Thursday Night, 6:34 p.m.
I was putting on my jacket when the office phone forwarded to my cell. The caller was a plant manager from a packaging operation outside Dayton. Their line had stopped at 2:10 p.m., and the maintenance crew had traced the alarm to the Balluff LVDT sensor that reports position on the forming press. Normal replacement time through their approved vendor was five business days. The line restart was scheduled for Saturday at 7 a.m., roughly 36 hours away.
In my role at an industrial automation distributor, I coordinate urgent replacements for this exact scenario. I've handled maybe 240 rush requests since 2017—could be a few less, I'd have to check—and most are straightforward. This one was not.
It's Tempting to Just Ship the Part
People think Balluff support is mostly logistics: find the part, confirm stock, arrange overnight shipping, done. Logistics matters, but it doesn't prevent callbacks. The fault history said the LVDT signal drifted by about 0.012 inch on one cycle, recovered, then disappeared after the third cycle. That pattern is a red flag. A worn or damaged LVDT usually fails gradually or acts up under a specific mechanical condition. An intermittent-then-open signal is much more likely to be a connection or cable problem. If we only swapped the sensor, we were betting that every wire, connector, and seal in the signal path was fine. I don't like betting a Saturday restart on that.
The maintenance manager asked why I didn't just release the order. I answered with a question: if the replacement gets there in 12 hours and the fault is still in the cable, what happens then? He wasn't happy, and he wasn't wrong to be unhappy. Every hour down was costing roughly $400. But saving one hour of investigation isn't a win if it turns into a second shutdown next month.
Eliminate What It Isn't Before Replacing What It Is
The motion controller on that machine also receives absolute position from an ATM60 encoder on the lift axis, and the alarm text mentioned a feedback mismatch. If the ATM60 encoder was faulty, we'd be sending the wrong part. We asked their electrician to watch the encoder count on the HMI while the technician moved the axis by hand. The count moved cleanly in both directions. That six-minute check ruled out the encoder and pointed us back to the LVDT channel and its cable.
Then we inspected the Balluff LVDT sensor itself. Once the unit was on the bench, the quality lab manager suggested putting the connector end under their VHX digital microscope. That was a no-brainer. At 50x magnification, we could see green corrosion on two pins and water stains around the shell. The connector seal had dried out, washdown water found its way in, and the sensor signal went from intermittent to open.
The VHX digital microscope didn't diagnose the machine on its own. It gave us eyes on a detail people usually ignore. The original Balluff LVDT sensor passed the bench output test. The transducer was still alive. The connector and cable were not. If we had followed the original plan, we would have replaced an expensive sensor with another expensive sensor and left the corroded cable in place to fail again.
Calipers, Brackets, and the Wrong Debate
Before reinstalling the original sensor, the maintenance lead measured the mounting bracket gap to confirm that the LVDT core was still in its correct mechanical range. That triggered the old debate: Starrett vs Mitutoyo digital calipers. The maintenance lead prefers Mitutoyo because of the feel and the IP rating; the quality manager has used Starrett for twenty years. If you've ever looked up Starrett vs Mitutoyo digital calipers, you know both camps have loyal followers.
On that job, both calipers agreed within 0.001 inch, so we ruled out misalignment. That check mattered because a core sitting outside its designed travel range can produce an erratic output that looks exactly like a sensor failure. The tools weren't the point. The discipline to measure twice was.
The Fix Was Not a Replacement
We ordered a new mating connector, a new cable seal, and re-routed the cable so water couldn't pool near the connector housing. We reinstalled the original Balluff LVDT sensor. Saturday at 6:48 a.m., the line jogged without an alarm. At 7:00, first shift made good product. The spare LVDT sensor we had pulled from stock never left the shelf.
That outcome didn't happen by luck. Our team now follows a simple rule when a customer asks for an emergency sensor replacement: inspect the failure before releasing the part, or document why that isn't possible. It took about twenty extra minutes. If we had skipped it, that machine would likely have failed again within a month, and the second repair would have cost far more than a connector.
Five minutes of verification beats five days of correction.
What I Want You to Take From This
Next time a position sensor goes intermittent, don't start with a part number. Start with the fault log. Look for damage or moisture in the connectors. Check other feedback devices on the same axis. If a VHX digital microscope is available, use it before you condemn the component. And measure the mechanical setup with a good caliper—Starrett, Mitutoyo, or anything else that repeats within spec—because the machine might be telling you that something moved, not that the electronics died.
The parts will still be there in twenty minutes. The expensive mistake is the one that fails after you've already celebrated the fix. Balluff support, done right, means closing the loop on why a failure happened. That's what separates a rushed repair from actual uptime.
