Behind the Patch Bay: How a Live Feed Is Kept on Air

People often assume the backup is an afterthought, added once the main route works. I do the opposite. The second path is planned first, because it is the one that has to behave when conditions are poor. Where possible it takes a different physical route and, if the venue allows, a different type of link, so that a fault in one place does not touch both.

A backup that has never carried a picture is only a hope. During setup I send the feed down the alternative route and look at it on a proper monitor for a sustained period, not a glance. I check that the picture holds steady, that audio stays in step and that the format arrives as the broadcaster expects. Then I switch back and confirm the main route returns cleanly.

Power gets the same treatment. Key units sit on separate supplies where we can arrange it, and I label which cable feeds which box, so a tripped breaker is a short conversation rather than a search.

Once the feed is running, my job turns into watching. I keep a monitoring wall that shows the signal at several points along its path, not just at the end. If the picture is fine at the source but degraded at the output, the fault lies between the two, and I already know which stage to look at. That narrows a problem from a whole system to a single link.

I pay attention to small things: a slightly late frame counter, a faint change in audio level, a link that reports more errors than it did an hour ago. None of these is an emergency, but they are the early warnings that let me act before a viewer notices anything. Most faults announce themselves quietly if someone is looking.

I also keep a short log through the day. It records what I changed and when, so that anyone taking over, or reviewing the broadcast afterwards, can follow what happened without relying on memory.

When a fault does appear, the calm approach is also the fastest. I decide first whether the picture on air is at risk right now. If it is, I move to the backup route and tell the vision team plainly what I have done, in as few words as possible. Diagnosis can wait until the feed is safe; fixing the main path while it is carrying the broadcast is how a small fault becomes a large one.

With the backup carrying the signal, I work through the route in order, starting at the source and moving toward the output. Because the drawing and the monitoring wall already show where the picture first goes wrong, this is usually a matter of minutes. Often it is a loose connector, a link that needs resetting or a unit that has overheated in a closed rack.

Afterwards I return to the main route only when it has run clean for a good while, and only at a moment the vision team agrees is quiet.

The handover to the broadcaster is the point where our responsibility meets theirs, so I treat it as a deliberate step rather than a hand-wave. Priya Nair, our distribution coordinator, confirms the delivery details with the receiving team beforehand, and I check on my side that what leaves us matches what they expect, in resolution, frame rate and audio layout.

When the broadcast ends, I do not strike the rig until the receiving side has confirmed they have everything they need. Pulling a link too early, even when the programme is over, can cut a recording or a late insert that someone still wants.

Then I write up the day: what was planned, what changed, what I would do differently. Those notes shape the next route drawing, which is how the work slowly gets calmer over time rather than merely repeating itself.