The 30-minute rule: when to stop trying and call for help

It's 1.15am. The labeller on line 4 has stopped on a servo fault. Your night fitter is a good mechanical hand. They reset it, it faults again. They check the wiring at the motor. They reseat the encoder plug. They find the drive manual on a phone and try a parameter.

At two, it runs for ten minutes and stops. At half two, they're sure it's the encoder, but there isn't a spare. At three, the shift manager rings the engineering manager, who rings round for help. The engineer who comes is an hour away. The line restarts at half five.

Nobody chose to lose four hours. They lost them ten minutes at a time, each one spent on "one more thing to try".

Why nobody calls

Calling for help feels like failing. A good fitter wants to fix it. The shift manager doesn't want to wake the boss. Nobody wants to be the one who paid for a callout that turned out to be a loose plug.

And every few minutes, something new looks hopeful. A fault that changes feels like progress. So the call keeps moving back.

The cost moves the other way. The line is stood the whole time. And the clock on getting help doesn't start until someone picks up the phone. If the person you need is an hour away, every minute of trying adds to that hour. It doesn't replace it.

Take the call off the fitter

Here's our view. The decision to call for help should not be left to the person with their head in the panel at 2am. They're the worst-placed person to make it. They're tired, they're close to a fix, and they don't want to give up.

So take the decision away from them. Set a time, agree it in daylight, and make it the rule. We would use 30 minutes. If the fault isn't diagnosed by then, the shift calls for help. Your site might pick 20 or 45, and that's fine. Pick a number and stick to it.

"Diagnosed" means you know what's wrong and what it needs. "Trying things" doesn't count. Nor does "it ran for a bit".

The clock starts when the machine stops, not when the fitter gets to it. And the shift manager, not the fitter, watches it.

Some faults skip the clock

Some breakdowns need help from minute one. Call straight away when:

• it's a PLC, drive, servo or safety circuit fault and nobody on shift is confident with that kit
• it's the same failure as last week and nobody found the root cause
• the fix needs work you can't do safely alone, such as live testing, work at height or anything in a confined space
• your fitter is already on another breakdown
• the machine is the one that stops the whole site

Competence is the part people skip. HSE's code of practice says maintenance "should only be done by people who are competent to do the work" (L22, paragraph 79). The Electricity at Work Regulations go further. Nobody should do work where technical knowledge or experience is needed to prevent danger unless they have it, or the right supervision (regulation 16).

A good mechanical fitter isn't automatically competent on a servo drive. Say that out loud. It protects them, and it gives them permission to make the call. HSE's own maintenance advice is short: "never take on work for which you are not prepared or competent" (HSE, maintenance).

Nights make it harder

On nights, your fitter is often on their own. HSE's lone working guide points out that lone workers "are usually unable to ask more experienced colleagues for help". It tells employers to set "the limits to what can and cannot be done while working alone". It also says lone workers should be able to recognise "when to seek advice from elsewhere" (INDG73).

The 30-minute rule is one of those limits. Write it into your night shift arrangements, so the fitter never has to decide alone whether they're allowed to call. We've written more on lone working for your night engineer and on why breakdowns on nights take longer.

Use the 30 minutes well

The rule only works if the first half hour is spent finding the fault, not guessing at it. That means:

1. Photo of the HMI and the alarm list before anyone touches reset.
2. Get the operator's account while it's fresh.
3. Work through it in order, from the symptom back to the cause. A half-split approach gets you there faster than swapping parts.
4. Write down each thing you've checked, so whoever comes next doesn't repeat it.

That last step matters at minute 30. The engineer you call can start where your fitter stopped, not at the beginning. Our post on writing a fault report an engineer can use has the template.

Do the sums before the night

The 30-minute rule is easier to hold when everyone knows what an hour costs. Work out your cost per hour of downtime on a quiet afternoon. Lost output, idle labour, overtime to catch up, scrap on restart, and any penalties for late delivery. Here's how to work it out.

Put that number next to the cost of a callout. For most critical lines, the answer is obvious. An hour of the right engineer is cheaper than three hours of the wrong one. And a call at minute 30 that turns out to be unneeded costs far less than a call at minute 180 that was.

The rule is only as good as the number you ring at minute 30. If nobody answers, or the engineer who comes doesn't know your kit, the rule just moves the wait to a different place. So sort out who you'll call, for which machines, before you need them.

What to do on Monday

1. Agree the time limit with your shift managers and engineers, and write it into your breakdown plan.
2. Make the shift manager the one who watches the clock and makes the call.
3. List the faults that skip the clock on your site, and the machines that stop everything.
4. Check you have a number for each of those faults that gets the right engineer at 2am.

* Paragraph numbers refer to L22, fourth edition (2014, amended 2018).