Keysight Device FAQ: Bi-Directional Power Supply, N5173B, Voltage Drop Calculator, and the Keysight vs Cisco Question

If you've ever searched for a Keysight instrument with a deadline breathing down your neck, you know the feeling. I've spent the last 10 years coordinating test-equipment purchases for labs, prototyping shops, and field-service teams. Most orders aren't urgent. The ones that stick in my memory are the ones that are.

Here's a quick list of questions I've answered more times than I can count:

• What is the Keysight bi-directional power supply used for?
• Is the Keysight N5173B enough for RF testing?
• How do I use a voltage drop calculator when choosing a power supply?
• Why does 'Keysight vs Cisco' show up in search results?
• What's the real total cost of a Keysight device?
• Can I get a Keysight instrument fast without overpaying?
• Do I need the top model, or is the mid-range okay?

The answers below are based on what I've learned from rush orders, spec-sheet mistakes, and one very expensive lesson about voltage drop.

What is the Keysight bi-directional power supply used for?

A normal bench supply only sources current. A Keysight bi-directional power supply can source current and sink it. That means it can act as both power supply and electronic load, absorbing energy from the device under test instead of just pushing power into it.

From the outside, that might sound like a marginal feature. It isn't. If you're testing batteries, fuel cells, supercapacitors, or power converters, a bi-directional supply lets you simulate charging and discharging without switching between separate instruments. In my role triaging rush orders, this is the device I recommend when a customer calls about a battery tester or PV inverter test and doesn't want to wait for a custom load bank.

But don't buy one if you only need steady DC for a circuit. The bi-directional functionality raises the cost. For plain bench work, a standard Keysight power supply is the better total-cost decision.

Is the Keysight N5173B enough for RF testing?

The N5173B is an EXG X-Series RF analog signal generator. Based on Keysight's public datasheet, checked in January 2025, it covers 9 kHz to 13 GHz, depending on the option you choose. It handles CW signals, AM/FM, phase modulation, and frequency sweeps. For a lot of RF validation work, that's plenty.

When I look at a request for an N5173B, the first thing I ask is whether the customer needs vector modulation. If you're testing a simple receiver, a filter, or an amplifier, the N5173B is usually the right tool. If you need QAM, 5G NR waveforms, or arbitrary I/Q modulation, you need a vector signal generator. That's a different model, like the N5182B, and no amount of hurry makes the N5173B do that job.

One lab I worked with in December 2024 had to replace a dead signal source before a customer acceptance test. The N5173B was in stock, it matched the old unit's frequency range, and it was delivered in two days. A vector generator would have delayed everything and wasted budget.

How do I use a voltage drop calculator when selecting a power supply?

Voltage drop matters more than most people think. If your power supply is set to 12V but the device under test only sees 10.5V because of thin or long cables, your test results are worthless.

The basic formula is:

VD = (2 × I × L × R) / 1000

where I is current in amps, L is total cable length in feet, and R is resistance in ohms per 1000 feet. Double the length because current flows out and back.

Here's a concrete example. Take 10 amps, a 10-foot cable run, and 14 AWG wire, which is about 2.5 ohms per 1000 feet:

VD = (2 × 10 × 10 × 2.5) / 1000 = 0.5V

Half a volt at 10 amps doesn't sound huge. But on a 5V logic rail, it's a 10% loss. I knew I should check the cable on a job last year, but I thought it was just a short cable. It wasn't. The device was glitching, and the problem wasn't the supply. It was the drop.

Before you order the Keysight equipment, use a voltage drop calculator to plan the cable. Then pick a power supply with enough headroom, or use shorter and heavier copper.

Why does 'Keysight vs Cisco' show up in search?

I'll be blunt: this isn't a direct competitor comparison.

Cisco makes networking infrastructure, like switches, routers, wireless controllers, and firewalls. Keysight makes test and measurement equipment, including network test platforms and RF instruments. When someone searches for Keysight vs Cisco, they're usually asking one of two things:

• Should I use a Keysight network tester to validate Cisco equipment?
• Should I buy Cisco gear or a Keysight tester for my lab?

The answer depends on which layer you're dealing with. If you're building a production network, you buy network infrastructure. If you're benchmarking, troubleshooting, or validating that network, that's a test and measurement job.

I've seen this confusion in person. A customer said network test and I heard network switch. Same words, different meanings. Once we clarified that the switch was the device under test and the Keysight tool was the measurement system, the whole project made sense.

What's the real total cost of a Keysight device?

Sticker price is the starting point, not the ending point. The total cost of any test device includes:

• Base instrument price
• Options and accessories
• Calibration, initial and ongoing
• Warranty or extended coverage
• Shipping and setup
• Downtime if it fails or arrives late

I've seen a quote for one power supply come in lower than a competitor's, then grow by more than 20% after calibration, rush shipping, and a cable kit that should've been included. The expensive quote was actually the cheaper one.

That's why I use total-cost math before comparing any quote. In the long run, a Keysight device with a proper calibration plan and a firm delivery date often has lower TCO than a cheaper-looking device that needs constant attention.

Can I get a Keysight instrument fast without overpaying?

Fast, good, cheap: you usually only get two of the three. But there are ways to avoid the worst rush-order markup.

First, ask about in-stock units. A distributor that has the exact model on the shelf can ship next-day for a flat rate, which is almost always cheaper than a special-order rush from the factory. Second, ask about demo and refurbished units. These are sometimes available with full warranty and a shorter lead time. Third, don't request overnight shipping on a Friday. I don't know why this always needs repeating, but a weekend in shipping terms means Monday.

One example: in March 2024, I had to get a Keysight N5173B to a client with 36 hours before a deadline. The normal lead time was two weeks. We found an in-stock unit, paid about 5% more for expedited delivery, and it arrived with five hours to spare. The client's alternative was a re-test slot three months out.

That's the thing about rush orders. Sometimes the cost isn't the fee. It's what happens if you don't move fast.

Do I need the top model, or is the mid-range okay?

Most projects don't need the flagship. A 13 GHz N5173B is a mid-range signal generator, and that's the right level for many labs. The same logic applies to power supplies: buy enough current and enough accuracy, but don't pay for a range you won't use.

I'm not a metrology engineer, so I can't tell you exactly which calibration option is right for every measurement. What I can tell you from a procurement perspective is that overspecifying is a form of waste. It makes each quote look expensive, it often extends lead time, and it doesn't make your test better.

Trust me on this one: a matching spec will get you faster, cheaper, and easier than a more capable device that sits idle at 20% of its range.