What happens when radio goes silent? Build the backup plan first
The studio's internet connection drops during a live interview. The presenter may still be speaking, but listeners are no longer receiving the audio. That is a poor moment for the team to decide for the first time who should act and which alternative to use.
A radio backup plan describes how a failure is detected, who decides to switch, what alternative is available and how normal service is restored. It cannot prevent every interruption, but it can make decisions clearer when time is short.
Is the backup really a separate path?
Trace audio from microphone to listener: source, mixer, encoder, network, streaming server and player. A second computer does not protect against a router failure if both depend on the same router. Two devices on one power circuit can share a single point of failure as well.
The EBU's recommendation on crisis distribution discusses diverse, resilient distribution infrastructure. A small student station may operate at a very different scale from a national broadcaster, but can still identify which backups share dependencies.
Internet, power and source failures are different
A failed microphone may call for a tested spare input or a short recording with appropriate rights. An internet outage may need a different connection path. If mobile data is the fallback, coverage, allowance and battery become relevant.
For power interruptions, assess any uninterruptible power supply against the actual connected load and manufacturer guidance. Do not assume the same advertised runtime applies to every setup; test safely. If an alternate streaming provider is available, also consider how listeners reach it and how the team returns to the main path.
Listen while exploring this story
Tune into campus radio with non-stop independent music and culture.
Disconnected is not the same as silent
An encoder can remain connected while sending silence. Some failover systems respond to disconnection but may not treat silence as the same event. For example, Radio Mast's backup-stream documentation describes using a backup when an encoder disconnects. That does not establish how every provider responds to silence.
Rehearse both cases separately on an authorised test stream and listen from an independent device. If silence detection is unavailable, decide who will notice the problem and how they can intervene manually.
One detail in the same Radio Mast documentation makes “we’re back on air” worth checking carefully. When the primary encoder connection returns, the service sends only **new listeners** to the primary stream; it does not automatically move existing listeners back, to avoid another interruption. The team may hear the restored primary while someone who connected earlier still hears the backup. This behaviour describes Radio Mast’s service and should not be assumed for another provider.
The documentation also describes sending from the same encoder to ingest servers in two regions. Separate ingest points can help with some network problems, but if both depend on one encoder, a failure of that encoder can affect both paths. That is an inference from the documented arrangement, not a finding about RadioTEDU’s setup or a completed test.
Practise the return as well as the switch
Fallback recordings need current information and appropriate usage rights. Agree who can make the switch, how the team communicates and what must be checked when the main service returns. Do not put passwords in a shared plain-text runbook.
A controlled rehearsal at a low-risk time shows what happens from detection until a listener hears the fallback. Record the date and observations, and do not describe a drill as completed before it happens. A useful plan includes the route back to normal broadcasting.
