Fine wine or snake oil? Ausenco's Grant Ballantyne helps miners sniff out real value in comminution tech before the hype goes straight to their heads

In the high-stakes world of minerals processing, separating fine wine from snake oil isn’t just a metaphor — it’s a business imperative. That’s the message from Grant Ballantyne, Technical Director at global engineering and consulting firm, Ausenco, who recently delivered a compelling presentation on how to critically assess the value and risk of novel comminution and classification technologies.

Speaking at a webinar hosted by AusIMM, Grant pulled no punches in his appraisal of the state of innovation in the sector. “Everything is actually oily wine,” he said. “You’re just as likely to get a comminution or classification technology that’s like fine wine as you are to get the snake oil. The idea is to work out where the oil is and where the wine is.”

It’s a timely reminder for an industry under pressure to reduce emissions, increase energy efficiency, and deliver more from increasingly marginal ore bodies — all while avoiding the pitfalls of over-promising technologies.

Getting the Numbers Right

Grant began by challenging a common industry myth: that comminution consumes 3 to 4 percent of global electricity. He pointed to more recent, ground-up studies that show a different picture. “If we’re looking just at the traditional mining areas, about 0.4 percent of global electricity is consumed by comminution. If we include coal, cement, and stone, it’s about 1.8 percent. It’s definitely not four — there’s very little evidence to say that it’s that high.”

This doesn’t diminish the urgency of the energy challenge. On-site, grinding circuits often account for more than 50 percent of energy consumption. But Grant stressed the importance of clarity and precision in how energy data is cited and used.

Measuring Energy the Right Way

When it comes to assessing novel machines, how energy consumption is measured can be just as important as the result itself. “If you’re using an electric meter, that motor isn’t normally operating under full load and full efficiency,” he explained. In a real-world example, he cited a mill drawing 0.5 kilowatts with only 70 percent efficiency: “Although it was drawing 0.5 kilowatts, it was actually only inputting about 70 percent of that into the pinion or drive.”

He also highlighted the broader system energy footprint — noting that ancillary equipment like fans and dryers in some vertical roller mills can offset any headline savings in mill energy.

The Steel That Eats Itself

One of the most surprising moments of the presentation came when Grant asked how long it would take for a mill to “eat itself” — that is, consume its own structural steel mass in grinding media. The answer? Seventeen days.

“Twenty kilograms per kilowatt is how much structural steel you need for a milling circuit. About 50 grams per kilowatt-hour is a typical consumption rate of steel media,” he explained. “That makes 17 days. I had to do this calculation multiple times because I didn’t believe it to begin with.”

The takeaway: grinding media is a huge contributor to emissions — and one that’s often overlooked. In emissions modelling, the steel media component (Scope 3) can dwarf the embodied carbon of the mill’s construction.

Circulating Load, “Fake Tonnes,” and the Trump Analogy

Circulating load and classification efficiency also came under scrutiny. In reviewing a circuit with six stages of hydrocyclones, Grant found the multiple layers added little value — and warned against overreliance on models that don’t account for “fake tonnes.”

“I like to think of them like Donald Trump would — they’re fake tonnes. They’re not real tonnes. Those tonnes are going around and around,” he said. “If you start playing around with the circulating load, it actually has a different effect on throughput than some of the models would predict.”

Grant described a novel classifier that was initially expected to boost throughput by 19 percent — but after correctly accounting for recirculation effects, the real gain was closer to 4 percent. “Which is nothing to be sneezed at,” he added, “but let’s make sure we’re assessing that correctly. Remember the axiom, classifiers don’t grind”

Particle Size: Don’t Rely on P80

P80 remains the industry’s go-to metric for particle size — but Grant says it’s often misleading when applied to novel technologies that produce non-standard size distributions.

“If the size distributions aren’t parallel in log-log space, you shouldn’t be using the P80,” he cautioned. He recommends instead using fixed sieve sizes, such as percent passing 75 microns, and compare energy input to fines generation.

The wrong measurement technique can also distort perceptions. In one case study, a new comminution machine appeared to produce fewer fines than a conventional HPGR — until it was discovered that dry sieving was underreporting fines. “We always recommend wet sieving below 38 micron,” he said.

Wear, Scale-Up, and the Long Road to Viability

Technologies like the HPGR and Vibrocone offer important lessons in the risk posed by excessive wear and unproven scale-up strategies. Grant revisited the infamous Cyprus Sierrita installation in the 1990s — where high wear due to smooth liners set back HPGR adoption by more than a decade.

On the Vibrocone, he noted: “It produced significantly more fines… but those fines came at the cost of wear.”

He urged developers to consider scale-up and wear as early as possible. “There’s a lot of risk associated with developing new technologies. Mining is already a risky business, and innovation can make it riskier.”

Pre-Weakening and Liberation: Overstated and Underdocumented

Claims of improved liberation or pre-weakening are common, but Grant warned they are difficult to quantify and often misinterpreted. “For mechanical breakage, unless you get to really high impact rates of 500 metres per second or really slow compression rates of millimetres per day, liberation by size is typically the same.”

Focus on Value and Risk Reduction

Ultimately, Grant’s presentation revolved around a central message: value and risk reduction must be addressed from day one.

“If your machine pre-weakens the rock amazingly but costs a fortune, it’s never going to be implemented,” he said. “You need to be looking at the economic value from day one — does it reduce your OPEX or CAPEX? Does it reduce your emissions, water use, tailings?”

He concluded with a tribute to Klaus Schönert, inventor of the HPGR, whose work took decades to gain industry traction. “You can see how long some of these things take — how invested these innovators are. Hats off to them.”

The Bottom Line

Grant’s address offered more than just technical insights — it was a call for sober, evidence-based thinking in a field prone to hype.

“Innovation is essential,” he acknowledged, “but only when it adds measurable value and is backed by transparent, system-wide data. Otherwise, we risk mistaking snake oil for the fine wine.”

To find out more on Ausenco, visit https://www.ausenco.com/