The 5,000 Hours You Lose When Nobody's Watching

(upbeat music) - G'day, and welcome back to Maintenance Break. Bites eyes to insights, one break at a time. I'm Pete. - And I'm Drew. Pete, I wanna start with a number because you told me this one off air and it stuck with me. Same site, same fleet, same final drives, same duty. One era, 20,000 hours before rebuild. A few years later, 15,000. What changed? - The oil, that's it. Same machines, same application, same operator group doing the same job. I was on that site through both eras. Early on, oil cleanliness was a daily, non-negotiable focus. Sampled every service, and if a result came back dirty, someone acted on it that week. Final drives ran to 20,000 hours routinely. - And then it changed. - Yeah, there was a change in site leadership and here's the thing Drew, nobody made a decision to stop caring about oil cleanliness. Nobody signed off on that. It just stopped being anyone's daily priority. Sampling remained, the results just sat in an inbox instead of triggering action. Dispensing got sloppy because nobody was checking it anymore. Wear rates crept up across the fleet. And inside 18 months, we were rebuilding final drives 5,000 hours earlier. - 5,000 hours of component life gone per final drive across a whole fleet, and nobody actually chose that. - That's the one that gets me. Nobody chose it, it just wasn't watched. So today we're talking about flak, fuel, lubricant, air, coolant. And I'll say this upfront, because people hear flak and picture someone wandering out once a fortnight to pull an oil sample. It's bigger than that. It's every point fluid touches an asset. From the moment you choose the product, right through to checking whether what you did actually worked. - Walk me through that because contamination control sounds like one thing, and I don't think it is. - It's not one thing, it's a chain. You select the right product for the application. You procure it. You receive it on to site without it picking up anything on the way in. You store it properly. You transfer it into the machine without introducing anything new. You monitor it while it's in service. And if a result comes back wrong, you respond, fix it, and verify the fix actually worked. Break any one link and the rest doesn't matter. - And that chain isn't one person's job. - No, and that's the part most sites get wrong. It's three people, minimum. Each controlling a different link. The maintainer, dispensing and handling the fluid at the point of use. The storman, keeping what's in the shed clean and correctly identified before it ever reaches a component. And the engineer setting the cleanliness target in the first place and making sure monitoring results are acted on. - All right, let's put a number on performance because dirty oil doesn't land with me as a mechanism yet. - Every component has a point where it's degrading but hasn't failed yet. What reliability people call the PF interval, the potential failure to functional failure window. Contamination doesn't cause an instant failure. What it does is drag that potential failure point earlier and it shrinks the warning window you've actually got to catch it and act. - So the final drive story isn't really about the oil going from plain to dirty. It's about how much runway you had left to notice. - Exactly, and it's not one cleanliness number for every system either. That's a mistake I see a lot. Hydraulics need to run cleaner than a gearbox. A gearbox needs to run cleaner than some other systems again. The goal is to keep them as clean as practically possible to minimize wear. - And that's on the engineer to set correctly. - That's the engineer's job specifically, yes. Set the right target for the right component and make sure whatever your commissioning is actually clean before it goes into service. Not cleaned up after the fact. Cost then, because I can already hear someone saying, it's oil, how expensive can getting it rolled B? - There's a rule reliability engineers use for exactly this. The earlier in that chain you catch a problem, the cheaper it is to fix. Ideally, clean fluids into components. But failing that, the next best time to act is to clean it up as soon as possible. The longer it runs contaminated, the more premature wear happens. - So the 15 versus 20,000 hour story is the cost-letting premature wear extract its price. - That's exactly what it is. Nobody signed off on an expensive decision. They just let it cheat problems sit long enough to become an expensive one. - Right, and what about risk? Because so far, this sounds like a maintenance cost story. - The risk here is that maintenance downtime's increase to change components early. And that can put pressure on production for assets that are the bottleneck. - Hang on, that sounds like cost again. Plan downtime's still downtime. - It's not the same thing. If you're monitoring properly, you're choosing when that component comes out on your terms scheduled around production. The risk is what happens when nobody's watching and that potential failure point I mentioned earlier runs straight past you. Then you're not scheduling a change out anymore. You're reacting to a failure on the assets timeline, not yours. - And if that assets the bottleneck, then the whole site's output is riding on a component nobody's had eyes on. That's the risk, not that it costs more that you don't get to pick when you find out. - Okay, time for this week's break tip. Tomorrow morning, go and have a look at your main dispensing point in the workshop or the quick fill points on an assets components. Look at two things. Are fill points capped or protected from the elements when not actively being used? And is the area dispensing and fill points clean? - And if the answers no on any of those? - Check if you have a standard for the cleanliness first. If you don't, start there. If you do, it's a work quality education opportunity. That's us for today. If you found this episode useful, like, subscribe and share it with a mate. Until next time, take care. (upbeat music)