Equipment Verification Checklist: 5 Things I Check Before Trusting Any Reading

Equipment Verification Checklist: 5 Things I Check Before Trusting Any Reading

When I first started in QC, I assumed calibrated equipment was trustworthy equipment. It took a $22,000 redo and a delayed product launch in 2022 to unlearn that assumption.

I'm a quality compliance manager at a contract analytical testing lab. I review every test report before it reaches clients—roughly 200 reports per year. In 2024, I rejected 9% of first-draft reports due to issues that traced back to someone skipping an instrument verification step. That's about 1 in 10 reports that would have gone out with questionable data.

So I built a 5-step verification checklist covering the analytical instruments, dimensional tools, and electrical test equipment in our lab. It's not a replacement for accredited calibration. It's a routine that catches the failure modes that calibration stickers don't.

Step 1: Verify Your Gas Chromatography System Before the Run

Our Thermo Fisher gas chromatography system handles the heavy lifting for volatile organic compound testing. It's also the easiest instrument to trust blindly, because the software makes everything look fine.

Here's what I check before every critical run:

Carrier gas pressure and actual flow. The gauge might read correctly while a partially blocked column or a worn septum alters real flow. I measure the flow at the detector vent with a digital flow meter, rather than relying entirely on the instrument's readout.

Column condition via a system suitability mix. Retention times shifting more than 0.5% from the previous accepted run is my trigger to investigate. I run a standard mix and compare retention times, peak areas, and tailing factors against the last accepted run.

Detector response. I inject a known concentration standard and confirm the signal-to-noise ratio meets our threshold. If the Flame Ionization Detector (FID) response has dropped, the column may be bleeding, the detector may need cleaning, or the gas flows are off. The instrument's built-in diagnostics won't catch all of these.

The Thermo Fisher diagnostics suite is genuinely useful. But I've seen systems report "OK" while column degradation caused a 20% error in quantitation. A system suitability check is the only way to know for sure.

Step 2: Verify Raman Microscopes—Including Leased Units

We lease a Thermo Fisher DXR Raman microscope through Excedr's rental program. It's a practical arrangement for managing instrument costs. But here's a misconception I've seen cause real problems: leasing does not mean someone else handles verification.

Excedr handles equipment maintenance and repairs as part of the agreement. Verification of instrument performance is the user's responsibility—the same as if you owned it. I've seen labs skip calibration checks on leased instruments because they assumed the leasing company verified everything before shipping. That assumption can produce weeks of questionable spectra.

My routine for the DXR, done weekly or before intensive use:

Wavenumber calibration using a silicon wafer standard. The expected peak is at 520.7 cm⁻¹. If it's off by more than ±1 cm⁻¹, I log it and run the instrument's built-in calibration routine before proceeding.

Laser power at the sample surface. I measure it with a power meter. The set point and the actual reading should agree within tolerance. Drift here affects both Raman and fluorescence signal interpretation.

Optics path inspection. A contaminated objective lens or filter produces artifacts that look like real spectral features. I inspect the optics and clean them per the manufacturer's procedure when needed.

That's a 20-minute routine. Skipping it means you might generate a week's worth of unusable Raman data before anyone notices.

Step 3: Check Dimensional Tools—Encoder Measuring Wheels

Encoder measuring wheels get far less attention than analytical instruments. They're simple tools, and that's exactly why they're dangerous. The failure modes are physically subtle.

The most common issue is wheel wear. An encoder wheel's circumference determines the distance it measures. As the wheel wears, the effective diameter shrinks, and every reading is off by the same percentage. A wheel that loses just 0.1mm in diameter produces roughly a 0.1% error in every measurement. On a 50-meter measurement, that's 5 centimeters of hidden error.

Here's my checklist:

• Verify against a known length standard. We use a certified 1-meter reference bar. Roll the wheel across it and check the readout. Do it from both directions to check for directional bias.
• Inspect the wheel surface physically. Run a finger clockwise across the entire circumference. Flat spots, cracks, embedded debris, or uneven wear change the rolling diameter. Any of those means the wheel needs replacement.
• Test repeatability. Measure the same standard length five times. If the range of readings exceeds the manufacturer's tolerance, investigate the encoder electronics and mounting hardware—not just the wheel.

Last year, routine verification flagged a wheel that was off by 0.3%. The operator had been using it for months, and the error accumulated slowly enough that nobody noticed. So glad I ran the verification when I did—it was one batch away from becoming a customer complaint.

Step 4: Multimeter Repair and Self-Check

Multimeters are the most abused instruments in our lab. They get dropped, leads get yanked, and nobody thinks twice—until a reading looks wrong.

Before any electrical testing, I run a 3-point self-check:

Leads and probes. I inspect both leads along their entire length for cracks, nicks, or exposed wire. Bent or corroded probe tips cause intermittent contact. I've caught more wrong readings from bad leads than from any other cause. If a lead looks suspect, I replace it. Leads are cheap; re-testing is not.

Battery. Low batteries affect accuracy, especially in resistance and capacitance modes. I replace batteries on a schedule rather than waiting for the low-battery warning to appear.

Known reference check. I test against a precision voltage reference and a known resistor. If the meter reads within its stated accuracy, it's ready.

What about multimeter repair? My rule is straightforward: if the meter fails the reference check, it goes out for repair or gets replaced. I'm not an electronics engineer, and I won't pretend to be. Opening up a multimeter to fix internal faults is beyond the scope of daily QC—and doing it incorrectly creates a hazard. Our multimeters go to an authorized service center when they fail. That's the extent of my repair policy.

Step 5: How to Test a Capacitor with a Fluke Multimeter

Capacitor testing is part of our incoming inspection for electronic components. The question comes up often enough that I'll give you the method directly.

1. Discharge the capacitor first. This is non-negotiable. I use a 10kΩ resistor across the terminals for several seconds. The exact time depends on capacitance and voltage rating—when in doubt, wait longer and verify zero voltage with the multimeter before touching anything.
2. Set the meter to capacitance mode (µF). Most Fluke multimeters have a dedicated capacitance setting. If yours doesn't, you can get an approximate reading in resistance mode, but capacitance mode is the right tool.
3. Connect the leads with correct polarity. Electrolytic capacitors have marked positive and negative terminals. Connect the red lead to positive and the black lead to negative. Reversed polarity on an electrolytic capacitor produces a misleading reading.
4. Compare the reading to the rated value. Most capacitors have a tolerance of ±10% to ±20%. The measured value should fall within that window.

A capacitance value within tolerance is a good sign, but it doesn't guarantee the capacitor works under load. A capacitor can have correct capacitance while suffering from high equivalent series resistance (ESR) or leakage current. If the component must perform reliably in a circuit, you need an ESR meter or a dedicated component tester. A multimeter is a screening tool—I wouldn't sign off on a critical component based on capacitance alone.

Common Mistakes in Equipment Verification

After four years of reviewing test reports, here are the recurring problems I see:

Skipping verification on leased or rented instruments. I've said it already, but it bears repeating: a leasing company is not your quality department. When we leased the DXR through Excedr, I confirmed exactly what the agreement covered. Routine verification wasn't included. That's our job.

Trusting calibration stickers. A calibration certificate says the instrument met specifications on a specific date under specific conditions. It says nothing about what happened to it since. Verification catches what calibration stickers don't.

Testing capacitors without discharging them. I've seen a multimeter's current-input fuse blown because someone skipped discharge and measured a charged capacitor. The meter survived; the technician's confidence didn't.

Giving the simple tools a pass. The encoder measuring wheel and the multimeter are just as capable of producing bad data as a gas chromatograph. They fail in different ways, but the consequence is the same: a report that doesn't reflect reality.

One final caveat: this checklist is not a replacement for accredited calibration or formal quality standards like ISO 17025. It's a practical routine for catching common failure modes that calibration doesn't cover. If you operate under formal compliance requirements, your documented procedures take precedence over anything in this article.