Keysight Power Sensors, Multimeter Accuracy, and Sensor Comparisons: A QA Manager's FAQ

Every week I review test equipment specs, calibration certificates, and measurement uncertainty budgets. After four years of that, I've answered the same questions about Keysight gear more times than I can count. Here are the ones that keep coming up, plus a couple you probably haven't asked yet.

What makes Keysight power sensors different?

Power sensors are the gatekeepers of RF and microwave metrology. A Keysight power sensor paired with a compatible power meter gives you a reading that can be traced back to a national standard—if the sensor is calibrated correctly. The sensor is the expensive part, and the meter is just the readout. That's why you'll see 'Keysight power sensors' as separate line items in capital budgets.

Keysight's E-series and U-series sensors cover a wide frequency range—I want to say up to 18 GHz or beyond, but don't quote me on that; check the current datasheet. The bigger point is traceability. I once rejected a calibration certificate because the lab didn't log the room temperature during the calibration. At higher frequencies, a one-degree drift can shift your measurement more than your uncertainty budget allows. That level of detail matters.

Is a Keysight 6.5 digit multimeter overkill for a small lab?

If I had a dollar for every startup engineer who asked me that, I'd have enough for a nice lunch. No, it's not overkill if you need repeatable low-level measurements.

The Keysight 6.5 digit multimeter (like the 34461A) isn't just about showing more decimal places. It's about the stability of the voltage reference and the internal ADC. A handheld meter gives you a number. A bench DMM gives you a number you can trust tomorrow and next month.

But I'll be honest: if you're only measuring batteries and LEDs, buy the $200 meter first. Upgrade when you hit a problem the handheld can't answer. Small budgets mean buying equipment that earns its keep, not demonstrating another digit. I say this as someone who has watched a small client stretch a $5,000 budget across five different instrument needs. They deserved straight answers, not an upsell.

How to test a capacitor with a multimeter

1. Discharge it first. This is not optional. Use a discharge resistor or a purpose-built tool—not your fingers.
2. If your multimeter has a capacitance mode, insert the capacitor or connect the leads. Wait two to three seconds for the reading to settle.
3. No capacitance mode? Switch to resistance mode. A good capacitor will show low resistance briefly, then climb toward open as it charges.
4. A reading that starts shorted or open at the beginning is a failure. Replace the capacitor.

Confession time: I knew I should discharge a large electrolytic before testing, but thought 'what are the odds?' The odds caught up with me when the meter flashed 'OL' and the internal fuse blew. $80 fuse, total embarrassment, and a very memorable lesson.

So if you're searching for 'how to test a capacitor with a multimeter,' start with the discharge step. Every other piece of advice depends on it.

What is a PMD 3D sensor?

PMD stands for Photonic Mixer Device. It's a time-of-flight sensor that measures distance by emitting modulated light and measuring the phase shift of the reflected signal. You'll find these sensors in industrial laser scanners, safety systems, and some automotive lidar modules.

The surprise wasn't the measurement principle. It was how many people assume a 3D sensor's accuracy spec is straightforward. It isn't. Temperature, target reflectivity, and ambient light all shift the final reading. When you're qualifying a PMD 3D sensor for production, you need a calibrated optical setup and an oscilloscope that can verify the modulation waveforms.

A PMD 3D sensor isn't a Keysight product. But Keysight instruments are exactly what we use to validate that sensor's performance before it goes into a product. And if you're comparing vendors, ask for raw data—not just a marketing curve. That single question saves more audits than any checklist I know.

How sensors compare with Omron and Keyence

If you're trying to decide how sensors compare with Omron and Keyence, the first thing to understand is that you're looking at two industrial automation giants, not Keysight. Omron has a broad catalog: photoelectric, proximity, pressure, ultrasonic, and more. Keyence is known for high-end laser displacement and vision systems, and a famously aggressive sales force.

Keysight doesn't build those industrial sensors. Keysight builds the test and measurement equipment that verifies their performance—power supplies, digital multimeters, oscilloscopes. So the real question isn't 'which sensor brand wins?' It's 'how are you going to verify the sensor's performance in your own environment?'

When I audit a supplier that uses Omron sensors, I check their calibration log. If they use Keyence, I check the same thing. The brand tells me less than the evidence does. If you want a comparison that matters, look at linearity, temperature coefficient, response time, and repeatability—then test the actual unit.

What's the real difference between a 5.5 and 6.5 digit multimeter?

At first glance, 'half a digit' sounds like nothing. But it's not one digit—it's a different analog front end. A 5.5 digit meter is often enough for pass/fail testing. A 6.5 digit meter gives you more averaging options, better noise rejection, and a lower measurement uncertainty for calibration work.

Let me put it in QA terms: we had a fixture that read 5.4999 V on a 5.5 digit meter and 5.50014 V on a 6.5 digit meter. Neither was wrong. The 6.5 digit meter just had more usable precision. We chose the 6.5 digit for the final test after a $22,000 redo caused by a borderline measurement. That redo taught me to pick a meter that gives you margin, not just digits.

Does Keysight take small customers seriously?

This one never appears on a spec sheet, but it matters. Three years ago I was with a five-person startup. We needed a 10 dB attenuator and a calibration adapter—under $400 total. The order went through a distributor, but the Keysight applications engineer still spent 20 minutes on the phone explaining a connector nuance. We later placed an $80,000 order when we needed a spectrum analyzer.

I've never had Keysight or its channel partners make me feel like my order was too small. Maybe that's luck. But I think it's culture. It's also one of the reasons I'm still a Keysight customer.