I'm a field support engineer handling network deployment orders for medium-scale tower projects. I've been doing this for about 6 years now. I've personally made (and documented) 8 significant mistakes, totaling roughly $14,500 in wasted budget and rework. Now I maintain our team's pre-deployment checklist to prevent others from repeating my errors.
This one's specifically for the Anritsu S362E Site Master — that handheld cable & antenna analyzer that's practically standard issue for tower work. But honestly, the steps apply to most Anritsu field analyzers (S332E, S412E, you name it).
Here are the 5 things I check every single time now. Do these in order, and you'll catch 90% of the common gotchas before they cost you a site visit.
It's tempting to think "the sticker says it's good for another 6 months, we're fine." That's the simplified view. But calibration isn't a light switch that's either on or off.
I learned this the hard way in September 2022. We had an S362E come back from annual calibration with a fresh sticker. First job with it — the return loss measurements on a new sector were all over the place. Spent 4 hours on the tower swapping jumpers before someone checked the calibration constants. The unit had been factory reset during calibration and the frequency correction factors were zeroed out. The sticker said it was calibrated, but the internal data said otherwise.
What to actually do:
I now include a screenshot of the cal status screen in our pre-deployment photos. It takes 30 seconds and saves the embarrassment of showing up on site with a paperweight.
This one sounds obvious, but I've seen it trip up people with 10 years experience. The problem is the default frequency range on the S362E is usually 2 MHz to 4 GHz or 6 GHz. That's too wide for most cable sweep work, and too narrow for certain filter tests.
I once had a tech tell me "the PIM test passed, but the VSWR looked weird." Turns out he had the analyzer set to sweep from 600 MHz to 2.5 GHz when the antennas were 3.5 GHz panels. The sweep stopped before the operating band. He would've spotted it if he'd set the range properly.
Quick reference (I keep this on a sticky note in the case):
The extra few seconds to confirm this before you close the case — that's where the efficiency win is. I've seen teams cut their on-site calibration time by 15 minutes just by doing this in the truck.
This is the one step that I'd say 60% of new users skip. And it's the one that causes the most rework.
The S362E will let you calibrate at the RF port on the unit itself. Don't do that — unless you're testing something plugged directly into the unit. For any tower work, you need to calibrate at the end of your test port cable (i.e., at the point where you'll connect the antenna or jumper).
Here's what I do:
The first time I skipped this (note to self: never rush the cal), the entire set of readings for a 6-sector deployment was off by roughly 0.8 dB. We'd "fixed" perfectly good jumpers because the analyzer said the return loss was marginal. That was a $3,200 mistake — 12 jumpers pulled and replaced, two extra site visits, and a lot of explaining to the project manager.
You'd think this is obvious, but I've had to chase down "trace 001" and "trace 002" on multiple S362E units and had no idea which was which. The default naming convention is useless for field work.
My naming convention (stolen from a senior tech who taught me):
[Site Code]_[Sector]_[Test Type]_[Date]_[Tech Initials]
Example: TWR-42_S1_RTL_2025-01-15_JD
TWR-42_S1_DTF_2025-01-15_JD
TWR-42_S2_RTL_2025-01-15_JD
Do this before you start sweeping. I set up the naming in the Save Setup screen before I even start the OSL cal. That way, when I finish each measurement, I can save with two button presses instead of fumbling with the touchscreen while the wind's blowing and the battery's dying.
I've also learned to save the trace before I disconnect the cable. Because the temptation is to disconnect, see the trace change, and realize you saved the wrong one. Save first, then disconnect.
The most frustrating part of field testing: getting back to the office and realizing the notes you took are useless. You'd think written notes on a clipboard would work, but in the rain, with gloves on, your scribbles become hieroglyphics.
Here's what I do now (after losing 3 hours trying to decipher my own handwriting from a Crown Castle site back in 2021):
I went back and forth between paper notes and digital for a long time. Paper notes offered speed; digital offered reliability. Ultimately I landed on a hybrid — paper for quick field notes, digital for the final records.
1. Using the wrong test port cable.
The S362E is sensitive to phase stability in the test cable. Using a cheap or worn cable introduces measurement uncertainty. I keep a dedicated set of Anritsu-spec'd test cables in the case and replace them every 12 months (or sooner if they get stepped on).
2. Forgetting to check the battery.
The analyzer will run for about 4 hours on a full charge. I've shown up with 30% battery and a 6-sector site. Now I charge it overnight and check the battery level before I leave the office.
3. Not updating the firmware.
Anritsu releases firmware updates that fix measurement quirks (not that the unit has many, but they do happen). I check for updates quarterly. The 2024 update fixed a bug with DTF range calculations on the S362E that I'd been working around for months.
4. Trusting the "Auto Save" feature blindly.
The auto-save is useful, but I've had it save corrupted traces when the battery died mid-save. Now I always manually save as a backup.
I keep this checklist laminated in the lid of my S362E case. It's saved me from repeating my own mistakes, and it's been a quick reference for the newer members of our team. Copy it, modify it, use it. But I'd recommend testing it against your own workflow first — every team's a little different.