.
When dealer management finally learns that just because someone like myself could solve this in less time than it takes a tech to do a set of brakes that in itself doesn't set the value of the effort. Think about this, the tech makes more money doing brakes then they do when performing diagnostics. They make more money flushing brake or transmission fluid then they do when required to perform diagnostics. If I'm right about your trailer harness, the tech didn't get a dime for solving a legitimate issue that wasn't even a warranty responsibility.
And, yes, it sounds a bit different now. 2 TIPMs and a ballast replacement later and I still have the problem. I wonder if they found the TIPM dead again and trying to figure out why that keeps happening.
'11 GMC Sierra 1500; '98 Alfa 156 2.0TS; '08 Maser QP; '67 Coronet R/T; '13 Fiat 500c; '20 S90 T6; '22 MB Sprinter 2500 4x4 diesel; '97 Suzuki R Wagon; '96 Opel Astra; '11 Mini Cooper S
https://www.eecs.mit.edu/ If you don't understand why the chip would not detect tin whisker growth, specifically, then the only thing that will help is education. The chip could detect if a circuit voltage is not within tolerance, but it would have no way to know why. Until you price out a repair. They can do that now and in fact are with a number of models. Now a headlight or even a side marker or turn signal replacement costs over $1000 on some of those cars instead of the price of a given bulb and a few tenths labor.