Why Would an Alternator Go Bad? 7 Questions From a Mechanic Who's Made $9,300 in Mistakes

Some questions roll into my bay on a weekly rotation. “Why would an alternator go bad?” “Do I really need a brake drum measuring tool, or can you just look at it?” “Are those cheap E36 headlights actually fine?”

I'm a technician at an independent repair shop—brakes, electrical, cooling, and lighting. I've been in this trade for about nine years, and I've personally made (and documented) seven significant mistakes, totaling roughly $9,300 in wasted budget. Maybe $9,800, I'd have to check my log. Now I maintain our shop's pre-install checklist so the younger techs don't repeat my dumbest moments.

Here's the FAQ I wish I'd been handed on day one. Jump to whatever's biting you:

• Why would an alternator go bad?
• How do you measure brake drums—and which tool is correct?
• Are Hella E36 headlights worth the upgrade price?
• Radiator checks before a Route 66 trip through Tucumcari
• Why did my brand-new alternator die after six months?
• How can I tell if a Hella headlamp is genuine?
• Should I machine my brake drums or replace them?

Why would an alternator go bad?

The short answer: alternators rarely die of old age. They die from one of four things:

• A failed voltage regulator. This module controls output. When it fails, you get overcharging (boiling battery, bright lights) or undercharging (slow crank, warning light).
• A battery with a shorted cell. The alternator runs at full output nonstop trying to charge a battery that can't hold a charge. Heat builds up, the regulator cooks, and the alternator is done. This is the hidden killer.
• Oil contamination. A leaking valve cover gasket drips oil over the alternator. The oil degrades the bearings and electronics. I'm not 100% sure why, but it's usually the driver's side.
• Poor remanufacturing. Some budget rebuilds are just old cores with cheap regulators. They fail in 10 to 14 months.

That last one is a total-cost lesson. I once installed a $140 reman alternator on a minivan; it died in 11 months. The redo cost the customer another $220 in labor plus $150 for the better unit. The quality unit would've cost less overall.

How do you measure a brake drum—and which brake drum measuring tool is the right one?

I've seen people measure a brake drum with a regular caliper on the outer lip. That's measuring the wear ridge, not the friction surface. You'll be off by 0.010 to 0.050 inch—enough to make a borderline drum look safe.

You want a proper brake drum micrometer (or an inside micrometer set). The sequence we teach:

1. Clean the friction surface. Grease, rust, or dust will throw off the reading.
2. Measure about 1 inch (2.5 cm) from the outer edge, at the deepest wear point.
3. Rotate 90 degrees and measure again. Also check for taper.
4. Compare to the discard diameter stamped on the drum.

I once measured a drum, got a reading I trusted, and machined it anyway. It was already at discard diameter. The brake pedal pulsed hard on the test drive, and we redid everything. That mistake cost $890 in redo labor plus new brake shoes. Now “measure twice, cut once” is a literal sheet of paper on our wall, not a motivational poster.

Are Hella E36 headlights really worth the upgrade price?

I have mixed feelings about the cost gap. Genuine Hella E36 headlights run two to four times the price of no-name projector sets. On one hand, I get why budget matters. On the other, I've replaced more than a dozen fogged-up, moisture-filled knockoff housings in the past three years—and zero genuine Hella units.

The difference isn't the brand sticker. It's the beam pattern. Cheap projectors look bright from the driver's seat but scatter light—oncoming traffic gets glare, and you lose the shoulder. A genuine Hella headlamp has a defined cut-off, UV-resistant coating, and proper ECE or SAE approval markings, per Hella's product data (hella.com). You can verify the markings directly on the housing.

Run the total cost. You save $200 on the buy. Then you pay $360 to redo the install after moisture gets in and output drops by half. The “expensive” unit ends up being the cheap one. To be fair, some higher-end aftermarket projector brands do solid work, and I understand buying on a budget. But the no-name $80 specials? I wouldn't take one for free.

What should I check in my radiator before a Route 66 road trip through Tucumcari?

You know the movie Cars? Radiator Springs isn't real, but the road it was inspired by—the Route 66 stretch that passes through Tucumcari, New Mexico—absolutely is. And that section of highway is brutal on radiators in the summer.

Before I let a customer drive that route, I run a quick cooling system check: coolant condition and level, the radiator end tanks for hairline cracks, the fan clutch and electric fans, and a pressure test. That last one is the big one—it takes 15 minutes and catches most cooling leaks before they leave you stranded.

My worst case: July 2019, a customer heading east had a cracked end tank let go just outside of Tucumcari. A local radiator shop couldn't get the part until morning, so he lost a night to a motel plus a tow bill that ate any savings. The upside of doing the pressure test is the certainty of not being stranded; the risk of skipping it is exactly what happened to him. I'd rather spend 15 minutes checking a $450 radiator than betting a $700 tow on it.

Why did my brand-new alternator die after six months?

This is the one nobody asks until it's too late. “But it's brand new!” is usually how the conversation starts. A new alternator that fails fast generally isn't the alternator's fault.

1. Battery health. A shorted cell cooks any alternator, period.
2. Belt tension and pulley alignment. An overtightened belt kills the front bearing in weeks.
3. Ground straps. A corroded ground makes the alternator overcompensate, which shortens its life.
4. The part itself. More than half of the “new” alternators we warranty in a year are budget remans. Don't hold me to this number, but that's what our counter log says.

I once replaced the alternator on the same van three times—no, I'm not gonna pretend I checked the ground strap before the third. That's the lesson: if the replacement fails young, stop replacing parts and start diagnosing.

How can I tell if a Hella headlamp is genuine?

Counterfeit headlights are everywhere. Flea markets, online marketplaces with “verified seller” badges—I've seen fakes in both. Here's what I check:

• Molded branding. “HELLA” should be embossed in the lens or housing. A sticker that scratches off is an instant red flag.
• Approval markings. ECE-approved lamps have an “E” plus a country code (E1, E4, etc.). SAE lamps say DOT. Fakes rarely bother stamping these.
• Packaging and part numbers. Genuine units match the part number on the box, housing, and barcode. Mismatched numbers mean trouble.
• Price. If it's 60-70% below what established distributors charge, it is not a deal. It's a counterfeit.

I learned this by buying a “factory direct” headlight for my own car. The beam pattern was terrible, and on closer inspection the lens said “HLLA”—missing the E. Cost me $75 and a wasted Saturday. Now I inspect every headlamp before it goes on a customer's car.

Should I machine my brake drums or replace them?

Machining is legitimate if the drum is within spec. But here's what gets people in trouble: new drums are not that expensive. When a shop charges $35 per axle to turn a drum that's already close to the discard diameter, that's a bad total-cost decision.

You're paying $70 in machining to save maybe $30 off new drums. Then you're running a thinner drum with less thermal mass, which means more brake fade on long descents. I've seen camper vans lose braking force on mountain passes because the drums were over-machined, and the owner blamed the master cylinder.

My rule: measure with the brake drum measuring tool, check the stamped discard diameter, and if the final machined thickness would be within 0.030 inch of that number, replace instead. The extra $25 now saves a brake job later.

And if a shop tells you your drums “can be turned,” ask them to show you the measurement. A competent tech will be happy to walk you to the bench and point at the micrometer.