LegitFix reads and decrypts every radio telegram straight from the meter — consumption, flow alarms, and diagnostic bits — and streams it to LegitOS the moment it happens. No batching. No estimates.
Every submetering operation runs on the same promise — real-time visibility. Most hardware quietly breaks that promise the moment it hits a real building.
Battery meters on NB-IoT or LoRaWAN batch readings to save power. By the time an anomaly reaches you, the leak has been running for hours — or the billing cycle has already closed around it.
An angry email or a call to the front desk is usually the first sign something's wrong. That's not a staffing problem — it's just what happens when the system only checks in a few times a day.
Consumption is a fact, not an estimate. If the reading behind an invoice came from a handful of scheduled polls, the number on that invoice was already an approximation before anyone opened a spreadsheet.
Wakes up a handful of times a day, reports whatever accumulated since the last check-in, and goes back to sleep. Between check-ins, you're blind — a quiet apartment and a broken meter look identical.
Streams the instant real consumption starts, and checks in every hour even when nothing happened. Silence with a heartbeat means idle. Silence without one means offline — the two never look the same.
That last number follows directly from the first two. Because every unit streams on the event instead of waiting for a scheduled read, LegitOS can tell a leak from normal use within hours — not when next month's bill arrives.
LegitFix ships in a compact black polycarbonate enclosure, wall-mounted, with a custom front membrane carrying status LEDs, the LegitFix mark, and regulatory certification.
| Enclosure | Black polycarbonate, wall-mount |
|---|---|
| Dimensions | 115 × 90 × 55 mm (body) |
| Mounting | Single wall-mount foot (2 holes), protruding from the base |
| Front panel | Custom polyester membrane, 95 × 70 mm, selective texture, 3M adhesive-backed |
| Indicators | 4 front-panel status LEDs (cloud, signal, RF, power) |
| Certification | ANATEL 04456-23-14750 |
LegitFix isn't the meter manufacturer's own gateway — it's a local, protocol-native concentrator that sits between the meter and LegitOS. It decrypts the full radio telegram on-site — consumption, flow alarms, every diagnostic bit — and forwards only a computed reading upstream. Nothing transparent, nothing discarded.
Compatible meters send a radio telegram at their own native interval regardless of consumption — 8s for the Diehl IZAR RCi and IZAR DP, 16s for the Diehl HYDRUS, up to 32s for some other meter brands. No provisioning, no DevEUI or AppKey, no vendor-specific decoder required.
Every telegram is decrypted and analyzed on the concentrator itself. Without consumption, only an hourly heartbeat goes out. The moment consumption is detected, every reading streams out, telegram by telegram.
Readings reach LegitOS over MQTT roughly 10 seconds after the meter's own telegram. Pricing rules turn it into an automatic invoice — like a gas pump showing the volume and the amount due side by side.
You don't wait for a monthly statement to know your balance — your bank shows it the moment it changes, on your phone. LegitFix does the same for water and energy: consumption shows up in the app as it happens, not as a number someone reads off a meter once a month.
That's only possible because the reading itself is event-driven. A concentrator that waited for a scheduled poll could never show a balance that's actually current — it could only ever show what the last poll caught.
Pump rooms, parking garages, machine rooms, buildings with reinforced concrete slabs and fire doors — the meter itself often has no line of sight to a cell tower or LPWAN gateway there. Get the concentrator wrong for that environment and you inherit hours of delay, dead zones, and DevEUI-style provisioning — right in the buildings where real-time visibility matters most. LegitFix's radio hop only has to cross the client's own walls, not reach outside the building, so it can sit wherever 4G signal is adequate, even if the meters around it are buried deep inside the structure.
| Constraint | NB-IoT / LoRaWAN | Wired M-Bus + Ethernet | LegitFix (local concentrator) |
|---|---|---|---|
| Latency | Meter batches readings to save battery/bandwidth — hours of delay between event and data arrival | Can hit ~10s updates, if nothing in the pipeline bottlenecks | Event-driven streaming, no batching required |
| Provisioning | Requires DevEUI/AppKey and vendor-specific onboarding per device | Requires rewiring the meter module and a gateway recognized by the wired network | Plug-and-play — meter transmits as soon as it's installed |
| Regulatory limits | 915MHz band (Brazil) is duty-cycle and channel-scheduled by Anatel — caps traffic per device, worse as more meters share a gateway | N/A | No duty-cycle constraint for typical submetering traffic |
| Physical penetration | Weak through reinforced concrete and metal structures | Requires new wiring through the same structures | 434 MHz penetrates concrete and metal reliably |
Same decoder, same data structure, tuned to the ISM band each market actually allows.
| Data layer | OMS-compliant (EN 13757-3), BLE |
|---|---|
| Radio | Wireless M-Bus at 433/434 MHz (Brazil) or 868 MHz (Europe) · adapted sub-GHz RF at 902–928 MHz, 915 MHz nominal (North America) |
| Regional variants | Same EN13757-3 frame and application layer across bands — the radio is tuned to whichever ISM band fits the market. In the US, that means the 902–928 MHz unlicensed band under FCC Part 15.247/15.249, suited to higher-duty-cycle reporting. |
| Payload | Encrypted; decrypted and computed locally by LegitFix |
| Transmission logic | Event-driven on consumption + guaranteed hourly heartbeat |
| End-to-end latency | Meter's own interval + ~10s for local processing, MQTT delivery, and platform receipt |
| Capacity | Up to 600 meters per unit |
| Meter-to-concentrator reach | The meter transmits, LegitFix listens: typically ~400m through light obstructions, ~100m through heavy obstructions, up to ~1000m with clear line of sight |
| Uplink | MQTT over 4G LTE |
| Provisioning | None required — no DevEUI/AppKey, no vendor-specific decoder |
| Deployment model | Hardware-as-a-Service (HaaS) |
| Data delivered | Consumption, flow rate, flow alarms, and other diagnostic bits |
LegitFix is the hardware layer behind LegitOS deployments across residential, industrial, and port environments.
302 smart meters across 370,650 acres, covered by 5 LegitFix units.
Read case study →A local LegitFix concentrator covers every consumption point in a facility where cellular and LPWAN signal essentially don't reach.
Talk to our team →Per-vessel water supply metered and billed automatically, with flow rate calculated for the full duration of each event.
"The supply ends, and the invoice is already generated. No delay."
Talk to our team →No. LegitFix decodes and decrypts the raw radio telegram and computes consumption on the device itself. Only the processed reading is forwarded upstream.
Yes. Because LegitFix decrypts the full radio telegram, it also captures flow alarms and the other true/false diagnostic bits the meter reports — not just the consumption value.
LegitFix still sends a guaranteed heartbeat once an hour, so an idle meter is never confused with an offline one.
In practice, around 4 hours from when it starts. Sustained flow that shouldn't be there shows up as an anomaly on LegitOS almost as soon as it begins, instead of waiting for the next scheduled read — which, on most other systems, means the next billing cycle.
No. The meter's native protocol doesn't require a DevEUI, AppKey, or vendor-specific onboarding process. The meter starts transmitting as soon as it's installed, and LegitFix decodes it automatically.
LegitFix streams processed readings to LegitOS over MQTT. LegitOS applies pricing rules against the data in real time and generates invoices automatically — no manual reading, no estimates.
Tell us about your site — pump room, port, parking structure, or residential portfolio — and we'll show you what real-time actually looks like there.
Talk to Us →