How many of the 50 to 200 generator hours a typical tower site burns through each year come down to a gap nobody actually checked? A dead battery gets swapped, a truck rolls, and the audit that would have caught the real problem never happens. By the time an outage forces the question, it is already the audit.
The Evoltix Zero-Glitch Power Module (ZPM) is not a UPS. It is what tower operators install when a UPS is no longer enough. But this checklist is not about any particular product. It is about the system already sitting at the site today, whether that is a ZPM, a legacy UPS, or whatever a previous vendor left behind five years ago. Run it before the next outage decides the answer for you.
Twelve things worth checking, grouped into four areas: what the site actually needs, what condition the batteries are really in, whether anyone would know before a failure happens, and whether the system can scale and hold up to a compliance review.
Start With the Load, Not the Nameplate
- Real load, not rated capacity. Audit the actual amp draw at the site today, not the number printed on the cabinet when it was installed. A system sized for a 2 kW site during initial deployment may now be running 4 kW after a 5G upgrade or added equipment, and nobody updated the backup power math to match.
- How many times has the grid actually blinked. Most sites log outage duration if they log anything at all. Frequency matters just as much. A site with twelve short interruptions a month wears differently than one with a single eight-hour event, and the difference changes what the site actually needs from a backup system. For context: one rural Alaskan WISP running remote tower sites cut generator runtime by up to 91% after deploying a battery-first system. That is the kind of gap a real interruption count can reveal, instead of a hunch.
- What the real runtime requirement is, not the industry rule of thumb. FCC and grant funding guidelines set a compliance floor, not a design target, and plenty of backup power systems are sized to just clear that floor. A site three hours from the nearest generator crew has a different real runtime requirement than one twenty minutes from a fuel depot. Audit against actual response time, not the minimum on paper.
Know What Shape the Batteries Are Actually In
- Battery chemistry and how it fails. Lead-acid batteries fail gradually through sulfation, often invisibly, until an outage exposes the gap. Audit which chemistry is installed, how old it is, and how that age compares to its expected service life, not just whether a status light says the system is fine. LFP batteries rated for up to 15 years change this math considerably, but only if someone tracks the install date against the clock.
- Ventilation and thermal exposure. Lead-acid batteries off-gas and need a vented enclosure. Temperature swings shorten their life fast, and a cabinet sitting in full sun in a gravel lot ages differently than one in a climate-controlled shelter. Audit whether the enclosure matches the chemistry installed and how much heat and cold cycling the site actually sees.
- Dead start capability. If the battery fully drains, can the system restart on its own, or does it need an external boost truck-rolled in? Audit for dead-start capability, AC or DC, before it gets tested by a real outage instead of a planned one.
Check Whether Anyone Would Know Before It Fails
- Real alerting, not a green light. Ask whether the monitoring actually reports battery state of charge and grid status remotely, or whether monitoring means a light on the front of the cabinet that requires a drive-out to check.
- Truck rolls tied to backup power in the last twelve months. Count them. If field techs are rolling out because a battery died silently, an alarm never fired, or nobody knew a site had switched to battery power until it ran out, that is data worth having before the next budget cycle, not after.
- Whether the monitoring system can actually reach anyone. Some agencies run backup power equipment on isolated networks, firewalled from the internet for security reasons that make sense on their own terms. That is a legitimate choice, but it means remote monitoring can really mean a local alarm panel nobody is watching. Audit what the system can actually reach, not what the spec sheet claims it is capable of.
Confirm the System Can Scale and Hold Up to Scrutiny
- Redundancy, or the lack of it. If the one thing keeping a site alive fails, is there a second path, or is a single point of failure the actual design? Audit for redundancy the same way you would audit for load.
- Consistency across the fleet. A backup power setup that looks different at every site, different vendors, different vintages, different failure modes, means every truck roll starts with a discovery phase before it can start with a repair. Audit how many distinct systems your team actually has to know how to service, not how many sites you have.
- Documentation for compliance and funding. If the agency is pursuing funding through BRIC, HMGP, or another federal program for backup power upgrades, confirm the system’s documentation actually supports what the application needs. Compliance minimums are a floor, not the target, and funding applications tend to get won or lost on documentation, not intentions.
None of this requires new equipment to check. It requires actually looking, on a schedule, not after an outage forces the question. When it does come time to replace what the audit finds, a power intelligence system that reports its own condition removes most of this list going forward, which is the category the ZPM was built to fill.
Quick Recap
- Real load, not rated capacity
- Grid interruption frequency
- Real runtime requirement vs. compliance floor
- Battery chemistry and age
- Ventilation and thermal exposure
- Dead start capability
- Real alerting vs. a status light
- Truck rolls tied to backup power
- Whether monitoring can reach anyone
- Redundancy
- Consistency across the fleet
- Documentation for compliance and funding
If this audit turns up more open items than closed ones, a TCO Studio session can map what a battery-first replacement looks like for the specific load at that site. Request a site assessment to start there.