Blowing a Diagnosis on a Road Trip
The weekend before Memorial Day, I took my customary road trip down to “The Vintage” in Asheville, North Carolina. This is the biggest vintage BMW event on the East Coast, with 600 cars in the village of Hot Springs nestled in the mountains north of the city, and the event hotel in Asheville is a non-stop, three-day hoopla where walking round the parking lot is as much fun as the official event itself. I’d missed it last year due to a family health issue, so I was looking forward to returning.
In addition, I decided to drive Hampton, my 49,000-mile survivor BMW 2002. I’ve written quite a bit about Hampton in these pages, describing how I’d bought the car from its original owner in 2019, how I revived it while taking care not to disturb its remarkable originality, how it didn’t sell on Bring a Trailer because people may say that they love survivor cars but what brings the money are shiny powder-coated vapor-honed mirages, and how I gradually warmed to the car’s survivor vibe. It’s not a quick 2002 like my 2002tii, but it’s an incredibly solid car, virtually free of the usual thunks, klunks, and rattles that haunt 50-year-old vehicles.
Even though it appeared that I would be keeping the car, the 50,000-mile rollover strongly affected how I used it. I know, it was stupid; it’s not like it was some ultra-low-mileage vehicle. It was already a survivor car, not some Cosmoline-coated hangar queen, but I still felt that the mileage was something to be hoarded like Bitcoin or virginity or something equally silly. But between one road trip to Vermont a few years ago and the required back-and-forth to the Monson warehouse on the MA/CT border where I store cars, the mileage had crept to 49,900. I had this epiphany: Do you want it to roll over on the way out to Monson, or do you want it to happen when you’re doing something big and fun?
So big and fun it was. Hampton was going to The Vintage. I took it for a shakedown drive, found a sticky front brake caliper, replaced it, drove it again, and by the time I got back, I was within 28 miles of the big rollover.
Then something unexpected happened. Two days before departure, one of my two road-trip companions called me saying that his BMW 2002 had problems and couldn’t make the trip. I thought about how I have these cars in the Monson warehouse gathering dust, and offered him my ’73 BMW Bavaria. After all, the Bavaria ran fine when I used it a few years ago for a mini-road trip to upstate New York to be used in a movie, and in my recent piece about how all my cars seemed to be rising in revolt, the only issues with the Bavaria were a dead battery from sitting and low-rpm buffeting from imperfectly synchronized Webers.
However, something occurred to me. I’m a big proponent of replacing convention mechanical ignition (points and condenser) with an electronic triggering unit such as a Pertronix (you can read about the debate here). The main reasons are A: points can wear down and close up, causing the car to die, and B: the quality of new points and condensers is absolute garbage these days. And yet I was about to head off on a 2,000-mile round trip in my only two vintage cars still running points. Why? Well, when I was trying to sell Hampton, I wanted to keep it original, and now there wasn’t time to order a Pertronix. With the Bavaria, after its first trip to The Vintage in 2014, I tried installing Pertronix, but for reasons unknown, the car didn’t want to rev over 4000 rpm with it installed, and I never figured out why (I’ve never had this happen on any other car), so I reversed back to points. So both of these cars were not only running points, but were still running the points that were in them when I bought them. (Spoiler alert: Point gap would figure prominently in repairs on this road trip, though not in the way I expected.)
So early on a Wednesday morning, my two companions met me at the Monson warehouse. We put a charged battery in the Bavaria and checked the fluids, then I checked the point gap in both cars with a dwell meter and adjusted it. Then we headed south for Asheville.
Oh, Hampton’s big mileage rollover? It happened 30 minutes into the trip. Over and done. I did my best impression of Paul McCartney singing “Let Me Roll It.” She’s a road trip car now.
We made it to the night’s destination Winchester, Virginia, a little over halfway, without incident. Hampton seemed genuinely happy to be free of its cloistered stored-in-a-barn-in-the-Hamptons-for-10-years-then-treated-like-a-wallflower existence.
When we were about to go to dinner, I got a phone call from a friend—professional vintage BMW mechanic Paul Wegweiser. He said that his friend and customer Mike was about 30 minutes south of me with a dead 2002, and asked if I could help. I called Mike and learned that he and the car were safe in a gas station parking lot with several hotels within walking distance. I said that it made the most sense for me to look at the car in the morning (daylight, it’s on my way to Asheville, auto parts stores are open, etc).
So the following morning I found Mike and his 2002. I’ve written over and over about the common things to strand a vintage car on a road trip (ignition, fuel delivery, charging, cooling, belts, and to a lesser extent clutch hydraulics). A car that goes from driving to dead is highly likely to be a victim of one of the first two. You can give a blast of starting fluid down the carb throat to test which it is (if doesn’t start, it’s ignition, but if it starts and runs for a few seconds, it’s fuel delivery), but for some reason I went right for the points—I yanked off the distributor cap and watched them while Mike cranked the engine. They clearly weren’t opening.
Setting the point gap is usually easy, as points usually have a notch that sits between two little bumps on the distributor plate that allows you to put a screwdriver in the notch and lever it against one of the bumps to increase or decrease the gap. However, the nylon block on these points was so badly worn that the slot wasn’t between the two bumps, and they didn’t really fit right on the plate. Plus, these were the unusual left-opening points used on 2002s with vacuum-retard distributors. I didn’t have a spare set of these with me, and the odds of any AutoZone having them was zero. It took quite a bit of fettling to get the point gap dialed in. When it was, Mike tried starting the car. The carb let out such a loud belch-and-backfire that it startled us all. I theorized that Mike had probably flooded it trying to get it to start with closed points. Eventually it started and idled, and a test drive verified that the car appeared happy. Mike joined our caravan, and we made it down to Asheville without further ignition-related issues.
It was a wonderful event. The organizers of The Vintage refer to it as “a gathering, not a car show.” It’s not a concours. There are no trophies. No one “wins” anything. While there certainly are some lovely restored high-dollar vintage BMWs there, it’s far more about shared passion and enthusiasm irrespective of budget. It’s the kind of event where, on the drive down or in the parking lot, if you need a part or expertise because your car is broken, there are hundreds of people who have your back, and that is a beautiful thing. My having helped Mike was part of the spirit that naturally flows out of the event.
There’s also a long history of my friend Paul Wegweiser pranking me at The Vintage. One year, he bombed my Bavaria with yellow chicken feathers that I’m still finding inside the car. Another year, he actually zip-tied burned-out wires under the dash of my 2002 and a burned-out fan motor under the front seat so I’d smell it on the drive home and wonder where the electrical fire is. He has threatened to put zip-ties on my driveshaft and half-axles so he can read about me going crazy trying to find the source of the noise. However, this year, he said that, since Hampton is such a lovely survivor example, he wasn’t going to screw with it. Like an idiot, I believed him.
The drive home hit a bump on our first stop in southern Virginia. Mike’s car had the good fortune of dying literally as we were heading into a gas station parking lot. Again, it was due to the points having closed up, but this time things were worse—the inside of the distributor cap was coated with soot, the points were noticeably more pitted than before, and I found that the thin braided wire grounding the distributor plate to its body had detached from its connector. And, to add insult to injury, we appeared to be parked near a leaking sewage line or septic tank.
My theory was that the detached ground strap was causing a much stronger spark across the points, which in turn caused both the pitting as well as the soot on the inside of the cap. I got everything buttoned back up, and we continued heading north. I rechecked the distributor on Mike’s car whenever we stopped, and it appeared to be soot-free with the point gap holding stable. One of my travel companions noted that another service area was also, uh, fragrant, but we were parked next to a drainage culvert at the time.
We arrived that night in Chambersburg, Pennsylvania. While we were unloading our bags from our cars, someone noted that the smell of Virginia rest stops appeared to have followed us. While we were waiting in line to check into the hotel, the red light went on in my head: It was my “friend” Paul. After all, someone who zip-tied burned wiring into my car certainly wasn’t above putting something foul-smelling into my BMW. After I checked in, I went back outside and did the nose test under the hood, along the rocker panels, and at the tailpipe, but nothing jumped out at me. I thought that maybe, whatever he’d done, it was heat-activated. He’s a clever guy.
Before we headed off in the morning, I re-checked Mike’s car. I pulled off the distributor cap and was relieved to see both the absence of soot and that my repair of the little ground strap was holding.
Then I borrowed his key and went to start the car so I could check the dwell. It clicked but didn’t start. I pulled out my voltmeter and measured the battery voltage. It read 13.1 volts. Standard resting voltage of a fully-charged battery is 12.6 volts, so it had plenty of voltage.
To make testing easier and eliminate the car’s ignition switch as the source of the problem, I connected a jumper wire to the starter solenoid. I touched the other end of the jumper to battery positive. Again, click but no start.
The no-start decision tree is pretty easy to follow and usually quite definitive. This was beginning to look like a bad starter motor. Pulling the starter isn’t a 10-minute job like the alternator, and we didn’t have a spare one with us anyway, so I wanted to be sure. I was about to swing my car in front of Mike’s to jump it when one of my other companions said he had a new fully-charged lithium jump pack. We hooked it up, it buzzed, and still… click, but no start. Just in case there was a bad connection in Mike’s battery cables, I used my jumper cables to connect the battery directly to the starter. It made no difference. And Mike’s car is an automatic, so there was no way to push-start it.
I lit the Hack beacon and posted a the “2002 down, 2002 needs starter motor” message on the Facebook page for The Vintage, then began removing the starter. With it out, I did the on-the-asphalt test of connecting it directly to the battery. It did spin, but the spin-up time seemed unusually long. Two people quickly answered the post, one of whom had two used 2002 starters at his repair shop just 20 minutes north. He said that we’d actually met once in the parking lot of a Sheetz convenience store nearby. When I got home after the trip, I looked through my old trip photos to The Vintage and found pics of the meeting. Incredibly, it was 10 years almost to the day, and I was driving the same Bavaria.
I tested both used starters by jumping them with Mike’s car’s battery. They both seemed to spin up a bit slowly, but one was obviously faster than the other. Installation, however, was a bear. The solenoid on the replacement starter was fatter than on the original one, and it couldn’t get past the bracket for the kick-down cable for the automatic transmission. I had to loosen the bracket to move it out of the way. It was the kind of bent-over pulling-up-wrenches work that angers up my aging back, but I seem congenitally unable to say “Good luck with AAA” when there’s a problem I can diagnose and fix.
Finally, with one of the starter’s bolts holding it snug enough to the bell housing to verify the repair, I reconnected the battery cables and again touched the jumper wire to battery positive.
Click, but no crank.
No. NO. Not possible.
My first thought was that the engine was seized or otherwise prevented from turning. I chocked a rear wheel with one of the other starter motors, had Mike put it in neutral, and manually rotated the engine (it’s easy to do this on a BMW 2002 by just grabbing the cooling fan and leaning on the belt with the heel of your hand). It rotated easily.
Stumped, I jumped in my car and swung it nose-to-nose with Mike’s to jump it. Why? Don’t know. Just to try something, I guess.
It spun instantly.
Wait, what?
As I put the car back together, I began to accept the idea that I’d gotten the diagnosis wrong. It probably never needed a starter motor. If it started with a jump, the problem was likely the battery. Just because the battery had more than the necessary 12.6 volts, that doesn’t mean that it was able to deliver the cranking amperage to spin the engine. I hadn’t suspected the battery since it looked new (Mike said he’d installed it when he bought the car last year). But it was a mystery why it didn’t start with my friend’s jump pack.
With the starter fully secured and the ignition switch reconnected, the started instantly with a jump and a twist of the key. I re-checked the point gap using the dwell meter, and it was still fine. I verified with my voltmeter that, with the engine idling, there was about 13.5 volts at the battery, indicating that the alternator was charging it. Mike and I said our goodbyes as he was peeling off to drive home to Pittsburgh, about 250 miles. I advised that, as long as he didn’t shut it off the car, he’d likely be fine.
Does anyone get it? Anyone see what I missed? I’ll give you a hint: It’s as plain as the nose on your face.
A few hours later, this text appeared on my phone: “Update! The good news: I am safe at a rest stop off the turnpike. Bad news: I am kaput! Car puttered out and battery is fried. Smoking and a little stuff coming out. I am 96 miles from home, which puts me within the free 100-mile tow! P.S. I think that [expletive deleted] smell was ME!”
Oh. My. God.
The smell! I can’t believe I missed this.
An old-school voltage regulator is designed to to rapidly open and close (not unlike ignition points), bringing the alternator in and out of the charging circuit so that the average voltage to the battery with the engine running is about 13.5 to 14.2 volts. When a regulator fails, it can fail in two ways. They usual “fail open,” which means they never bring the alternator into the charging circuit, so the battery runs down and eventually the car dies (or won’t start). But if they “fail closed,” they cause the alternator to always feed the so-called full-field voltage (about 17 volts) to the battery. This over-charging boils the sulfuric acid in the battery and produces gaseous sulphur which smells like rotten eggs. THAT’s what we all were smelling. It wasn’t sewage. It was the battery being fried.
If someone had said “I smell sulphur,” or “I smell rotten eggs,” my voltage-regulator-stuck-closed neuron would’ve fired, but I missed it. This is why the car’s resting battery voltage read 13.1 volts instead of 12.6 (I can’t believe I missed this one too). And, most important, this is why the battery wouldn’t crank the starter in the car—it was ruined. It’s also why, when removed, the starter was slow to spin up. Had I dropped my own battery in Mike’s car, or used my battery to bench-test his starter, it would’ve spun fine. It was also likely a contributor to why the points were pitting and the distributor cap was coated with soot.
I think that part of the reason I got it wrong was that it was just a few months ago that I wrote about buying a new battery for Hampton when the problem turned out to be a bad starter motor, but I felt like an absolute idiot. The entire episode could’ve been avoided had I simply jump-started the car like anyone who doesn’t pretend to be a know-it-all would’ve done, and if, once it was running, I checked the battery with a voltmeter both while the engine was idling and while it was revved up. I would’ve seen the over-voltage. I had a spare regulator in my trunk. That and a trip to an auto parts store for a battery… it would’ve been so easy.
I still, though, didn’t understand why the car didn’t start off my friend’s lithium jump pack.
A day after we got home, my friend messaged me:
“So I figured out why the starter didn’t crank with the jump pack. It’s a ‘smart’ jump pack that sensed that the battery was at 13.1 volts. That’s the buzzing we heard when you hooked it up. Per the instructions: ‘HOMPOW [brand] car jump starter with intelligent clamps provides protection against over-charging, over-discharging, surge voltage, overload, over-voltage, short-circuit, reverse polarity, and high-temperature protection, making your devices jump faster in a safe way.’”
Oh, my two cars, with their decades-old points? Flawless. Absolutely flawless.
When you blow a diagnosis, all you can do is learn the lesson, and hope that the consequence of being wrong isn’t too painful in time, effort, money, and the degree to which you’ve caused yourself or someone else a pain in the butt. At least this one made for a good story, and two good arrows in the diagnostic quiver.
***
Rob’s latest book, The Best Of The Hack Mechanic™: 35 years of hacks, kluges, and assorted automotive mayhem is available on Amazon here. His other seven books are available here on Amazon, or you can order personally-inscribed copies from Rob’s website, www.robsiegel.com.
Check out the Hagerty Media homepage so you don’t miss a single story, or better yet, bookmark it. To get our best stories delivered right to your inbox, subscribe to our newsletters.
’64 Buick would not fire up when not run over night, clean points, change condenser, check ign switch, replace coil, clean and check all associated wiring connections, checked plug wires all measures 15K ohms, checked distributer cap and rotor, plenty of gas to carb, still crank but no start. It turned out to be leaky HV from old sparkplug wires. Probably would have shown up in the dark.
Might get your ol’ factory sensor checked too Rob, I would never confuse the smell of a boiling battery with sewage….A handy ignition test I came up with decades ago: pull the longest plug wire off its plug, stick a spare plug into it and lay it on the windshield against a (grounded metal) wiper blade. You can visually check for spark from the driver’s seat.
I have jump-starters that have an override switch that will still jump the car even if the sensor says “no.” I needed that on my 911 to use the jump-starter as a battery for the hood latch. The battery was low, but you can only open the hood with power. If that doesn’t work, there’s a procedure to jump there hood latch in the fuse box (+ terminal supplied and labeled) so that you can pop the hood and get to the battery. But the safety sensors wouldn’t fire up. You might look into one with an override.