Three Things I Learned the Hard Way About Diesel Generator Fuel Systems (and One I Still Get Wrong)
The Problem You Think You Have
I got a call last March from a facility manager in Houston. Their Cummins 750kW generator had been running fine—on paper. Weekly exercises, no fault codes, clean oil. But when Hurricane season prep started, they discovered the real problem: their fuel system was slowly killing the generator.
The symptom they saw was a 'Low Fuel Pressure' alarm. The solution they thought they needed was a new fuel pump. They were right about the symptom but wrong about the cause—and that mistake is more common than most people realize.
I know because I've made it myself. Several times. With receipts.
The Deeper Cause: It's Not the Pump
Here's the thing about diesel generator fuel systems. They’re incredibly robust until they aren't. And when they fail, the first thing everyone blames is the fuel pump.
But in my experience working with systems ranging from a 400 kW Cummins standby generator to the big 2.5 MW data center units, the root cause is almost never the pump itself. It’s three things:
- Contaminated fuel. Diesel grows algae (diesel bug). It collects water. It gets stale. And it doesn't take much to clog a filter.
- The wrong filter. A surprising number of places specify a 10-micron filter when they should be using a 2 micron diesel fuel filter. The difference between a generator that starts and one that doesn't is often that simple.
- Neglected fuel polishing. Running the engine is not the same as maintaining the fuel. The fuel in a standby generator's tank can sit for months. That's an eternity for biological growth.
The Fuel Pump Impeller Trap
One of the most common components that gets replaced unnecessarily is the fuel pump impeller. I did this myself in 2022. A 500 kVA unit at a distribution center kept losing prime. The symptoms pointed to the pump. I swapped the pump assembly—$1,200 for the part, plus labor and downtime.
The problem? A piece of debris from a deteriorating fuel hose had partially blocked the pump inlet. New pump, old problem. The impeller was fine. The system wasn't.
I didn't fully understand the value of a pre-filter inspection until that $1,200 mistake. Now I check the fuel lines before I check the pump. — Me, kicking myself for not doing it right the first time
The Cost of Ignoring the Supporting Systems
Fuel is one half of the equation. The start system is the other. And if there's one part of the generator setup that people underinvest in, it's the battery and charging system.
Battery Charger vs Jump Starter: A False Choice
A conversation I hear periodically goes something like: “We’ll just keep a jump starter in the shed in case the battery dies. Saves us from buying a battery charger.”
I used to think this was a reasonable take. It's not. It's a gamble with a terrible payoff.
In January 2024, a client had a 400 kW Cummins standby generator fail to start during a scheduled test. The battery voltage was 11.8V. They grabbed their jump starter. The generator started—once. But a jump start doesn't fix the underlying cause: a failing battery or a charging system that isn't keeping up.
The generator ran for 4 hours. Then shut down. The battery had not been recharged by the generator's alternator because the load wasn't high enough. The jump starter was back in the shed, fully charged but useless for the real problem.
A proper battery charger (not just a maintenance charger) does three things:
- Maintains the battery at optimal charge level
- Desulfates the plates to extend battery life
- Provides a diagnostic signal—if the charger is working hard, the battery might be failing
A jump starter does one thing: gives you enough juice for one attempt. And that's not a backup plan. It's a lottery ticket.
Why This Matters More Than You Think
Here's the uncomfortable truth: the cheapest generator—or the cheapest solution to its problems—is almost never the actual cheapest. Ask me how I know.
In 2023, I oversaw a project where the team decided to skip the fuel polishing service to save $400 on an annual maintenance contract. The generator was a Cummins 750kW for a telecom facility. The result? Clogged 2 micron diesel fuel filter during a power outage. The generator ran for 2 minutes, then shut down. The facility lost power for 3 hours.
The cost of that outage was estimated at $17,000 in lost operations and emergency service calls. The fuel polishing contract would have been $400. The fuel system flush to clean up the mess? Another $3,200.
“In my experience managing over 30 generator installations in the past 6 years, the lowest quote has cost us more in roughly 60% of cases—usually because of hidden issues with supporting systems like fuel and charging.”
That's not an opinion. That's a running tally I keep. The outliers were the cases where the cheap quote got it right—usually because they happened to be good at the exact systems that matter, not because cheap is generally safe.
What I Do Now (and What I Wish I'd Done From Day One)
If you're managing a Cummins generator—whether it's a 17kW home standby or a 2500kW data center monster—here's what I've learned to focus on. It's not complicated, but it's consistent:
- Test the fuel, not just the generator. Sample the tank every 6 months. If you see water, sludge, or bacteria, treat it before it reaches the filter.
- Know your filter specs. A 2 micron diesel fuel filter is the standard for modern common-rail injectors. If your system uses 10 micron, you might be asking for trouble with newer fuel.
- Invest in the charging system. A reliable battery charger (like a Cummins or compatible brand) is cheaper than a generator failure. A jump starter is a false safety net.
- Don't replace the pump impeller without looking upstream. Check hoses, check the tank, check the pre-filter. The pump is rarely the problem.
One thing I still get wrong: I have a tendency to assume that if the system ran fine last quarter, it'll be fine this quarter. That's not how diesel works. The fuel degrades. The bacteria grows. The battery sulfates. The filters clog. The only maintenance that's predictable is the kind you schedule.
I still occasionally skip a fuel test because I'm busy and the numbers look good from last time. It's a bad habit I'm working on. (Note to self: schedule the next sample drop-off before I leave the office today.)
The generator isn't the weak link. The systems that support it are. And those systems don't need expensive solutions—they need consistent, informed attention.