Why I Now Believe Preventive Maintenance Beats Disaster Response (A $6,800 Lesson)

I believe most cone crusher failures are preventable. Not all, but a significant majority. I didn't always believe this. In fact, I used to think that the best approach was to react swiftly and competently when something broke. I was wrong.

This opinion isn't based on a textbook. It's based on a single, expensive mistake I made with a Sandvik CH870 cone crusher's Hydroset system back in Q3 2021. That mistake cost my operation roughly $6,800 in direct downtime and replacement parts. More importantly, it destroyed a week's worth of production schedule. That's what I want to talk about.

The Mistake That Changed My Mind

I was a maintenance supervisor handling equipment orders for about four years at that point. We were running four Sandvik CH870s in our secondary crushing circuit. The Hydroset system on one unit was behaving strangely—the pressure was fluctuating a bit more than usual on start-up.

Everyone told me to check the hydraulic accumulator pre-charge before looking at anything else. It's a simple check: nitrogen pressure should be within spec. I didn't listen. I assumed it was a main control valve issue and ordered a replacement kit. The kit cost about $2,400. We had to shut down the crusher for two shifts to swap it.

The result? The new valve didn't fix the problem. The pressure fluctuation was still there. We then did the pre-charge check (ugh) and found the accumulator was flat. A $40 recharge and a $15 seal kit sorted it out. The $2,400 valve kit sat on a shelf, and we lost an additional 12 hours of production. Total waste? The parts cost plus 30 hours of lost crushing time. I calculate that at roughly $4,400 in lost production at that time, plus the valve cost. Maybe $6,800, give or take.

That's when I learned a hard lesson: check the simple, documented thing first. It's tempting to think you can diagnose complex machinery by feeling it out. But the 'always check the pre-charge' advice isn't just a saying—it ignores the consequence of not doing it.

Why a Simple Checklist is Your Most Effective Tool

After that disaster in September 2021, I created a 12-point pre-start checklist for our CH870s and all Sandvik drills in the fleet. The checklist itself is boring. It lists things like verifying accumulator pressure, checking the oil level in the lubrication tank, and running a calibration test on the ASRi system.

The numbers said to spend 20 minutes on these checks. My gut said we didn't have the time (unfortunately). The first week, we found three issues on four crushers: a slight drop in lube oil flow, a leaking hose fitting on a Hydroset unit, and a worn-out tramp release cylinder seal that was weeping oil.

To be fair, the team didn't adopt it happily. They saw it as extra paperwork. (Should mention: we had to adjust the checklist twice in the first month because we'd included a couple of steps from the wrong manual.)

In the 18 months since, we've caught 47 potential errors using that checklist. Most were minor—a loose bolt, a slightly low accumulator charge. But three were serious issues that could have led to a catastrophic failure, like a cracked main frame that we spotted during a visual inspection prompted by the list. The cost analysis was clear: we spent roughly 60 man-hours per quarter on checks. Before the checklist, we were averaging one 'surprise' failure every quarter. Each failure cost us an average of $2,500 in downtime and parts. The checklist costs us about $600 per quarter in labor. That's a 4:1 return. In my view, a checklist is the cheapest insurance you can buy for your Sandvik mining equipment.

The Complexity Most People Ignore

It's tempting to think a modern cone crusher like the CH870 is 'smart' enough to self-diagnose. The ASRi system reports a ton of data—power draw, pressure, CSS. But the data can be misleading. The system tells you there's a pressure deviation, but it doesn't always tell you why. It could be a rock box issue, liner wear, hydraulic contamination, or an accumulator problem. The data points to the symptom, not the cause.

Everyone I talk to says, 'We just watch the ASRi screen.' That's a huge risk. The ASRi system is brilliant for optimizing performance, but it's not a replacement for physical checks. The 'just trust the sensor' advice ignores the reality that sensors drift, get clogged, or fail. Our checklist includes a manual verification of the CSS with a physical lead brick once a week. It takes 10 minutes. In August 2022, that brick check revealed the ASRi was reading a 6mm CSS when it was actually 9mm. We were crushing too fine without knowing it.

For Sandvik parts like the Hydroset system, the same principle applies. You could order a new set of HPU components based on a symptom (like slow response), or you can check the simple things first: oil viscosity, filter condition, and accumulator pre-charge. Most issues (I'd estimate 70%) are traced back to these fundamentals. The official Sandvik manual says to check these first. I finally took that advice a bit too late.

Addressing the Obvious Objection: 'We Don't Have Time'

I hear this a lot. 'We can't spare 20 minutes to check every crusher.' I get it. Production pressure is real. But consider this: a catastrophic failure on a CH870 can mean a 24-48 hour complete shutdown to replace a main shaft. That's a loss of, say, 8,000-10,000 tonnes of throughput.

Let's do very rough math. If your operation makes, say, $5 per tonne, that's $40,000-$50,000 in lost revenue. A 20-minute daily check costs you about $10 in labor. The argument that you can't afford the 20 minutes is the exact thinking that leads to $6,800 mistakes. It's not a cost. It's a risk mitigation investment. I've seen it on a $3,200 order of wear parts where someone skipped verifying the part number (the CH870 and CH440 use similar liners) and we had to wait 48 hours for the right ones.

Some people argue that experienced operators don't need a checklist. I used to think that. It's wrong. Checklists are not for training; they are for ensuring consistency. My most experienced operator once forgot to close the lube oil drain valve after a filter change. A checklist would have caught it. He didn't. We lost a $1,200 bearing. He's the best we have, but humans make errors. The checklist is the safety net.

So I'm not softening my view here. I believe a robust, documented, and enforced checklist for your Sandvik crushers and drills is non-negotiable for anyone serious about reducing unplanned downtime. The data from my operation over 18 months proves it. It's not a suggestion. It's a core operating procedure.

The 12-point checklist I created after my third mistake has saved us tens of thousands of dollars in potential rework. Five minutes of verification beats five days of correction. Every time. Learn from my $6,800 mistake: check the accumulator first. Use the checklist.