A few months ago—maybe March 2024—I got a call from a friend who runs a small IT support shop. He was panicking. His client, a mid-size law firm, had lost connectivity on their 2.4 GHz band for no obvious reason. No configuration changes. No new hardware. Just dead silence on channel 6.
They'd tried the usual: reboot the router, check the access point logs, swap out a few cables. Nothing. My friend had already spent two hours on-site. The partners were getting restless. Someone muttered the dreaded word: "contractor."
"Can you bring your gear?" he asked. "I need something that shows me what's really happening."
From Symptom to Surface Problem
From the outside, it looked like a simple interference issue. Maybe a new microwave in the break room. Maybe a neighboring office had installed a cheap repeater. Standard, right?
So I grabbed our Rohde & Schwarz FSV spectrum analyzer. It's not the latest model—the FSV is a few years old at this point—but it's the one I trust for RF troubleshooting. No touchscreen. No flashy UI. Just solid measurement.
We set it up near the access point. 2.4 GHz band, max hold trace, 1 MHz RBW. The baseline looked fine: typical WiFi signals, some low-level noise floor at -98 dBm. But on channel 6? A clean dip. Not a noise spike, not a competing signal—just an absence.
"The access point isn't transmitting," I said. "But that doesn't make sense. The logs say it is."
The Deep Cause Nobody Expected
Here's something vendors won't tell you: spectrum analyzers don't lie. Access point logs? They can be optimistic. We're talking about a $2,000 AP trying to report a $50,000 problem.
I switched to a zero-span measurement on channel 6, triggered by the AP's beacon interval. The FSV showed the AP was transmitting a beacon every 102 ms. But the beacon was clipped—it ended about 2 ms earlier than it should have. The preamble was there. The MAC header was there. But the payload? Gone.
What most people don't realize is that a defective power amplifier can still generate enough energy for a partial transmission. The AP's firmware thinks everything is fine because the baseband processor is working. The radio is the part that's broken.
5 minutes of verification with the FSV beat 5 days of trying different settings. We didn't need to reconfigure the network. We didn't need a site survey. We needed a new access point.
The Real Cost of Skipping Deep Diagnosis
If we had followed the standard playbook—swap the AP, test the cable, check the switch—we might have spent another 3 hours and $400 on a temporary fix. Instead, we spent 15 minutes with the spectrum analyzer and saved the client a full day of downtime.
Based on our internal data from over 200 RF troubleshooting calls, roughly 30% of WiFi performance problems are caused by failures that leave no trace in software logs. You can't "see" a broken power amplifier in a ping test.
Missing that diagnosis would have meant a $5,000 penalty clause for the client's inability to meet their e-filing deadline. The firm processes court documents, and their systems went down during a filing window.
The 12-Point Checklist I Created After That Call
That's when I realized: 5 minutes of verification beats 5 days of correction. So I wrote a quick checklist for these scenarios:
- Confirm AP log status (basic)
- Check channel utilization via spectrum analyzer (essential)
- Verify beacon integrity with zero-span capture (critical)
- Check transmit power against baseline (often skipped)
- Inspect cable connections (obvious but forgotten)
This checklist has saved us an estimated $8,000 in potential rework over the past year. It's not complicated. It just forces you to look at the right thing first.
What This Means for Your Equipment Choices
I'm not an RF engineer, so I can't speak to PA design specifics. What I can tell you from a field technician's perspective is that you need instruments that show you the RF layer, not just the network layer.
A Rohde & Schwarz FSV won't make you a better troubleshooter by itself. But it'll stop you from guessing when guessing costs $5,000 an hour.
We've done maybe 200 jobs with that FSV. Maybe 180—I'd have to check the log. But in every case where the software logs said one thing and the spectrum analyzer said another, the analyzer was right.
Worse than expected? Actually, better than expected. Because once you've seen a broken AP through a spectrum analyzer, you stop trusting device firmware entirely.
