Alternators, A Good Example of What's Wrong, GM, Ford, and Chrysler

Don't kid yourself. The "new" GM compact alternator that they started putting in cars back in 1987 is a piece of crap. It always has been, and it always will be. It is DESIGNED to fail; and the DESIGNED failure mode is catastrophic - to such an extent that the unit usually cannot be rebuilt - period, let alone by a properly equiped shop. Let GM up a "reconditioned" one in, and you are looking at a $300 + repair bill.

I did that ONCE. The alternater has a 12 month warranty on it. It lasted precisely 13 months. And GM said too bad, so sad - "would you like to book your car in for another non-warranty repair???". Hell no. I went to the local jobber and got a Dixie rebuilt. Ironically, it lasted for enough years that I finally traded in the piece of junk, um, I mean car, on a newer model. A Corsica. 60,000 km warranty. You guessed it! THe compact piece of shit alternator lasted 61,000 km!

After some severe cussing and swearing at GM, my local dealer gave me the part, but I had to pay for a tow, and for the intallation of the part. I was greatful for small mercies.

I have also come to the conculsion that GM, and others, have gone over to "the dark side" of Quality Assurance - otherwise known as "reliability engineering". ANd having done that, they have created designs that will maximize their profit by exploding just after the warrnty expires.

Well my pockets are not that deep. I cannot aford to be paying $600 per month on a car loan so that I can drive a $30,000 piece of JUNK.

I will never buy GM EVER AGAIN. This is the ultimate statement I can make as a consumer.

I now drive Chrysler. And I am very happy with it. Sure there are problems. Most are well known and documented. There is a huge support group, and parts are relatively inexpensive.

If they are now further complicting the alternator designs, then I guess I will just have to keep my old van running for a while longer until they smarten the hell up. K.I.S.S. Keep It Simple STUPID!

/rant off

You didnt have to explain it.. we know what you are talking about.

You are right, of course, the idea that the ECM is smarter than a simple integrated regulator is ridiculous.

Here's what I heard my mechanic discuss with his assistant: If you experience a fuel problem with your new Toyota Camry, they will hook it up to a diagnosis computer. The computer will diagnose a faulty injector. So you tell the service manager to change the faulty injector. Er, sorry sir, the computer doesn't say which injector it is.

You do the math!

With the North American Big three, they have to pay those salaries and benefit and please the stock holders at the same time. So any way to boost the bottom line will be considered, even if it is using less quality parts.

The solution is simple, drive old cars forever. But the consuming populace isn't about to do that because they don't want to maintain vehicles.

That said there is some good as well as abysmal engineering in modern cars, and DIY is possible if you RTFM and have a basic electronic background, which mechanics today do not have, and never did. When they start in on me with their bullshit I always ask them to explain the difference between an NPN and PNP transistor, or if I have a NAND gate with one leg HIGH and the other LOW what the output is, then they get quiet.

You're a cruel man! :)

Bill Putney (To reply by e-mail, replace the last letter of the alphabet in my address with the letter 'x')

In 15 years of driving the Bonneville 305,000 miles - I'm on alternator # 3. The original & the second had growling bearings - other than that, they were charging just fine.

Harryface

Simply put, we have no reason to discern the difference between a PNP and an NPN transistor.... the majority of automotive module controlled circuits are ground side switched, anyway.... nor do we need to understand the nuts and bolts of a NAND gate or any other binary devices.... We are not charged with repair the modules... we are to repair the circuits they control and replace the modules if need be. It doesn't matter to me that NPN transistors are easier (and therefore cheaper) to "build".... but it does matter to me that they are used in voltage divider circuits. While I need to know "what" a module does, I don't really have any need to know "how" it does it.

Having said that, you would be caught at at least as much of a disadvantage if I were to ask you about the design of a Ravigneaux gearset, why it is important to know why a shift is synchronous or non-synchronous or the reason a solenoid "dithers"..... along with the multitude of more urbane "old school" tasks that many "lay people" (including those well versed in micro-circuitry) seem to have a certain amount of difficulty performing without destroying ther parts in the process.

The techs part is to ascertain whether a module is receiving an input (a coherent input) and whether or not it is producing an output (which may or may not be modulated) ...... How the module processes information is of only passing importance to the tasks we perform.

The little diatribe that makes you feel so smug and superior proves absolutley nothing in the grand scheme of things...... It can only make sense that my time (and the time of other techs) is spent on developing skills and experience for the ever-changing offerings in the automotive world, while you can rest assured that a transistor (be it NPN or PNP) will work the same, year in and year out.

Auto repair is a moving target... one only has to compare the level of technology present in todays vehicles to that of vehicles from 10, 20 or even 50 years ago. Your dashboard clock/thermometer/compass has more computing power than the US sent to the moon in the 60s. Do I need to understand how this device processes it's inputs? No.... I only need to be able to ensure that it has adequate power and ground connections and coherent inputs. If it is faulty, I am going to replace it.... not fix it. And the reason for that is "I am a mechanic and I fix cars.... not clocks".

Jim,

The problem that old No-Man is talking about is the disappearance of the modular design.

I understand where your coming from and you are right - however what is happening with the technology in electronics is the "all in one" designs are becoming a big fad item.

This isn't limited to car computers. Take a look at the latest Macintosh. Go into any office supply store and look at printers - quite a lot of all-in-one designs too.

In the olden days, the CPU's used in car computers were pretty slow so you couldn't do much with them. Control a transmission - well slap a computer on it. Control a remote unlock - slap a computer on it. Control a speedometer, slap a computer on it. Tie all of them together with a low-speed buss like the CCD bus and your in business.

Today even the cheap CPU's run lickety-split. So fast in fact that after getting done with determining the next shift point, the computer sits around loafing. So the manufacturers have figued out that they can collapse all the computers into one biggie computer that all the inputs run into and all the outputs come from. It saves money, since you only have 1 computer instead of a dozen. Until that is, something breaks - and then there's no modules to examine the input and outputs, save one super giant big one. Thus making troubleshooting a complex system - which is the process of breaking a complex system into smaller more easily troubleshot systems - next to impossible. The mechanic is reduced to replacing the one super duper computer and crossing his fingers and hoping that will fix it. If it doesen't, well then that customer is likely never to get his car really fixed.

We are seeing the process of collapsing all those computers into a massive single one going on today. They haven't yet got there, but it's coming. And every year more and more systems in the vehicle are tied into each other, making it harder and harder for a repair tech to break them down into subcomponents.

You mean like your little diatribe here? ;-)

I got news for you - the automakers frankly don't want you guys. What they want working on their designs are pure parts-changers. And, like some have observed, they only want their products to last a bit past the warranty then go kaput, so the customer is stuck buying a brand new one.

Take a look at an old time auto service manual. Hell they contained instructions on how to tear down and rebuild the engine, transmission, even the air conditioning compressor. Hell, in the 60's you could get a Sams Photofacts for the car radio that would even show you how to do board-level repairs to that.

Today, the service manuals are mostly full of procedures on how to take out and put back in entire subassemblies. Like the transmission. They expect that if you want that fixed your going to take it to a specialist. Oh sure, you can still get the repair manual from the factory that will show you how to tear it down and rebuild it, but that's not part of the regular FSM.

In the olden days a lot of garages had full machine shops. If you bent a tie rod, they would pull it off and fire up the forge and bend it back, then probably even peen it for you. Today, people buy a ball peen hammer because they think that it's somehow more special than a regular claw hammer for hammering steel, and they don't even know how to use it.

No, the reason is that the procedures for tearing apart the dash to get the damn clock out are so complex these days that you spend all your time in training on how to tear apart dashboards in cars. You are being reduced to the status of a parts-changer and you don't even really see that happening to you. I'll bet that it's been years since you touched a forge, for that matter, or done any real repairs. Your stuck subbing everything out to rebuild houses because the automakers have got their designs so cocked-up that it now takes a specialist simply to change parts.

I'm not saying that Bret is any better with his rediculous PNP/NPN nonsense. Sure, he may understand how the clock actually works - but if he attempted to change out his clock he would probably break some hidden fastener in the dash and end up with the interior rattling like a Mexican Mariarchi. Whereas you could probably get it out and a new one in and the dash would end up looking like it had never been taken apart. But, look man, every year the automakers take a bit more of the brainwork in repairing cars away from your job. The new designs have better than ever diagnostic computers simply because the automakers don't trust 3/4 of the mechanics out there with an oscilloscope probe.

Ted

I agree that the failure rate of GM alternators is too high, but my experiences with alternator replacement is drastically different than yours. I put GM remanufactured alternators and starters in my vehicles and my alternators typically hover right around $100 from the local Chevy dealer. I own 4 GM products (recently down from 5), and so I do a fair amount of parts business with the local dealer. As well, my neighbor is a lead mechanic in that shop, so for years I have gotten discounted pricing. That does help some, but my discounts are not so big that I get a $300 alternator for $100. I've maintained a Park Ave Ultra, a Chevy 1500 Pickup, a Pontiac Sunbird, a Chevy Malibu and a host of other GM cars that I've worked on for other people.

I'm not so sure about the "just after the warranty expires" part, but otherwise I do agree that the parts business is a recognized profit center and failures are higher than they should be.

Well, the good part is that the GM problems are pretty well known and documented as well. That should put their failures on par with Chrysler failures.

Agreed - the type of changes spoken of by the OP seem just plain dumb. But... I know that all car manufacturers are going to bigger and different electrical system designs because the traditional 12v system has reached the end of its life. A lot are looking at 48v systems. I'm sure there will be lots of new designs hitting the streets in the charging systems as these new power sources become mainstream. Oh joy.

disadvantage

Hell, I just couldn't snip any of that diatribe Jim. I sit on both sides of the fence - I do understand NPN and PNP transistors (though I fail to see the relevance of that), and I fix my own cars. I no longer own an oscilloscope and it wouldn't help me if I did, since I don't have any logic diagrams or advanced schematics anyway. I have however, gone to my local Chevy dealer and presented a failure scenario to a mechanic and sought advice. Sometimes his first suggestion conflicts with the total of the symptoms I've observed and we delve in deeper.

I've seldom found such an ignorant mechanic that he could not work through a logical progression of alternative failure possibilities with me. We put heads together and we work at it. I have almost always found that while the mechanic may not understand how things work inside a computer, they certainly do understand what tends to cause certain failures and to a degree - why.

I've experienced that they can typically tell me that "there is a component in there that opens up", or something similar. They demonstrate that they understand to a reasonably finite level of detail, what is really happening and the effect that the failure has on the car as a system. It does not matter to me that he really can't tell me the color code of a resistor or maybe even not be able to tell me the difference between a transistor and a resistor. Hell - what good would it do anyway? What electronic geek is going to be able to replace a PNP transistor inside an integrated circuit anyway?

Bloating one's chest about one's electronic prowess is nothing more than hot air.

Scrape...

Still begs the point on whether 'local' voltage regulation is superior or not.

I say it is... because the fail-safe design is built in!

Excatly when did auto manufacturers start doing this? What makes you think it's only GM, Ford, and Chrysler that does this? How hard would it be to swap it out with an alternator that has an internal regulator?

Exactly where on the internet can I find more information about these new alternators? I am not exactly sure what to search for (is there a generic name for this).

Not the rambling nonsense from this clown again.

If you've been around "these parts" at all, you'll recognize the OP, and expect nothing different from him. Totally "out of the loop"

~snip~

Like all trolls, NN craves attention. From the size of his threads, we seem to be happy to oblige him.

Dave

I would have to agree. The last thing you want in a moving vehicle where there's lots of vibration is a bunch of connections between the regulator and the field coil.

Ted

So true, that's why I decided to keep my '95 Chrysler Concord beyond my normal maximum. As for maintenance it's simple, go to a non new car dealer garage interested in maintaining older cars.

I left for these garages a few years ago because based on the pressure for me to buy one of those ugly (German styling) Chrysler 300s with the drive at the wrong end, I no longer trusted my Chrysler dealer.

It doesn't apply to cars at all.

Take a look at the latest

It is still done that same way today.

Don't know were you get your information from but it appears to be wrong. The only computers that have been combined are the trans and engine control computers. This was supposedly done because they share alot of information from the same sensors anyway. By combining both computers in one box it eliminates alot of duplicate wiring. From what I have seen the manufactures are going to more computers not less. Are you aware of the CAN-buss systems ?

It doesn't matter if you have seperate modules or one big module, when it gets down to trouble shooting at that level you are looking for a wire/wires controlling a single component. You are still checking inputs and ouputs.

Thus making troubleshooting a

Wrong. See above comment.

If it doesen't, well then that customer is likely never

And just what manufacturers are using one giant computer instead of seperate units. It has been my experience that every year cars are equipped with yet another computer, not fewer computers.

And every

The computers are usually tied into each other to save on wiring. For example, you have five computers that need info from the same sensor. You have one computer that proccesses that information and sends it to the other computers over the buss ( buss= 2 wires - twisted pair), You can see the savings in running seperate wires to each computer.

It's not that hard for a trained tech to repair any of these systems. The only problems I have seen is the location of some of the computers, you have to take a good portion of the vehicle apart to gain access to the computer.

I don't believe that at all. If that was the case they would be building cars with all the electronics that they now have. If you don't have a pretty good understanding of electronics nowadays you may not be able to fix anything by replacing parts either.

And, like some

I do agree with this. It seems they are getting very good at making parts last till just past the warranty period.

And things were easier to repair back then. You could take any car to the corner garage and get it repaired. Look at all the special tools required today. You would be hard pressed to find an independent shop that could afford all the special tools for just one car line. Add to that how complicated everything has gotten, and every manufacturer does things a diferent way.

Agree. It's getting so that you cannot even purchase a manual for su assemblies anymore. The only way to get them is to attend the manufacturers training school. If you don't work for a dealership you don't attend the school.

It's getting to be a throw away world, and it's not just cars. As parts prices come down due to mass production and labor rates got higher and higher it's just not economically feasable to repair something. It is usaully cheaper to replace it.

Your dashboard clock/thermometer/compass has more

The new designs have better than ever diagnostic computers because the systems keep getting more and more complex. The better computer self diagnostics are required to repair the vehicle properly because the systems are getting so complex. Every year it takes more and more brainwork to repair vehicles, not less.

simply because

It's just the opposite. An oscilloscope is required more than ever to repair todays computer controlled vehicles. If fact for some repairs, if you don't have a scope or know how to use one, you will not be able to make a proper diagnosis, never mind a repair.

Mike