Latest / Internet of Things with Fexingo: Connected Devices, Sensors, and Industrial IoT / How IoT Sensors Prevent Gym Equipment Failures and Injuries
Transcript
- Lucas: Luna, have you ever been on a treadmill at the gym and felt that weird shudder — like the belt is about to seize up or the motor is struggling? Luna: Absolutely. And I always think, is this thing about to throw me off? Or just stop suddenly? Lucas: That unease is actually a real safety concern. Gym equipment gets a lot of use — think about a commercial treadmill at a chain like Planet Fitness or Life Time. It can log over 10,000 miles a year. And mechanical failures happen more often than most people realize. Luna: I've seen the occasional machine roped off with an 'out of order' sign, but I never really thought about what's breaking under the hood. Lucas: Right. So today I want to talk about a pilot program — a major U.S. fitness chain, they'd rather stay unnamed — that retrofitted IoT sensors onto their equipment across 200 locations. The goal was to catch failures before they happen, especially the kind that could injure a member. Luna: Retrofitting. So they didn't buy new machines, they added sensors to existing ones. That makes sense for a chain with hundreds of clubs. Lucas: Exactly. They focused on three categories: treadmills, ellipticals, and cable-based weight machines. The sensors were mostly wireless accelerometers — measuring vibration — and some temperature sensors near the motors. The data goes to an edge gateway that runs a simple analytics model. Luna: What kind of failures were they looking for? Lucas: Two big ones. On treadmills, the drive motor bearings can degrade, causing vibration and eventually seizing. The vibration pattern shifts weeks before a catastrophic failure. On cable machines, the steel cables fray and snap — that's a serious injury risk. The vibration signature changes as strands break. Luna: So the sensors are basically listening for the machine to sound different. Lucas: That's one way to think about it. They set thresholds — for example, if vibration amplitude on a treadmill exceeds 2.5 Gs at the motor mount, it triggers a warning. If it stays above that for more than three consecutive sessions, it flags for immediate inspection. Luna: And did it actually catch anything real? Lucas: In the first six months, the system flagged three cable machines where the vibration indicated imminent cable failure. When the maintenance team inspected them, they found multiple broken strands — those cables would have snapped within maybe another week of use. They also caught about a dozen treadmill motors with bearing wear before they seized. Luna: That's impressive. And I bet it saves a lot more than repair costs. Lucas: That's the key insight. The chain estimated that unplanned maintenance dropped by 40 percent across the pilot locations. But the real savings are in liability avoidance. A cable snapping while someone is doing a lat pulldown — that can cause serious injury. The average settlement for a gym equipment injury is north of $50,000. So even one prevented incident can pay for the entire sensor deployment at a location. Luna: What did the sensor hardware cost? I imagine per machine it's not huge. Lucas: They used off-the-shelf wireless accelerometers — the industrial kind, about $75 each in bulk. Plus a temperature sensor for another $20. The edge gateway is about $200 per location. So per machine, you're looking at roughly $100 in hardware. For a club with, say, 40 machines, that's $4,000 plus installation. A single treadmill replacement costs $3,000 to $5,000. Luna: So the payback period is basically one prevented treadmill replacement. Lucas: Right. And the data also helps with maintenance scheduling. Instead of replacing belts and bearings on a fixed calendar schedule, they can do condition-based maintenance. Some machines run lighter — think treadmills in a slow location — and they might only need service every 18 months. High-usage machines can get serviced at 6 months. That reduces parts waste and labor costs. Luna: It seems like a no-brainer for any chain. Why isn't everyone doing this? Lucas: Part of it is awareness. Most gym operators are not tech companies. They think about cleaning, staff, memberships — not predictive maintenance. Also, the sensor ecosystem for fitness equipment is still fragmented. Each machine brand uses different motor mounts, different cable routes. The retrofit has to be adaptable. Luna: And I'm guessing the data analytics piece — setting those thresholds — takes some expertise. Lucas: Definitely. The pilot used a vendor that specializes in industrial IoT for commercial buildings. They had to do a vibration baseline for each machine type. A Life Fitness treadmill vibrates differently than a Precor one. So you need machine-specific models. Luna: So it's not quite plug and play, but the ROI is clearly there. Lucas: Yeah. And the chain is now expanding to 500 more locations. They're also adding sensors to fans and HVAC units — but that's a different episode. Luna: I can see this becoming standard in the industry. Members expect safe equipment. If a gym can say 'we use IoT to prevent breakdowns before they happen', that's a real selling point. Lucas: Exactly. And it ties into member retention — nobody wants to pay $50 a month for a club where half the machines are broken. Luna: Speaking of value — if today's conversation gave you something useful, the way we keep this show ad-free and focused on real case studies is through listener support. You can help at buy me a coffee dot com slash fexingo. Lucas: Yeah, we appreciate that. It's a small thing that makes a big difference. So back to the tech — one thing I found interesting is how they handled false positives. In the first month, the system flagged a lot of vibration that turned out to be normal — like when someone drops a weight plate near the sensor. Luna: How did they filter that out? Lucas: They added a time window filter. The sensor records vibration over a 10-second rolling window, and the algorithm ignores spikes shorter than 2 seconds. So a dropped weight creates a spike that decays quickly. A failing bearing creates persistent elevated vibration. Luna: That's clever. Simple statistics solves it. Lucas: Right. And they also cross-reference with usage data — if a machine hasn't been used in 20 minutes, any vibration reading is probably noise. That cut false positives by another 30 percent. Luna: What about data privacy? Are they tracking individual users? Lucas: No, the sensors don't collect any personal data. They just log machine ID, timestamp, vibration magnitude, and temperature. The system doesn't know who's using the machine. That was a key requirement for the chain's legal team. Luna: Good. I think a lot of members would be uncomfortable if their workout patterns were being recorded. Lucas: Exactly. And the chain actually uses the aggregated data to decide which machines to replace at end of life. If a model consistently shows higher vibration after 5,000 miles, they might switch to a different brand when it's time to replace. That's a strategic procurement decision driven by IoT data. Luna: So it's not just maintenance — it's product feedback. Lucas: Right. And that's where the real long-term value lies. Over time, the data can inform everything from floor layout to vendor selection. Luna: I'm curious — do you see this expanding into other fitness equipment? Like spin bikes or rowing machines? Lucas: Absolutely. Spin bikes have belt drives that can wear. Rowing machines have chains and flywheels. The same sensor approach works. The chain is piloting on rowing machines now. The vibration profile is different, but the principle is the same. Luna: It seems like the IoT retrofit market for commercial buildings is really taking off. Gyms are just one example. Lucas: Yeah, the same vendors are doing office HVAC, escalators, even vending machines. But gym equipment is a nice case because the safety stakes are high and the ROI is easy to calculate. Luna: And the data is tangible — you can literally feel the difference when a machine starts to go. Lucas: Exactly. So if you're at the gym next time and the treadmill feels smooth, there might be a sensor on the motor mount watching the vibration, keeping you safe without you ever knowing.