Predicting Maintenance Needs

by Rolin Lintag

[May 2026] You normally do not wait for your vehicle to break down to do maintenance repairs, do you? So, what about your program and transmission systems? Predicting potential problems, as Rolin explains, can prevent downtime for your station – and problems for you.

Have you ever wondered how automobile manufacturers can tell you what maintenance to do on your car after so many thousand miles or after 6 months?

Intuition tells us that they know that out of experience, but how do they do that?

Can we apply that for our broadcast facilities and how?

THE BIG QUESTION

One question that Chief Engineers are asked is, “When do you think we need to replace such-and-such equipment (transmitter) so we can start saving money for it?”

Accountants call it depreciation. They ask the question as if I hide a crystal ball of fortune in my office. I just say, “I did not go to Hogwarts for college nor do I use a magic wand in my pocket.” Although some humor can help sometimes, it can only take me so far.

Before accepting the job offer, I told my prospective employer that there are two things I cannot do: (1) I cannot do magic. I use science, reason, logic, and my experience and that of others to solve problems for a living. (2) I cannot read minds.

I need to know if something broke or if a workflow is no longer working. I cannot hit a target that is not on my radar so, “Please tell me if something went wrong and the sooner the better!” This is one of my pet peeves that I openly tell my Engineering staff: “I’m going to find out about it anyway so better you tell me now than I hear it from the GM during our Management meeting tomorrow.”

THE PATH TO KNOWING WHEN

Historical Information is the raw ingredient for predictive tasks.

Information is one commodity for which Chief Engineers need to have access, so they can be able to plan and make the right decisions – or even to package a report to other stakeholders of which regular engineers in the station may not even be aware.

“Information is power,” Sir Francis Bacon famously wrote. I would like to add to this “… actionable information is power, and acting on it is power displayed.”

Rule #4 of The 48 Laws of Power by Robert Greene says, “Always say less than necessary,” but his broader philosophy reminds us that “Actions speak louder than words.”

NOT KNOWING LEADS TO BAD OUTCOMES

We simply need information (historical) to predict outcomes. If we do not learn from history, then it will just repeat itself.

If we do not document a certain failure and rectify what caused the problem, then it will just recur.

One big and very expensive mistake that someone made for not paying attention to history was during Operation Barbarossa.

In 1941, Hitler ignored the cautionary tale of Napoleon’s 1812 invasion of Russia (as reported in The Second World War by Antony Beevor). Some military strategists say that it was the start of the irreversible defeat of the Nazis at the Eastern front. If Hitler had learned the lesson from Napoleon – not to attack Russia during winter – then he could have averted his defeat and spared 3 million men and about 3,300 tanks lost in January 1941.

As it happened, inevitably, the Nazis lost to what the Russians nicknamed “General Winter.”

It may seem quite dramatic to use this as an example, ‘eh? But it did happen in history, so do not kid yourself that such a tragedy will not happen to you and your station if you do not deliberately work on it.

Recording history and acting on it is a very important task of Broadcast Engineering.

KEEPING A TIGHT GRIP ON THINGS IS NECESSARY

While some problems seem to be intermittent, can they actually be predicted? Yes.

Aluminum terminal strips and copper wires are known to not like each other as far as thermal linear expansion is concerned (NEC Section 110.4). What started as a very good mechanical and electrical connection can become loose after a few years due to “cold flow.” That is why they need to be checked for tightness, even if no one has touched them for years.

One of these days, such an intermittent problem will drive you nuts—a problem you could have prevented if you only knew this tiny bit of information.

EVEN TIGHT CONNECTIONS CAN FAIL

The same goes for metals in the Galvanic series (ASTM G82-98 Standard) .

If you join dissimilar metals in a moist transmitter site, one will eventually sacrifice itself to the other via electrolysis. We usually watch out for this at tower guy anchors where dissimilar metals at different levels of Galvanic series may be deployed.

Now, I am not saying that intermittent failures can be altogether predicted and prevented. There are failures that border in the area of “freaks of nature” or some poltergeist haunting your station, but we do not want to leave any stones unturned.

This is why we strategize our method of attack to overcome such failures, or even impending failures.

www.digitalalertsystems.com

GEP – GOOD ENGINEERING PRACTICE

We try to keep one step ahead of the problem as much as we can, hence predictive, by identifying such things in our maintenance program. The old term from the FCC was GEP – Good Engineering Practice.

It is important to keep networking with other engineers. Robert Greene mentioned in his laws of power to “Avoid the unhappy and the unlucky” (Rule #10). He simply meant ‘hang out with winners and move away from “rat poison” personalities.’

In other words, you need to surround yourself with other engineers who figure things out and eat problems for breakfast. (There were some early mornings in my job when I ate problems for breakfast, chewing some microwave STL puzzles while driving to the site.)

There are people who just roll up their sleeves and get things done rather than start a pity party and just complain. Be with the thinkers, the doers, the movers and shakers and leave the others behind who do not really even want to follow. You have got somewhere to go to – attend web seminars and conventions – and meet people who are like-minded as you.

LOGGING – THE FORENSIC INVESTIGATION

You should treat your station log like a cold case investigator treats a crime scene.

Every “low RF” alarm or “high temp” spike is a fingerprint left by a future catastrophe. If you are not logging the “near misses,” you’ll never see the direct hits coming. The event log stored by your computer servers is a powerful tool. Get with the app tech support, analyze the data and find out what is going on.

We found some practices need to be included in our routine like re-booting Microsoft OS run devices on a regular basis. We also found out that we need to “bump” the NOM of our NRCS (News Room Computer System) once in a while. We did not wait for apps to hang up and cause confusion of our workflow. Forensic analysis of event logs showed us that buffer overflows do happen through the public Internet once in a while, so rebooting the encoder/decoder solves the issue.

There are so many other things that make troubleshooting of impending failures doable – if the event log is open to receive data and for you to draw the information out of it.

“GUT FEEL” VERSUS DATA DRIVEN

If you think that experience and “gut feel” alone can help you predict things, well and good for you. It does not work for me, however, so I might need that magic wand after all.

What I am blabbering about here is that data can be my magic wand. It does not work and give answers instantaneously but it will do on this planet.

I want the data collected by others, data that I collected over the years and data that can be collected in the station that I am working on. Some call it statistics. I call it evidence where I can draw conclusions from and ACT on it. Analyze, Categorize and Take action = fix it before it recurs!

So, can we really predict failures even without a crystal ball or a degree from Hogwarts? To a certain degree, the answer is, simply, Yes.

Being on top ahead of potential problems was all we needed before. Now, with AI and telemetry sensors that monitor real-time current draw and component heat, we can definitely do a much better job.

History is not just in books; it is in your maintenance logs. Study it, or be forced to relive it during your drive to the station at 3 AM.

– – –

Rolin Lintag, CPBE, is Director of Engineering for KSTF and Lincoln Asian Media in Brisbane, CA. You can reach him at rlintag@ktsftv.com.