Physical layer faults undermine the whole data chain
RS485 is a differential signalling standard used widely for BMS, metering, inverter, BESS and substation communications. Its reliability depends on correct cable type, topology, shielding, termination and separation from electrical interference. Where any of those conditions are not met, the data chain above it, gateways, integrations, dashboards and billing systems, becomes unreliable in ways that are hard to trace from software alone.
Field investigations frequently reveal that cabling decisions made during installation were not reviewed at commissioning and were not visible in the documentation handed over to operations teams. By the time data loss or latency is noticed, the physical cause is buried in trunking, exposed to weather or obscured by years of fault attribution to software, polling settings or network configuration.
Five installation errors found on a single site
The same installation errors appear across different sites and building types. Indoor-rated cable is used for outdoor runs, leaving it exposed to moisture ingress and UV degradation. TPU-jacketed cable without shielding is used in place of screened, shielded twisted pair, removing the protection against electrostatic and electromagnetic interference that RS485 links depend on. Communications cabling is routed parallel to and across power cabling, inducing electrical noise directly into the signal conductors. Terminations are inconsistent: joints poorly made, connectors loose, impedance matching at line ends absent. Wiring is installed in a star configuration from a central point rather than in a daisy-chain, breaking the differential bus structure that RS485 requires.
None of these errors is individually subtle. Each represents a deviation from the installation standard that should have been caught at inspection or commissioning. In combination, they produce a bus that is structurally unable to support reliable communication, regardless of the devices connected to it or the software configured above it.
From unstable comms to incorrect billing
The operational consequences on this site were consistent with the physical faults: intermittent communication loss, periods where data from PV inverters and metering devices was absent or corrupted, and increased latency causing polling timeouts and missed readings. Because the gaps were intermittent rather than total, the symptoms were initially attributed to gateway configuration and network behaviour rather than the physical layer.
The commercial consequences followed directly. PV generation data was inaccurate, making performance monitoring and yield calculations unreliable. Metering data fed into billing was incomplete, producing incorrect invoices that required investigation and correction. The cost of rewiring the site to a compliant installation, screened twisted pair in the correct topology with proper terminations and appropriate cable selection for internal and external environments, substantially exceeded what correct installation would have cost at the outset. The evidence case for that remediation spend required a structured diagnostic review to separate physical, configuration and integration faults and produce documentation that could support supplier and contractor accountability.