Most tower operators replace a failing UPS the way they’d replace a failing car battery: same size, same specs, same brand if they can still get it, and move on. Evoltix builds power intelligence systems for the sites where that habit breaks down: where the load has grown, the runtime requirement has changed, or the real ten-year cost of the last replacement was never tracked.
The Zero-Glitch Power Module (ZPM), Evoltix’s battery-first power intelligence system, is not a UPS. It’s what tower operators install when a UPS is no longer enough for what the site actually needs. That distinction matters here, because a straight swap assumes the answer is another UPS. It might not be.
Before signing off on a replacement, and before assuming the replacement has to be another UPS, these are the questions worth answering first.
What’s Actually Forcing This Decision?
A UPS gets replaced for one of three reasons: it failed, it’s aging out of its service life, or a budget cycle finally opened up. Only the first is a true emergency. The other two are decision points, not foregone conclusions, and the decision isn’t automatically “buy the same thing again.”
If the trigger is age, ask what changed at the site since the original UPS went in. Sites that added remote radios, expanded to more channels, or picked up higher-density equipment often carry more critical load today than the original unit was ever sized for. A like-for-like replacement inherits that undersizing instead of fixing it.
If the trigger is a failure, ask what actually failed. A dead battery inside an otherwise sound UPS is a different problem than a UPS that has failed twice in five years. The second pattern usually means the architecture is the issue, not the battery.
What Does the Load Profile Actually Require?
Most legacy UPS battery backup systems in tower environments were sized to bridge five to fifteen minutes: enough time for a generator to catch, or a brief grid blip to clear. Tower sites rarely fail for five minutes. Storms, wildfire events, and grid disruptions that actually matter operationally run for hours, not minutes.
Ask directly: how many hours of runtime does this specific site need at its actual current load, not its original nameplate load? At 2 kW of critical load, for example, a ZPM 1 unit provides close to 3.84 hours of runtime; a legacy double-conversion UPS at the same load is typically measured in minutes. If the honest answer is “we need hours, not minutes,” a UPS replacement is solving the wrong problem.
The question that matters isn’t which UPS to buy next. It’s whether a UPS is still the right category of system for what this site needs to survive. The full side-by-side on what a rack-mount UPS can and can’t do at a tower site is worth a look before that gets decided in the field instead of on paper.
What’s the Real Cost of the Battery Chemistry You Already Have?
Most tower UPS systems run on VRLA lead-acid batteries with a three-to-five-year replacement cycle, and that cycle cost rarely gets tracked as its own line item. It gets buried in general maintenance spend. Ask what a single replacement cycle actually costs once the battery, the truck roll, the labor, and disposal are added up. On a remote site, that number commonly lands between $2,100 and $5,600, and it repeats every three to five years for the life of the site.
Lithium iron phosphate batteries change that math: up to fifteen years between replacements instead of three to five, with no on-site ventilation required for off-gassing. That’s not a reason to buy any particular product. It’s a reason to ask your current vendor, and any replacement vendor, what the chemistry actually costs over a ten-year horizon, not just at the register. For the ZPM specifically, operators typically see a payback estimate of roughly three years, driven mostly by eliminated battery replacement cycles and fewer reactive truck rolls.
If you’re trying to figure out whether the battery you already have is close to that point, the visible warning signs, swelling, extended charge times, runtime noticeably shorter than rated, are worth checking before you assume a full replacement is even necessary.
Can You See the System Before It Fails, or Only After?
A legacy UPS mostly tells you it has a problem once the problem has already started. Ask what visibility the replacement gives into battery condition, grid status, and power conversion events before something trips an alarm on-site.
IntelliCore, the monitoring platform behind the ZPM, surfaces trending data and automated alerting, backed by up to 26 mappable dry contacts for SCADA and alarm integration, so a NOC (Network Operations Center) sees a battery trending toward end-of-life or a grid event in progress, not just a site that’s already dark. A ZPM can also dead-start from either AC or DC input, so a site that loses grid and battery at the same time still has a path back online.
None of this means the answer is always the ZPM, or always a battery-first system. It means the question is bigger than “same UPS, same brand.” A rural Alaskan WISP running remote tower sites in some of the most demanding conditions in North America cut generator runtime by 91% after making that bigger question its starting point instead of an afterthought.
Talk to an Evoltix application specialist about your site’s actual load profile, or run the numbers yourself first in the TCO Studio calculator to see where a battery-first power intelligence system pays for itself.
Related Pages
ZPM vs. UPS Battery Backup vs. DC Power System
UPS Battery End-of-Life: Warning Signs and What to Do Next
Why Tower Operators Are Replacing Rack-Mount UPS Systems with the ZPM