The batteries were brand new. Two of them, top of the line, installed that morning. The generator still wouldn’t crank.
Not a click, not a groan — well, actually there was a groan. That slow, dying whrrr every generator owner knows in their bones. The site manager looked at me like I’d sold him the batteries personally. “They’re new,” he said. “They can’t be the problem.” He was right. They weren’t the problem. The two meters of copper between the batteries and the starter were.
Here’s what nobody tells you about starting a diesel: the starter motor is the hungriest thing on the whole machine. It can pull several hundred amps for those few seconds of cranking. At that kind of current, every scrap of resistance in the cable turns into lost voltage — and voltage is what the starter actually runs on. Your batteries can be perfect and the starter can still starve, if the road between them is bad. If you’ve already ruled out the usual no-start suspects, the road is where I’d look next.
I learned this the expensive way — or rather, a customer did, and I got the phone call. A 400 kVA set at a logistics yard, and the crew had already replaced both batteries. Twice. Six batteries total, because the first new pair “must have been defective.” When I got there I put the meter on the battery terminals: 25.4 volts, beautiful. Then I had somebody hit start while I watched a second meter clipped onto the starter terminals. Fourteen volts. Eleven volts of drop, gone into heat somewhere along two meters of cable.
The cable told the story once I looked at it. Years earlier, during some long-forgotten repair, somebody had replaced the original heavy positive cable with a thinner automotive one — probably because it was what was on the truck. It had worked, sort of, for years: slower cranking that everyone blamed on aging batteries. Each new battery masked it for a while. Until it didn’t. I laid my hand on that cable after a crank attempt and it was warm. A battery cable should never be warm. Warmth is voltage turning into heat instead of turning the engine.
One proper cable later — right gauge, right length, proper crimped lugs — the engine spun like it was showing off. Six batteries, two service visits, and the fix was a cable.
There are three, and I’ve seen all three enough times to check them in this order.
First: too thin. Copper costs money, so there’s always temptation to go one size down. At starter currents, “one size down” can mean a volt or two of drop you never budgeted for. The engine doesn’t care what you saved. It cares what arrives at the terminals.
Second: too long. Somebody relocates the batteries to “a better spot” — farther from the starter, routed prettier — and doubles the cable length. Double the length, double the resistance, double the drop. Batteries belong as close to the starter as the layout allows. Every extra meter is a tax.
Third: the connections. This one’s the big one, honestly — more common than the other two combined. A lug crimped badly, a terminal bolt left a quarter-turn loose, corrosion blooming white-green under the heat shrink where you can’t see it. I check connections before I check cable size now, because connections fail slowly and silently while the copper itself is usually fine.
Current has to get back, and the return path is usually a single ground strap from the engine block to the frame or straight to battery negative. I spent half a day once on an intermittent no-crank — started fine sometimes, dead others, no pattern at all. Turned out the ground strap was bolted through a painted frame rail. It had worked for years because the bolt teeth bit through the paint just enough. Then vibration and moisture did their slow work, and “just enough” became “sometimes.” Scraped it to bare metal, star washer, proper torque — never came back. These days I check the ground before the positive. The positive cable gets all the attention. The ground does half the work.
On 24-volt sets there’s one more trap: the short series jumper between the two 12-volt batteries. People obsess over the big cables and ignore it, but it carries the same hundreds of amps. A corroded series link drops voltage exactly like a bad main cable, and it’s the last place anyone looks. And — I say this having seen the aftermath — don’t pair an old battery with a new one in series and expect the cables to make up the difference. The weak battery becomes a resistor. The cables get blamed. The cables were innocent.
The crank-voltage test. Five minutes, one meter, no guessing. Put the leads directly on the starter motor terminals — not the battery, the starter — and have somebody crank. Watch what the voltage does under load. On a 12-volt system, if it collapses deep into single digits, something between the battery and the starter is eating your volts. On a 24-volt system, same idea at double the numbers. Then walk the meter backwards: starter terminal, far end of each cable, battery terminal itself. Wherever the voltage jumps back up, the bad stretch is between your last two probe points. Fastest honest diagnostic I know. Physics doesn’t negotiate.
Cruder version for when the meter’s back at the shop: after a crank attempt, lay your hand along the cables, carefully. A cable that’s warm — or a lug that’s hot — is pointing at exactly where the resistance lives. Heat is voltage you paid for and didn’t get to use.
Go one size heavier than you think you need. Keep the run as short as the layout allows. Use proper crimped lugs with sealed heat shrink — not the hammer-crimp-and-hope kind. Route away from exhaust heat and anything that chafes. It costs a little more, once. It pays back every cold morning for a decade.
An old-timer I know insists battery cables are lifetime parts — “copper doesn’t wear out,” he says, and about the copper he’s right. Where we disagree is everything around it: the lugs, the crimps, the corrosion, vibration working the strands at the terminal until half of them let go. I tell customers to put eyes on the cables every couple of years — flex them gently near the lugs, look for green, tug the terminals, check the ground. He says I’m inventing work. Maybe. But I’ve never had a callback on a start failure after that check, and he doesn’t tell me about his callbacks.
The batteries at that logistics yard are still there, by the way — the second new pair. Perfectly good batteries. They just needed a road worth traveling. Give your cables ten minutes on your next maintenance visit: feel for warmth, look for green, meter the crank voltage at the starter. Cheapest diagnostic on the whole machine. It’s the one that would have saved that customer six batteries.
Starter still acting up after the cables check out? Get in touch — sometimes the problem really is the starter, and that’s a different story.
Email：[email protected]
Work：Monday to Friday,9:30-18:30
