It Started With Two Quotes
It was March 2024 when our RF engineer dropped a quote request on my desk. "N6752A and an 81150A," he said. "And while you're at it, tell me why everyone in the forum is arguing about the 117 multimeter."
I've managed this lab's equipment budget for seven years. We're a 45-person engineering company, and I've overseen about $180,000 a year in electronics test gear. I negotiated through the 2020 component shortage, through the shipping delays in 2021, and through the annual budget drama ever since. I am not easily surprised by a quote. But that week, I got a lesson I keep re-learning.
The Cheap Quote That Wasn't
Let me start with the hardware. The Keysight N6752A power module is a 50V, 10A, 100W module that drops into the N6700 mainframe we already own. Per Keysight's published datasheet, it's designed for clean DC rail testing—which is exactly what a 5G IoT sensor design needs. The Keysight 81150A is a pulse/function/arbitrary waveform generator, exactly what our team needed for generating glitchy, noisy, real-world signal conditions.
Vendor A sent me a quote for both. The price looked great. I actually opened a purchase order. Then I read the fine print and felt my stomach drop.
Vendor A's quote said "calibration included." Vendor B's quote said "C210 calibration included." They sound similar, but they're not. Here's something vendors won't tell you: "calibration included" can mean completely different things. One vendor might mean a basic calibration certificate. Another might mean Keysight's C210 accredited calibration—the one with measurement uncertainties, traceability documentation, and a report that quality auditors actually recognize. For us, that difference can make or break a product release.
Vendor A's quote was lower. I'm a procurement person, so I looked at total cost, not the first line. Shipping? Vendor A charged freight. Vendor B folded it into the price. Calibration after 12 months? Vendor A's instrument would need a separate calibration cycle before its first annual check. Vendor B's C210 report covered the first cycle and gave us a baseline. By the time I added that up, the gap was only about $600—not $3,000. I almost ignored Vendor B because they looked more expensive on the first screen. That's a mistake. The total cost of ownership made them the cheaper choice. As I tell our new hires: the cheapest equipment is the equipment you have to buy twice.
So glad I checked the quote line by line. I was one click away from approving Vendor A's order on a Friday afternoon. That would have meant an $1,800 surprise six months later, plus a tense email to our quality team. Dodged a bullet.
Why C210 Made the Difference
What most people don't realize is that C210 isn't a luxury. It's a paperwork decision. If you are building automotive parts, medical devices, or anything a customer might audit, the calibration certificate is as important as the measurement itself. The instrument could be perfect; without the right certificate, it's just a box with a blinking number.
If you're prototyping and nobody external needs to see your data, you can skip C210 and save money. You can also buy a used generator and hope for the best. But "hope" is not a calibration plan. In regulated work, the C210 report is the difference between a clean audit and a long email thread with your quality manager.
C210 is not a model number. It's a service option in Keysight's calibration portfolio. I mention that because c210 gets searched by people who don't know what it is—I was one of them. It's easy to assume C210 is a hardware variant. It isn't. It's the calibration report that tells you the instrument's uncertainties on the day it left the factory.
We chose C210 on both units. That meant an extra line item on the PO, but it also meant the engineer didn't have to stop and wonder whether the measurement was defensible. It simply was.
The Waiting Game
One thing that caught me off guard was lead time. Vendor A listed the N6752A as "in stock." Vendor B said "ships in two weeks." For a lab, two weeks is usually nothing. It gave us time to prepare the slot in the mainframe, update the test procedure, and train one of the technicians on the 81150A's software. In hindsight, that buffer was a feature, not a bug. If Vendor A's module had arrived in three days, we would have been the bottleneck.
What most people don't realize is that "in stock" on a quote often includes buffer time to manage a vendor's queue. It's not necessarily faster. It's just less committed. A concrete delivery date is worth more than a vague stock marker.
The Multimeter Question That Followed
The engineer's second question turned out to be the more interesting one. Every few weeks, someone in our shop searches "which multimeter to buy." In those threads, the 117 multimeter is what almost everyone recommends. And for a lot of people, they're right.
Look, I'm not going to say the 117 multimeter is bad. It's rugged, easy to use, and reliable. If you're a field technician working on 480V panels, an electrician-style meter is a perfectly sound choice. I've bought them for maintenance staff, and they do exactly what we need. If your job is checking wiring and you ask "which multimeter to buy," a 117 multimeter is a genuinely sensible answer.
But in our lab, we're measuring millivolt-level sensor drift and checking DC power stability on test boards. A handheld multimeter—even a very good one—isn't the right first tool for that. We use a benchtop Keysight meter with higher resolution, data logging, and a display you can actually read at 2 a.m. The 117 multimeter is great for what it's for. It's just not what we're for.
The "you have to spend a fortune for lab accuracy" thinking comes from an era when multimeter options were basically "cheap handheld" or "expensive lab rack." That's changed. A modern bench meter is affordable enough that the choice is about application, not budget scare stories. The right answer isn't "buy the 117" or "always buy bench." It's: know what you're measuring, then choose the meter that can see the problem.
What I'd Buy Again
The N6752A power module has run almost daily for seven months. It holds a 5V rail with the kind of stability that our old linear supply could only approximate. Or rather, the old supply might have been as good, but we couldn't prove it—there was no fresh calibration record. The 81150A has generated maybe a thousand ugly waveforms. Both came with the C210 calibration certificate, and that bit of paper has already been pulled into two audit packages. No one has argued with it.
If you're doing serious DC or signal-integrity work, put the N6752A and the 81150A on your short list. Pair them with C210 if your work will ever be audited. If not, ask yourself whether the certificate is worth it, and make that decision consciously. Do not let a low quote decide for you.
And if you're trying to decide which multimeter to buy? Start with your work, not the forum. A 117 multimeter can be the right call for the right person. For a precision lab, a bench meter makes more sense. There's no single best instrument. There's only the best one for the measurement you need to defend.
At least, that's been my experience in a 45-person lab where "close enough" doesn't make it into a report.
