Insights on being well-grounded
![Hack-Mechanic-Broken-Cable-Lead](https://hagerty-media-prod.imgix.net/2023/06/Hack-Mechanic-Broken-Cable-Lead.jpg?auto=format%2Ccompress&fit=crop&h=481&ixlib=php-3.3.0&w=640)
Whenever anyone mentions having an electrical problem with a car, someone always gives the well-meaning advice to “check the ground.” My experience is that this isn’t the slam-dunk many think it is—I’ve found that most electrical issues are instead tied to bad components or the connections between them.
Of course, sometimes it is the ground. The most dramatic example is with a poor engine ground connection when trying to start the car.
Witness the situation that occurs in the vintage BMW world where I spend most of my time. Someone will pose a question to bmw2002faq.com—the hive mind of 2002 information—asking, “My 2002 wouldn’t start. I turned the key and it wouldn’t crank at all. Then I hit the accelerator pedal, tried again, and it started right up. But afterward, there was a bit of a burning smell from under the hood. What’s going on?” Those of us who have seen this before can answer the question immediately because the symptom is so specific.
On every car, the battery’s negative terminal is attached to two places. The first is the car’s metal body. This provides the common ground path for every single electrical device.
Well, except for one—the starter motor (and any other electrical devices that are integral with the engine). Because engines have rubber mounts to prevent their vibration from telegraphing through the body of the car, they are usually not grounded without a second ground connection from the negative battery terminal directly to the engine. Whether there’s a single Y-shaped ground cable connected to the negative battery terminal where one strap goes to the body and the other to the engine (as is the case on a BMW 2002), or a ground strap from the negative terminal to the body and then a separate one connecting the body to the engine, varies according to make and model.
![Mini battery](https://hagerty-media-prod.imgix.net/2023/06/Rob_Siegel_DSC_0563_annotated.jpg?auto=format%2Ccompress&fit=crop&h=426&ixlib=php-3.3.0&w=640)
Both of these ground connections must be present for the car to start and run properly. With age, it’s not uncommon for one or both of them to fail due to the braided strap corroding and breaking. If the engine strap is intact but the strap to the body has failed, the engine should still crank, but other electrical components may work intermittently, if at all. Conversely, if the chassis ground strap is fine but the strap to the engine has failed, the lights, fans, etc., should all work fine, but with no direct grounding of the engine, the starter probably won’t crank.
I use words like “may” and “probably” and phrases like “if at all” because if one of the ground straps fail, there can be “unintended ground paths” that may allow current to flow though the other one, at least until enough current flows through something that it’s not supposed to that something burns up.
And, with that background, I can explain the strange condition where a BMW 2002 won’t crank until you hit the accelerator pedal, then starts, followed by a burning smell afterward. As shown above, the negative battery cable on a 2002 is a braided Y, where one braid goes to the chassis, the other to the block. When the braid to the block corrodes all the way through, the engine has no direct ground path. After all, the engine is on rubber mounts, as is the transmission, the shift platform, and the exhaust.
But there is an “unintended ground path” and it’s quite a surprising one. It’s the accelerator linkage. The linkage is a series of metal ball-in-socket rods. There’s a lower rod that connects the accelerator pedal to the throttle linkage rod. One end of the throttle linkage rod is connected to the carburetor or injection pump (which, in turn, is connected to the block via the intake manifold and head). The other end goes into a plastic bushing in the firewall that supports it and aids in rotation.
![Mini coil](https://hagerty-media-prod.imgix.net/2023/06/Rob_Siegel_DSC_0585_annotated.jpg?auto=format%2Ccompress&fit=crop&h=426&ixlib=php-3.3.0&w=640)
Wait. Plastic doesn’t conduct electricity, right?
No, it doesn’t.
So how is this a ground path, and how does hitting the accelerator pedal make the starter crank?
This is the fun part.
There’s a throttle linkage return spring about the diameter of a fountain pen connecting the linkage to the firewall. Unlike the rotating end of the throttle linkage rod itself, the spring isn’t sitting in a bushing. The top is just hooked over the cowl and sits in a small hole in the metal. Similarly, the bottom spring hook is in a hole in the linkage. Over time, the paint in the upper hole wears away, allowing metal-to-metal contact. Poor metal-to-metal contact with an inner coating of rust, to be sure, but contact just the same.
![Mini coil](https://hagerty-media-prod.imgix.net/2023/06/Rob_Siegel_IMG_3114_enhanced_annotated.jpg?auto=format%2Ccompress&fit=crop&h=480&ixlib=php-3.3.0&w=640)
So, hit the gas pedal, it stretches the spring, which pulls the hooked ends tighter against their metal holders, maybe jostles them around in the holes. Turn the key, and suddenly hundreds of amps of current find a ground path. They flow through this little piece of metal like an entire city’s water supply being shoved through a soda straw. With all that current, the spring can literally glow like the similarly-shaped elements in a toaster. So, yeah, there can be a smell.
Credit for discovering this rather remarkable failure mode and disseminating the knowledge goes to my BMW 2002 colleague Mike Self, but being an old guy who has seen it happen, I get to act like I’m a wizard when someone describes the symptom and I blithely nail the solution as if it’s the product of deductive reasoning instead of my friend Mike having told me.
I just encountered a pretty cool variant of this issue on a modern car. My friend and neighbor Dave has a 2014 MINI that wouldn’t crank. I walked two doors down to check it out. The battery read 12.6 volts, so it appeared to be fully charged. I connected my battery jump pack, and tried four times. On three of them, nothing happened, but on one, there was the CLICK of the starter solenoid. The rest of the car’s electronics seemed to work, just not the starter.
![Mini Cooper front](https://hagerty-media-prod.imgix.net/2023/06/Rob_Siegel_IMG_5424.jpg?auto=format%2Ccompress&fit=crop&h=480&ixlib=php-3.3.0&w=640)
As I was musing whether the problem could be caused by the interlock that prevents the starter from engaging unless the clutch pedal is depressed, Dave said, “When the car was in for service last year, the mechanic said that the ground strap was corroded. Do you think that could have something to do with it?” He then showed me a video that the mechanic shot, clearly showing a ground strap green with corrosion, and a weak spot forming at the strap’s connection to the chassis.
![Mini wiring cable fray](https://hagerty-media-prod.imgix.net/2023/06/Rob_Siegel_mini-cable.jpg?auto=format%2Ccompress&fit=crop&ixlib=php-3.3.0&max-h=383&max-w=640)
It made perfect sense to me that this was the problem—the strap had corroded through, or nearly so, and the engine no longer had a valid ground connection.
To be clear, Dave wasn’t asking me to repair the car. He was just wondering what I thought was going on, and how serious the problem was. And besides, in order to replace the ground strap, I’d need to jack the car up, set it on stands, and pull off the engine’s under-cover—and his driveway, where it was parked, was too uneven and sloped to safely do any of that.
Dave isn’t a car guy, but he’s smart enough to have asked me the following very interesting question:
“Is there some way to ‘jump’ the ground path?”
A light bulb went off in my head.
“Do you have a set of jumper cables?”
“Sure.”
Like most cars these days, the MINI is a front wheel-drive car with a transverse-mounted engine, so instead of the engine mounts both being just above the oil pan like they are on an old-school rear wheel-drive car, one of them was up high, connecting to the right inner fender wall. I clamped one end of a jumper cable to the big aluminum bracket connecting the engine to the mount and the other end directly to the battery’s negative post.
![Mini clamp](https://hagerty-media-prod.imgix.net/2023/06/Rob_Siegel_IMG_5422.jpg?auto=format%2Ccompress&fit=crop&h=480&ixlib=php-3.3.0&w=640)
I then had Dave start the car. It took a couple of tries with me reseating the jumper cable’s clamp and pulling the ends of the handles apart in order to get their teeth to bite more firmly into the bracket, but the MINI started. It certainly removed any question that the problem was due to the engine not being grounded. And from the video, the root cause of that was almost certainly the corroded ground strap.
Dave asked if I could drive the car to the repair shop. I carefully disconnected the jumper cable providing the ground path. The car continued to run, but we both heard a sharp click, likely an electrical relay—one that also relied on the electrical ground path through the engine—de-energizing. If this was an emergency, I could’ve gone to AutoZone, bought a beefy generic ground strap, and connected it semi-permanently from a good ground point on the chassis to a good ground point on the engine. But it wasn’t, plus Dave had AAA, so after a free tow and a small repair bill at a local shop, the original ground strap was correctly replaced.
This is a good thing to be aware of, and a good work-around to be able to put in both the mental and physical tool kits. Now, you can be the wizard who pulls this out of his bag of tricks at a car event and saves the day. Extra credit if you can point to a spring glowing red like a toaster.
Hey, everyone likes to feel well-grounded.
***
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.
I just experienced the need for proper ground with my recently purchased C5 Corvette. I had a couple codes pop up for lost contact. I cleaned the ground needed and it cured the issue.
The car has 9 grounds and each can affect different parts of the car.
While many grounds may not look as bad as some here they can still use a bit of cleaning with todays cars. Body control modules can really do some crazy things with bad grounds.
Fiberglass bodies = no grounds
I was bitten by this in my 1980 (E12) 528i. Unfortunately, it quietly failed on the way to the airport, and didn’t present itself until I returned a week later and tried to start the car (around midnight, in a January freeze.) I didn’t find the issue until I got the car home the next day. The fix itself was cheap; the assorted transportation fees, somewhat more pricey.
I remember listening to the Click and Clack (the Magliozzi brothers) radio show one time. The caller had replaced their CV axles twice before 100,000 miles and was needing to to have them replaced AGAIN. The mechanic said all the grease was dried up. Ray (or Tom – can’t remember) quickly diagnosed the problem as a bad engine ground, which left the current running through CV axles for ground and cooking the grease!
Oh, that’s good!
Now that’s an interesting one! I’ll file that away for future use. 🙂
Remember… the advice is ‘check the ground’ not ‘it’s a ground’… I see you mentioned the jumper cable trick, that should always be one of the early steps in odd electrical problem diagnosis
One of my more odd ground related problems was in an early 90s Firebird. The car was running very rich once warmed up and was leaving obvious soot marks on the (white) bumper. I got my hands on a scan tool (not as easy as it is today)… the ones with cartridges almost like an NES, and you have to have the right cartridge, and the right diagnostic adapter… the memories… At any rate, scan tool tells me the engine is 60 deg-F when the dash gauge is reading between 180 and 220 (verified independently using a thermocouple probe). Apparently the gauge and computer used two different sensors. The sensor for the engine went to a harness with it’s own independent ground… and that ground was bad. Cleaned it up and reinstalled… happy sailing
Good work there! So the computer thought the engine was still cold, so it was in full rich mode. Excellent diagnostic there.
That’s a good one. I’ll keep it in mind. Thanks!
i recently worked on a 76 MG for a neighbor. When you turned on head lights they did not work and the engine cranked! I disconnected the starter solenoid wire and pulled a headlight out. Touched a wire from headlight ground socket terminal to engine block and headlights came on. found 3 ground wires going to a common terminal that was not grounded to the inner fender. I hooked up the grounds and everything on the car worked properly again!
100%!!! I just discovered this on my ’57 F-100 this past Saturday! The body to engine ground was finger loose and corroded.
It was an AHA! moment after being in the garage dormant for 6 months.
Geez..
Had a ’88 Toyota Pick up back in the day. Wore out the transmission. Got it replaced and hard starting ensued. The replacement bell housing was painted all over including the starter flange. A 6awg cable between the battery ground and a starter bolt cured it. No more having to park on a hill or push starting required.
It’s not always the ground, but it is enough times that you should start there… I learned this most recentyly when I had a no-start issue on the hotrod a couple years ago. Basic car – glass ’33 Ford Hiboy coupe body on an original frame with 350/350 drivetrain. You know – budget fun. Anyway, everything was great till it wasn’t. Parked it in the driveway while running fine one day, then it barely cranked when I went to move it an hour later. Hmmm – battery was a few years old, so I charged the it overnight. Next morning it still wouldn’t start. My cousin suggested it was the ground. So I pulled my truck up beside it, hooked the positive side of the jumper cable to the hotrod’s starter and the negative side to the frame. Bingo – fired right up. Pulled into the garage and eventually found that the block-to-frame ground cable had rusted and broken. In typical overkill reaction, I ran a 1 gauge wire from the trunk-mounted battery to the frame and another from the engine block to the frame. Zero troubles from then to the time I traded it for my latest indulgence. 🙂
I had a big piece of heavy equipment. At only 1200 hours, the oil was full of metal. Turbo diesel. My mechanic figured out that the factory had installed a loose ground and failed to clean the paint off block. When we tore down the engine, every bearing surface looked like it had been ‘tapped’ with a stick welder all over. Had to replace engine with new, clean, multiple grounds.
Wow…
That braided ground cable is a very common problem with just about all newer BMW products in the Northeast. I have replaced several of those braided straps. A voltage drop test will quickly find bad power or grounds. Anyone who works on vehicles should learn how to voltage drop a circuit. Once you understand the theory of voltage drop and you know how to do the test, people will think you are an electrical wizard.
Gilbert, agreed. I discuss voltage drop measurements in detail in my electrical book (https://www.amazon.com/Mechanic-European-Automotive-Electrical-Systems/dp/0837617510), but here I was more going for the optics of photos of broken braided cables and the imagined image of a spring glowing as bright as the elements in a toaster.
Electrical issues are sometimes the most frustrating thing to diagnose but the most satisfying to fix also.
Very true. I kept blowing fuses on the radio of a Plymouth Acclaim I owned back in the 1990s. My father-in-law helped me trace the problem to a door speaker wire. I was happy after that discovery.
I remember second-gen Camaros had an issue where the automatic shifter cable would become stiff to move after a period of time. Usually, the case was a bad chassis-to engine ground, so the current was going though the shifter cable.
Just yesterday my brother and I are re-wiring my basket case Bradley GT. The headlight relay is buzzing and the lights are dim. My brother says we have a ground problem. Hooked up the ground “properly “ and of course we now have lights!
On older British and Swedish rally cars we ran a short braided cable bridging the Left motor mount. Not only an auxiliary ground but also a safety in case of a sheared rubber mount.
I was told long ago that all electrical problems are simple – after you figure them out
I had a friend long ago with a rear-drive Chevy-something that wouldn’t start one night. I checked it out while he held the key in “start”. No crank, but I smelled burning rubber, and saw a red glow from under the rear of the car. Looking underneath, on the rear axle I saw the metal spring surrounding the rubber brake hose was glowing red hot. It actually set the brake hose on fire before I could get the friend to let go of the key. This was very close to the gas tank! The hose fire went out quickly when the key was off, so the worst didn’t happen. Found and fixed the bad engine-to-chassis ground strap under the hood, which was forcing current to flow through the only path it could find- through that spring- to get to the starter, but not enough current to crank the engine. I told the friend not to drive this old car until you change that rear brake hose, explaining that it was a single-master cylinder brake system- any leak anywhere and you lose all braking. He didn’t listen, and about a week later, yeah it popped, and he crashed. Thankfully no injuries, but saving on a $100 repair cost him $1000 or so.
Another great example. Glad your friend wasn’t hurt.