Application guidance for water, heat and gas monitoring programs with audit-ready evidence.
Itron application note cover
Application Note

Three Instrument Purchases, Three Different Mistakes: A Scenario Guide

2026-11-03 by Marcus Feld

The first time someone asked me which instrument to buy, I gave a clean, confident answer. That was 2017. I was wrong in about four different ways, and it took three separate failures before I noticed they were not the same failure at all.

I keep the receipts now. Fourteen logged mistakes between 2017 and 2025, roughly $11,400 in wasted budget, and one handwritten pre-check list that has caught 47 problems in the past 18 months. Most of those were paperwork. A few were genuinely expensive.

Here is what nobody told me early on: buying instruments is not one skill. It is at least three, and they look identical from the outside because they all end in a purchase order and a spec sheet.

So I do not sort instrument purchases by product category anymore. I sort them by what happens when the number is wrong.

  • You get audited. Someone outside your company will eventually question the reading, and the paper trail matters as much as the device. Water metering lives here.
  • You get dependent. The tool is only half the purchase. The other half is training, accessories, software, and spare parts, and all of that can evaporate when a model line dies. Thermal imaging lives here.
  • You get hurt, or you get a bad reading on something that matters. Handheld electrical gear lives here, and the failure mode is lopsided: usually nothing happens, occasionally something very bad happens.

Three scenarios. Same supplier catalogs. Completely different advice.

Scenario A: The meter that will be audited (Sensus water meters and flow metering)

In September 2022 I approved a 60-unit order for a meter replacement program. The spec sheet said the right thing. The order form had the wrong output option. We found out during install, when the first six units refused to talk to our reading system.

That one cost $4,800 in restocking and air freight on replacements, plus a nine-day hole in the schedule that I got to explain to a very patient customer. The units themselves were fine. The configuration was not.

Three things I do differently now, in roughly the order I wish I had learned them.

1. The dealer matters more than the price list

Search for Sensus water meter dealers and you get a long list, and not everyone on it is authorized in the way you need. I am not talking about counterfeit hardware (I have not seen that). I am talking about who handles firmware updates, who processes the RMA, and who knows which configuration option your AMI system actually expects.

Two questions I ask on every first call: are you an authorized distributor for this product line, and who performs warranty service — you or the factory? If the second answer is vague, that is the answer.

From the outside, dealer quotes for the same model look interchangeable. The reality is the quote covers the box, and the relationship covers everything after the box.

2. The model naming will trip you up if you search instead of reading a catalog

Sensus flow meter is a search term that returns everything from residential meters to commercial flow measurement, and SM flow meter pulls in yet another family with its own configuration ladder. I have assumed a model name carried over from one product family to another. It did not.

I do not have hard data on how often this catches people, but anecdotally it is the most common way I have seen six-figure programs lose a week. The fix costs nothing: pull the actual datasheet for the exact model number, not the family page, and read the ordering matrix line by line out loud. It feels stupid. It works.

3. The standards are the part that is not on the spec sheet

AWWA's C700-series standards for cold-water meters, NSF/ANSI 61 and 372 for drinking-water contact, and MID (2014/32/EU) if the units are headed to Europe are the things your utility's engineer will ask about in month three, not month one. None of these are printed on the marketing datasheet. Ask for the compliance documentation with the quote, not after the invoice.

The reverse of that assumption also matters: people assume the cheapest dealer is sourcing gray-market units. Sometimes. More often the hardware is identical and what is missing is exactly this paperwork plus the person who would have answered the phone.

Scenario B: The diagnostic tool that gets discontinued (thermal imaging camera TI125)

We bought three thermal imaging camera TI125 units back in 2016, when that was a current model, and they did exactly what we needed for six years. Then we tried to buy two more, and to source a couple of spare batteries, and discovered that the world had moved on without us.

That is not a hardware problem. It is a platform problem, and it cost us a lot more than the cameras.

What I ask before ordering any diagnostic tool

What is the current successor model, and when did it launch? If the answer is a launch date more than four years ago, I assume a replacement is coming and I plan for it.

What is the detector resolution, not the display resolution? A claimed high-resolution image is often interpolation. Ask for detector pixels in plain numbers and watch how the sales conversation changes.

Who recalibrates it, at what price, and how long will they keep doing that? I do not have hard numbers on how long discontinued instruments stay serviceable — my experience says five to seven years for batteries and software, less for calibration support — but that is a sense from a small sample, not data.

The uncomfortable lesson here is that emissivity and reflected temperature settings matter more than the camera does. A $10,000 imager with the wrong emissivity gives you a wrong number with great confidence, and a mid-range unit set up properly beats it every time.

The fundamentals of thermography have not changed. The execution absolutely has — detector prices dropped and report software got dramatically better, so what was a reasonable spec in 2018 is now underpowered at the same budget.

Scenario C: The handheld where brand loyalty is the wrong axis (Klein vs Fluke multimeter)

In 2019 I bought fourteen Fluke meters on the strength of the name. Half of them never left the shop, because the field crews did not want to carry something that expensive, so they used older meters instead — and one guy was checking a 480V panel with a no-name meter from a hardware store. That was the actual risk in our shop, and I had created it by optimizing for the wrong thing.

The Klein vs Fluke multimeter argument is usually about accuracy. That is the wrong axis. Two others matter more:

Safety category rating

The CAT III or CAT IV rating and the transient rating under IEC 61010-1 have to match where the meter will actually be used. This is not a marketing badge — it is the thing that decides whether the meter survives a transient or becomes shrapnel. A meter with a genuine CAT III 600V rating for the panels you actually open beats an overrated meter from a brand nobody can certify.

The calibration path

If your quality system requires traceable calibration certificates, check who can actually calibrate the model and what the certificate costs on your cadence. Fluke's service network is wide. Klein's is narrower. Over a five-year cycle that difference can flip the math entirely, and it is invisible if you only compare purchase prices.

Here is the counterintuitive part: for a twelve-person field crew, buying all Fluke was the wrong call for us. Mid-tier Klein for general service work, plus a couple of high-end Flukes for the measurements that get signed off, was cheaper and got used more. The expensive meter sitting in a cabinet protects nobody.

Flip that completely if you are in a lab or a calibration-driven environment. There, saving money on the meter is not saving money.

How to work out which scenario you are in

Three questions. Thirty seconds.

  1. Will anyone outside your company ever question the number? If yes, you are in Scenario A and paperwork goes on the purchase order, not on a to-do list.
  2. Will you need this exact model in six years, or just the capability? If it is the capability, you are in Scenario B and you need to know the successor before you buy the predecessor.
  3. If it reads wrong, does somebody get hurt or does a job get redone? If it is the first one, you are in Scenario C and you start from the safety rating, not the brand.

If two of them apply — and they often do, especially in a utility maintenance department — treat it as the higher-risk scenario and let the other one ride along. Adding the paperwork discipline of Scenario A to a handheld purchase costs you an hour. Skipping it costs you a quarter.

My current checklist has eleven items. The first one is not about the instrument at all. It is: who answers the phone in year three? Anyone can compare spec sheets. What separates a purchase from an expensive lesson is what you arranged around the instrument before you signed.

Marcus Feld

Marcus Feld

Marcus Feld is an electrical test and measurement analyst specializing in multimeters, oscilloscopes, clamp meters, insulation testers, spectrum analyzers, and data loggers. He applies IEC 61010-2-030 and IEC 61010-031 concepts while examining measurement category, bandwidth, true-RMS response, input loading, and stated uncertainty. His work helps maintenance engineers and test teams choose safe instruments with performance suited to the signals and environments they actually measure.

Leave a Reply