7:40 on a Monday. The network closet smells like warm dust. The Tripp-Lite unit we installed two years earlier is dark. Silent. The green LED on the front panel—the one that's been glowing since before I took over purchasing in 2021—is out.
"The UPS won't turn on," the office manager said. "Order a new one."
So I did.
The first time this happened, I didn't question it. A UPS that won't turn on is a dead UPS, full stop. I grabbed a replacement from our vendor, arranged the swap, and put the old unit on the loading dock. Total time spent: maybe three hours of emails and coordination. Not ideal.
Except the "dead" UPS wasn't dead at all. The battery was.
What "Dead" Actually Means
When I finally opened that old unit instead of sending it to the recycler, the battery cartridge was swollen. Not a little—it was visibly bulging, and it took some wiggling to get it out of the tray. The date code said three years and four months. That's past the expected service life of most UPS backup batteries, and nobody had told me. Not the vendor, not the manual nobody reads, not the person who installed it before my time.
Here's the thing I wish I'd known back then: a UPS that won't turn on is usually a battery problem, not a unit problem.
The unit sits plugged into the wall for years, keeping the battery on a steady charge. The network closet gets warm. The battery chemistry degrades. Eventually the internal protection circuit decides it can't safely store energy anymore, and the whole unit refuses to start. It's not a malfunction. It's a 3.5-year-old consumable part saying it's done.
If you're reading "replace the battery" and thinking "that's a technician's job"—fair. I thought the same. I'm not the person who knows how to replace a fuel pump. That's a mechanic's job, and I'm an office administrator. But a UPS battery cartridge? That I can swap in about ten minutes with a screwdriver. Most Tripp-Lite units have user-replaceable cartridges designed exactly for this.
Looking back, I should have put battery replacement dates on the calendar for every UPS in the building when I took over that role. At the time, I didn't even know UPS batteries had a lifespan. Now I do.
The Two Ratings Nobody Checks
Batteries aren't the only reason a Tripp-Lite UPS looks broken. There are two other culprits, and both come down to reading the label.
First: VA is not watts. A 1000VA unit sounds impressive until you find the real wattage rating, which is usually lower—sometimes by a third. If the gear plugged into it pulls more than the wattage rating, the UPS will shut down under load or refuse to turn on. It looks like a failure. It's actually math.
I learned that one the hard way. We have a satellite office with a UPS that kept cutting out under load. One day it appeared completely dead. I drove out on a Saturday—an hour each way—and found the UPS was fine. The switch and router plugged into it were drawing more than the unit could deliver. I'd undersized it when I ordered it. A lesson learned the hard way.
Second: simulated sine wave vs. pure sine wave. This one is stealthier.
Budget UPS units usually output "simulated" or "modified" sine wave when they switch to battery. Older, clunkier power supplies don't care. But modern server power supplies—especially ones with active PFC—can reject that power. They reboot. They shut down. They make the UPS look like the problem when the problem is the wave shape.
The UPS isn't broken. The load is refusing the power it's being fed.
This is why I now buy pure sine wave for anything that matters. Tripp-Lite's SmartOnline line includes 1000 watt pure sine wave inverter models, and the output spec is right there in the data sheet. It costs more on paper. It costs less in reality.
There's a third trap, too: buying a bigger UPS than you need. It sounds like the safe move, but bigger units are less efficient at light load, and the replacement battery costs scale with capacity. The Tripp-Lite UPS selector and calculator walk you through actual load and runtime, which is more than I can say for my instincts.
What a Silent UPS Costs
Let me put some numbers on this, because the price of the replacement unit is the smallest number in the story.
I manage all IT and facilities ordering for a 90-person company—roughly $250,000 a year across fifteen vendors, reporting to both operations and finance. I think about cost differently than I did five years ago.
The replacement unit I ordered that first time cost a few hundred dollars. The battery cartridge that actually fixed the second unit ran maybe a third of that. I'd have to check the invoice for the exact figure. Still: I spent money I didn't need to spend because I didn't check the simplest thing first.
Then there's labor. Every hour I spend coordinating an unnecessary swap, or driving to a satellite office on a Saturday, is an hour I'm not managing the other 60-80 orders I process annually. It adds up.
But the cost that genuinely changed my thinking was the perception cost.
Clients visit our office. Vendors visit. People notice when the phones dip during a storm and the file server restarts itself at 2:00 PM. Nobody walks into a network closet and evaluates the UPS on its specifications. But when a customer's order processing stalls because the power blipped and the backup didn't hold—that's visible. That's a brand hit.
From my perspective, saving money on backup power is a false economy. You buy a UPS for the few minutes a year when the power does something ugly. The quality of those minutes is the entire product. The extra $80 or so per unit for pure sine wave went from "an indulgence" to a line item I stopped questioning after two outages with zero client-facing incidents. Nobody sends a thank-you email when the backup power works. But they notice when it doesn't.
When I weighed the risk—$80 saved per unit on a simulated-sine model versus the possibility of a client-facing system going down—the math didn't hold. The upside was $80. The risk was being the person who caused a preventable outage. Not worth it.
Before You Order a Replacement
Quick caveat before the checklist: I'm not a certified electrician, and I'm not telling you to ignore your vendor's advice. I'm telling you what I wish someone had told me.
If your Tripp-Lite UPS won't turn on, do these three things before you buy anything:
- Replace the battery first. A dead battery cartridge will prevent the unit from starting even when it's plugged into wall power. In my experience, this is the fix in most cases.
- Add up the real wattage of everything connected. If it exceeds the UPS's watt rating, the unit is doing exactly what it should.
- Check the outlet feeding it. This sounds ridiculous, but I once spent an hour troubleshooting a UPS that was plugged into a switch-controlled outlet. Somebody flipped the switch over the weekend. Worse than expected, but a good story now.
If you do need a new unit because the old one is undersized or genuinely past its useful life, here's what I'd spec today, from one buyer to another:
- For a server rack: a Tripp Lite 1U UPS with pure sine wave output. The rack-mount form factor keeps power next to the equipment, and it fits the rack like any other piece of gear.
- For a small server room: a 1000 watt pure sine wave inverter from the SmartOnline series. That's my default for anything that can't be offline.
- For general office equipment: a SmartPro unit is a solid workhorse. Just put the battery replacement date on the calendar.
And use the Tripp-Lite UPS selector or calculator before you order. It sounds like extra work; it took me five minutes, and it showed me where I'd overbought and where I'd underbought. I just wish I'd done it before that first dead Monday.
Set the reminder for year three. Batteries are consumables. So is the assumption that everything works until it visibly doesn't.
Leave a Reply