Latest / Internet of Things with Fexingo: Connected Devices, Sensors, and Industrial IoT / How IoT Sensors Are Preventing Grain Silo Explosions on Farms
Transcript
- Lucas: So there's this number that's been stuck in my head for weeks: five people died in a grain silo explosion at the Didion Milling facility in Wisconsin back in 2017. And when OSHA finally released its investigation report, they found that the temperature inside the grain storage tank had been climbing for three days before the blast. No one noticed. Luna: Three days? That's not a sudden event. That's a slow-motion disaster. Lucas: Exactly. And that's the thing about grain silo explosions — they're not random. They follow a very predictable chain. You need three things: grain dust suspended in air at just the right concentration, an ignition source, and oxygen. Grain elevators have all three, constantly. The only variable is whether you detect the heat spike early enough. Luna: So how does an IoT sensor break that chain? Lucas: By giving you continuous, real-time data on the two things that matter most: temperature trends and methane concentration. Grain naturally off-gasses methane and carbon dioxide as it spoils. If you're measuring both — plus particulate density in the air — you can spot the conditions for an explosion hours or even days before they become critical. Luna: Right, because by the time a human walks into the silo and feels the heat, it's often too late. Lucas: And walking into a grain silo is already one of the most dangerous things you can do in agriculture. You've got the risk of engulfment — grain acts like quicksand — plus toxic fumes. So the whole point of IoT here is to keep people out of the silo entirely. A wireless sensor node, about the size of a fist, sits inside the storage tank and pings data to a dashboard every five minutes. Luna: Is that technology actually being used, or is it still in the pilot phase? Lucas: It's in production. Companies like BinMaster and 4B Components have been selling these nodes for years. But here's the catch: adoption is shockingly low. A USDA survey from 2024 found that less than 12 percent of U.S. grain storage facilities use any kind of continuous temperature monitoring. Most still rely on manual checks — someone walks the bin once a shift, sticks a probe in, takes a reading. That's like checking your car's tire pressure by kicking the tire. Luna: Why? Cost? Or just inertia? Lucas: Both, but cost is the big one. A single wireless temperature sensor node used to run about two thousand dollars. For a farm with a dozen silos, that's twenty-four grand, plus the gateway and the software subscription. For a lot of family operations, that's real money. But the cost has been dropping fast. The latest generation of LoRaWAN-based sensors — long-range, low-power — can be had for about a hundred and fifty dollars per node. Luna: That's a huge drop. What changed? Lucas: Two things. First, the sensor hardware itself got commoditized. The MEMS temperature and humidity chips you find in a twenty-dollar smart home sensor are basically the same ones you'd put in a grain silo. You just need a more rugged enclosure and a different wireless protocol. Second, the telecom industry built out LoRaWAN networks for smart agriculture — irrigation, soil monitoring, cattle tracking — so the infrastructure already exists. You don't need to install a cellular modem; you just piggyback on the farm's existing IoT network. Luna: So the unit economics actually works now. But you said adoption is still only 12 percent. Lucas: Yeah, and that's the puzzle. The cost of retrofitting a single silo with sensors is roughly the same as the cost of one hour of a grain elevator's downtime after an explosion. The math is obvious. But agriculture is a thin-margin business and farmers are famously skeptical of anything that feels like a new expense. You have to prove the ROI in a really concrete way. Luna: Which brings us back to Didion Milling. That case was the catalyst for something? Lucas: It was. Didion ended up paying over ten million dollars in OSHA fines and settlements. And in 2025, OSHA proposed a new rule specifically requiring continuous monitoring of temperature and combustible dust levels in grain handling facilities. It hasn't been finalized yet, but the direction is clear: regulators are moving toward mandating the kind of data that IoT sensors already deliver. Luna: Interesting. So you've got a regulatory push meeting a technology pull, and the price is finally right. That sounds like the ingredients for a real shift. Lucas: It does. And there's one more angle I think is really cool: the technology transfer from industrial flour mills to farms. Flour mills have been using vibration sensors and temperature probes for decades — it's standard practice. But the sensor packages were custom, expensive, and wired. What happened in the last five years is that companies took those industrial-grade sensors, made them wireless, and repackaged them for the agricultural market. Same sensor, different label. Luna: So the farm gets industrial safety tech at consumer electronics prices? Lucas: Pretty much. And not just temperature. Some of the newer nodes measure methane, carbon monoxide, and airborne dust concentration simultaneously. If the methane level spikes above five percent of the lower explosive limit, the system can automatically trigger ventilation fans or shut down augers. That's an active safety system, not just a warning light. Luna: Which brings it closer to what you'd see in a chemical plant. But on a farm. Lucas: Yeah, and that's the trajectory. The interesting business question — the one we haven't figured out yet — is who pays for this on the thousands of small family farms that don't have the capital for a full retrofit. Is it the farmer? The grain buyer? The insurer? Because insurance premiums for grain elevators are going up fast. If a sensor package can lower your premium by, say, fifteen percent, the payback period drops to under two years. Luna: That's exactly the kind of incentive that could tip the scales. It's not just about safety compliance — it's about the bottom line. Lucas: And that's the thing about good IoT applications. They don't sell on altruism. They sell on a spreadsheet. Luna: You know, speaking of spreadsheets and sustainability — this is a tangent, but it connects — I was reading about a pilot project in Iowa where they're using the same temperature and methane sensors to monitor grain quality for ethanol production. If the grain spoils, the ethanol yield drops. So the sensors pay for themselves just by preventing spoilage. Lucas: That's a great example. It's the same sensor, same installation, but a different value stream. Suddenly the ROI isn't just about explosion prevention — it's about product quality. That makes the decision easier for a farmer. Luna: Right. And I think that kind of multi-use case is what will drive adoption faster than any OSHA mandate. Lucas: Completely agree. Let me zoom out for a second. If you look at the broader IoT in agriculture market, grain storage monitoring is a small slice — maybe three hundred million dollars globally. But it's growing at about 18 percent compound annual growth rate, and the regulatory tailwind is just starting. I'd expect to see more startups targeting this niche, especially ones that bundle hardware, connectivity, and insurance discounts into a single subscription. Luna: So the business model is evolving from selling boxes to selling outcomes. Lucas: Exactly. And that's where IoT really hits its stride: when it stops being a gadget and becomes an input to a financial decision. Luna: On that note — and I'm not trying to sell anything here — but one of the reasons we can dig into this kind of specific technology angle without ads or sponsors is that a handful of listeners support the show directly. It's a small group, but it means we can spend ten minutes on grain silo sensors without anyone telling us to keep it short for a commercial break. Lucas: It's true. We keep the show ad-free and the episodes focused, and that's only possible because of the listeners who chip in at buy me a coffee dot com slash fexingo. If that's something you've been meaning to do, it really makes a difference. And if not, no pressure — we're glad you're here either way. Luna: So back to the silos. One thing I've been curious about: are there any startups specifically building IoT for the grain storage aftermarket, not as a OEM install? Lucas: Yeah, a few. One that caught my eye is a company called GrainSense — they make a handheld probe that you stick into the grain, and it measures moisture, protein, and temperature in about thirty seconds. That's more for quality assessment than explosion prevention, but it shows the trend: cheap, instant data. Luna: But it's not continuous. So you'd still miss a three-day temperature climb. Lucas: Right. Continuous is the key. And there's a startup called SiloSense that's doing exactly that — wireless nodes that sit in the bin and report every ten minutes. They claim a 40 percent reduction in spoilage and a 60 percent reduction in manual inspection hours. Those are the kinds of numbers that get a farmer's attention. Luna: And presumably the insurance companies are paying attention too. Lucas: They are. I spoke with a risk manager at a major ag insurer — off the record — and she told me they're actively looking at telematics data from grain storage to adjust premiums. If they can verify that a farm has continuous monitoring, they're willing to offer a discount. That's a direct financial incentive, not just a safety one. Luna: So the insurance industry is effectively acting as a subsidy for IoT adoption. Lucas: In a way, yes. And that model has worked before — think of how auto insurers drove adoption of telematics devices. The same dynamic is playing out in agriculture, just slower. Luna: What about the smaller farms that can't afford the upfront cost even with the discount? Lucas: That's the hard part. Some states have started offering grants. Illinois, for example, launched a program in 2024 that covers 50 percent of the cost of grain bin monitoring systems for farms under a certain size. But it's limited. I think the real solution is going to come from the grain buyers — the ethanol plants, the flour mills, the feedlots. If they start requiring monitoring as a condition of purchase, adoption will accelerate fast. Luna: That's a powerful lever. It's not regulation, it's supply chain pressure. Lucas: Exactly. And we're already seeing that in the organic grain market. Some buyers require documented storage conditions to maintain certification. Once that becomes standard, the sensors are no longer optional. Luna: So we're at an inflection point: costs down, regulation looming, insurance incentives appearing, and supply chain requirements starting to emerge. Lucas: I think so. And the technology is solid. The sensors work, the networks exist, the data is actionable. The remaining gap is just distribution and awareness. Which is where a show like this comes in, I suppose. Luna: I hope so. Because the alternative is more incidents like Didion, and that's just not acceptable when the solution is already on the shelf. Lucas: Yeah. It's one of those rare situations where the tech is actually ahead of the adoption curve. The sensors are ready. The question is whether the industry is ready to use them.