The Pre-Installation Checklist I Use for Every SICK Sensor (After Making 11 Mistakes)

Why Are You Searching for "Sick Thermometer with Fever"?

If you've landed here searching for a sick thermometer with fever, this might not be the article you're looking for. We're not that kind of SICK. SICK AG is a German industrial sensor manufacturer. We make encoders, photoelectric sensors, moisture meters, and safety laser scanners. If you're an engineer, commissioning agent, or maintenance tech who's about to install one of these, this checklist is for you.

I've been handling service orders for SICK products for about seven years. Actually, eight, if you count my internship in 2016. In that time, I've personally made eleven significant mistakes that totaled roughly $23,000 in wasted budget. That's not counting the quiet embarrassments or the 2 a.m. callouts. So, I figured I'd write down the five-step checklist I run through on every single installation now. I'd rather you learn from my scar tissue than make your own.

Step 1: Verify Your Supply Voltage (and Don't Trust the Old Wiring)

In 2017, I was swapping out an old SICK WLL260 photoelectric sensor on a packaging line. The cable label said 24V, so I hooked it up. The moment I powered it, there was a pop and a faint puff of smoke. Turns out, that '24V' line was actually a 110V line that a previous electrician had rerouted for a temporary heater. I killed an $890 sensor in about 0.3 seconds.

The lesson? Always measure the supply voltage with a multimeter before connection. Don't trust labels. Don't trust the previous guy. Measure it. I mean, sure, it was a dumb mistake, but it was also an expensive one.

Step 2: Read the MR176 Moisture Meter's Output Range (I Learned This in September 2022)

The MR176 moisture meter is a compact humidity and temperature sensor that SICK sells for HVAC and industrial drying processes. It's a great little device. But it does not output the standard 4-20mA range you might expect. It can be configured for 0-10V or 4-20mA, but they have specific scaling caps.

I once configured the range incorrectly on a 47-piece order. Every single unit was scaled to the wrong humidity range, so the PLC was seeing 45% RH when it was actually 98% RH. The mistake affected a $3,200 order and took a week to rework. The fix? Don't assume your standard scaling. Pop open the datasheet, check the specific device's setpoints and the PLC's input scaling. They need to match. It sounds simple, but I get at least one frantic call a year from someone who's sure the sensor is broken when it's just a scaling mismatch.

Step 3: Ground Everything (Even the Stuff You Think Doesn't Need It)

Let's talk about power quality analyzers. We don't make them—we typically use a Fluke unit for troubleshooting—and the same hidden pitfall applies to them and to SICK encoders. I had an encoder signal that was randomly dropping out. The logic said it was the encoder. The encoder said it wasn't. I swapped it out—again, keep it on my tab—and the new one did the same thing.

It turned out to be a ground loop. The control cabinet had a 120V potential difference between the sensor ground and the PLC ground. I saved $80 by skipping the grounded cable shield on that installation. Ended up spending $600 on a replacement encoder and a day of troubleshooting. The rule now: the shield is not optional, and the ground bus bar is your best friend. Verify the potential between grounds before you power anything up. This goes for SICK programmable encoder setups and any other sensitive instrument in the cabinet.

Step 4: Program Your SICK Encoder Before You Mount It (Or Pay the Price)

This is the big one. SICK programmable encoders—like the ATM60-P4—are incredibly versatile, but they need to be configured before they're installed. In Q1 2024, we had a project with a heavy turntable. The supplier told me they had a pre-programmed spare encoder just in case. It looked fine on my screen.

We installed it, powered it up, and the turntable spun in the wrong direction and then jammed. The spare wasn't programmed for the right resolution or even the right code sequence (Gray code vs. binary). The result was a 3-day production delay, a $2,900 replacement motor breakdown, and a wasted encoder. We caught the error when the drive faulted out from overcurrent.

That's when I learned the golden rule: never trust a 'pre-configured' device. Plug it into your PC with SOPAS, verify the resolution, the rotational direction, and the code type before you install it. It's a 10-minute check that saves weeks of pain.

Step 5: Thermal Cameras See Surfaces, Not Walls (A Note on FLIR and IR)

I get this question constantly: 'Can thermal cameras see through walls?' The answer is no. I've had a client try to return a rental thermal camera from FLIR because they thought they could locate a hot water pipe inside a brick wall. They couldn't.

Thermal cameras see surface temperatures, not what's behind them. If you're looking at a wall, you're seeing the heat signature on the wall, not inside it. If you are using a thermal camera for a panel inspection, you're looking for hot spots on the terminals, which is the surface contact resistance. Don't blame the camera for physical limitations. This applies to all brands, not just FLIR. Also, for the love of everything, don't point a thermal camera at your own face unless you want a shockingly high number to stare back at you. The emissivity of skin is around 0.98, so your face reads pretty accurately, but it's still not a fever thermometer. If you need a sick thermometer with fever, we're a solid no.

The Common Thread: Assumption

Here's what connects all these failures: assuming the voltage was right. Assuming the configuration was right. Assuming the shielding didn't matter. I've been doing this for a while, and honestly, I'm still not sure why we keep making these basic assumptions. My best guess is that we're all just a little too optimistic.

So, the checklist is simple:

• Verify voltage.
• Verify scaling.
• Verify grounding.
• Verify program.
• Verify your thermography expectations.

That's it. It's not a 10-point quality system, but it's the five checks that have saved me from making my 12th significant mistake. We've caught 47 potential errors using this checklist in the past 18 months.

This was accurate as of early 2024, but SICK's software and part numbers change fast. Verify current specs on SICK's official support portal before you buy or install. If you're still running a fever and need a thermometer, I'm sorry—we can't help you. But if you need an industrial sensor that works the first time, try following this list.