Telecom towers
Less diesel. Fewer truck rolls. Sites that stay up.
A tower site is a small power plant with no one on it. The grid, the generator, the rectifier, the batteries and the air conditioner all have to work for the radios to stay on. We keep a physics twin of each one, read the controllers you already have, and tell your field team which site needs a visit and why, before it goes dark.
Telecom site power chain
The grid or the generator feeds the site through the transfer switch. The rectifier makes −48 V DC for the radios and keeps the battery charged. The air conditioner keeps the battery cool.
- Air
- AC power
- −48 V DC
- Diesel
- Has its own twin page
What costs you money
Where the money goes today.
Diesel you can’t account for.
Generators run at low load, burn more per kWh, and tanks drop faster than the load explains.
Truck rolls to healthy sites.
A visit to check, a visit to refuel, a visit because an alarm cleared itself.
Batteries that die early.
A hot shelter cuts battery life in half, and you find out during the next outage.
Alarm storms.
One grid drop at a site raises dozens of alarms across the power and radio systems. The real cause is buried.
The machines we know
Six machines on every site, one chain.
Each one has its own twin with its nameplate, its equations and the ways it fails. Together they show the whole site.
- Lead assetDiesel generator setA generator that runs light burns more fuel for every kWh, and wears faster too.20 kVA / 16 kW prime, 400 V 3-phase, 50 Hz, 1,500 rpmOpen the diesel generator set
- AssetRectifier systemEvery module you leave running at light load is heat you pay for twice.−48 V DC plant, 6 × 3 kW modules (N+1), float 54.0 V, boost 56.4 VOpen the rectifier system
- AssetBattery bankA battery bank fails during the outage it was bought for, unless you watch its temperature and its float.48 V, 24 × 2 V VRLA cells, 600 Ah at the 10-hour rate, 10-year design life at 25 °COpen the battery bank
- AssetShelter coolingThe air conditioner protects the batteries. When it fails, the clock starts.2 × 3.5 kW split units (lead and lag), DC free-cooling fan, 2.5 kW equipment heat, 12 m² shelterOpen the shelter cooling
- AssetFuel tankFuel that leaves the tank without the engine burning it is the most expensive litre on the site.500 L horizontal cylinder, resistive level sender, refuel by truckOpen the fuel tank
- AssetTransfer switch and mainsCount the outages, time the transfer, and know when the grid is the problem.63 A 4-pole automatic transfer switch with mains-failure controller, 400 V, 50 HzOpen the transfer switch and mains
Also watched on site
- Solar panels and hybrid controllers
- Aviation obstruction lights
- Earthing and surge protection (inspection records)
- Tower structure (inspection records through Field)
- Access control and door contacts
- Smoke and water sensors
How a job goes
One cluster, one generator setting, one ledger entry.
- 01 Watch
TwinEdge OS reads the generator controller over Modbus and the rectifier over SNMP at the site, and keeps history through backhaul drops.
TwinEdge OS - 02 Find
In the Historian, the twin sees the generator running at 20% of prime on most outages, burning about 0.6 L per kWh, and the battery reaching only part charge before the generator stops.
Historian - 03 Fix
Agentic Analytics drafts a change to the hybrid charge settings and a work order for the field team. A person approves it.
Agentic Analytics - 04 Prove
Run hours and litres per outage hour go into the ledger against the baseline from the first 30 days. The field team closes the work order in Field.
Field
What we connect to
We read what you already have.
At the site
- Generator controllers (Modbus RTU and TCP)
- Rectifier and battery controllers (SNMP poll and traps, Modbus)
- Fuel senders and energy meters
- Door and smoke contacts
- LoRaWAN sensors through your network server (MQTT)
- Store-and-forward at the edge when backhaul drops
Your systems
- Network EMS and NMS: Huawei iManager U2020, Nokia NetAct, Ericsson OSS-RC and ENM, ZTE NetNumen, Fujitsu PNMSj, Samsung SKA, NEC iPASOLINK
- Your CMMS, or AssetOps EAM
- Fuel-delivery records
What we put a number on
Worked out from the equations, before we start.
Fuel per kWh, site load against 75%
0.60 against 0.34 L/kWh
The reference site’s generator averages 20% of prime. At 75% the same kWh takes 44% less fuel.
Modeled
Battery-hybrid cycling
35% less fuel
67% fewer run hours at a 4 kW site, charging a lithium string at 80% load and then stopping.
Modeled
Battery life per °C above 25 °C
0.67 years
6.7% of a 10-year VRLA string for the first degree. 5 years lost at 35 °C.
Modeled
Rectifier efficiency mode
57 W per site
At a 2.4 kW load: 170 W of losses with all modules on, 113 W with 2 on.
Modeled
Modeled on the reference machine. Results vary by site.
What we do not do
Clear lines, so you know what you are buying.
- We do not control the radios or change network configuration.
- We do not start or stop generators remotely unless your own controller and policy allow it, and a person approves.
- We do not replace your NOC tools. We feed them a cause instead of a storm.
How you pay
A 90-day pilot. The fee follows the result.
A 90-day pilot on one cluster of sites. We baseline first, then measure the change. The fee is tied to the measured change, under the standard pilot terms.
What we baseline first
- Generator run hours per site
- Litres per site
- Truck rolls per site
- Battery autonomy test results
What we measure after
- Run hours and litres per outage hour
- Truck rolls avoided
- Fuel that left the tank without being burned, found and closed
- Autonomy at each site against the twin
Questions
Straight answers.
Do you need new sensors at each site?
Usually not. We read the generator, rectifier and battery controllers that are already there. Fuel senders often need a calibration check against a metered fill.
What happens when backhaul drops?
The edge keeps reading and stores history, then sends it when the link returns.
Does it work with our NOC?
Yes. Our tower fault service reads alarms from the EMS and NMS systems listed above, builds a cause graph across the power and radio equipment, holds back the storm and sends one incident with a root-cause note. Each connector’s limits are written down before we start.
Can you tell why fuel went missing?
We can show fuel that left the tank without being burned, how much and when. Whether it was a leak, a bad sender or something else is for your team to confirm.
Do we need a server at every site?
No. A small edge gateway per site, or the controllers’ own network link to a regional gateway.
Will you start our generators?
Not on our own. A start or stop only goes through your controller and your policy, and a person approves it.
What we do for you
We watch it. We catch it. We prove it.
Our reliability engineers run the twins, review every finding, and help your crew fix it. After the repair we compare the meter to the baseline and write the result in the ledger.