Drift behaviour is known to two or three technicians: HR for diagnostic equipment makers

A district laboratory calls on a Thursday. Its semi-auto biochemistry analyser reads creatinine controls slightly high, every morning, then drifts back by afternoon. The service engineer on site has swapped the lamp and cleaned the cuvette. Nothing changed. Back at the factory, only one calibration technician has seen this before. He knows it happens on that model when the incubator block sits near a window and the morning room is cold. He is away at his village till Monday, and the lab is threatening to return the analyser.

Our note on people issues at diagnostic equipment makers opens with this pressure. Which calibrator to believe. How peristaltic pump tubing stretches over weeks. Why one batch of photodiodes wants a different gain. Hardly any of this is on paper. It lives with the few people who have nursed hundreds of units, and it leaves whenever one of them resigns.

• Warm-up behaviour. How long each model needs before readings settle, and how that changes in cold or humid rooms.
• Ageing parts. The way a halogen lamp, pump tubing or probe seal shifts readings slowly before it fails outright.
• Reference sample quirks. Which control serum lots run slightly off, and which ones the senior technician quietly avoids.
• Component batch effects. A new lot of sensors, strips or optical filters that needs a different offset at calibration.
• Firmware compensation. Why a temperature correction or filter threshold was set at a particular value, and what breaks if it is changed.
• Field symptoms. The pattern in a lab's complaint that points to drift rather than operator error or a bad reagent.

The cost rarely appears on one line. It shows up as service visits repeated because the first engineer guessed. It shows up as loaner units stuck in the field for weeks. It appears as burn-in racks held while a junior recalibrates a board three times. It shows in a hospital tender where the buyer asks about field support and the owner hesitates. And it shows in pay. Once two technicians realise nobody else can tame a stubborn analyser, they can name their price. Or they accept the offer a bigger instrument house has dangled for months.

Drift is learned unit by unit, cuvette by cuvette. A technician notices, adjusts, notices again, and builds a feel over years. Writing that down takes hours nobody grants. Bench technicians and service engineers sit in separate rooms under separate bosses. So a symptom reported from a lab never meets the person who saw it at burn-in. Seniors also sense that their value lies in being the only ones who know. Without a reason to share, the safest move for them is to keep it.

1. 1

   List the models and the holders

   For each analyser, meter or reader you sell, write which technicians and engineers can settle a drifting unit without help. Any model with one or two names is your first priority.

2. 2

   Start a drift casebook

   One entry for every unit that fails first calibration or comes back from the field. Symptom, room conditions, what was checked, what fixed it and who was present. Keep it on a shared drive that service engineers can search.

3. 3

   Pair juniors on every failed unit

   No senior closes a stubborn case alone. The junior watches, then does the next one under supervision, and the casebook records both names.

4. 4

   Bring bench and field together

   A short weekly call where service engineers describe open complaints and bench technicians suggest causes. It turns scattered symptoms into shared patterns.

5. 5

   Record firmware reasons

   Each time a threshold, filter or correction factor changes, the engineer notes why in plain words. A one-line reason beats none.

6. 6

   Name a second holder per model

   Agree when a named backup should settle that model unaided, then prove it on a genuine reject.

7. 7

   Reward the teaching

   Count casebook entries and juniors brought to sign-off in the senior's appraisal, with a visible pay rise attached.

Model, serial number and firmware version: ____.
Where found: burn-in / first calibration / field complaint.
Symptom in plain words: ____.
Room temperature, humidity and warm-up time noted: ____.
Reference or control lot used: ____.
Checks done, in order: ____.
What fixed it, and why the technician thinks it worked: ____.
Senior and junior present: ____.
Similar earlier cases in this casebook: ____.

Few people combine electronics with a working sense of laboratory instruments, and larger instrument firms know it. Retention starts with defining the roles clearly. A calibration technician who can settle three models deserves a higher grade than a newcomer. Our piece on writing technician roles for analyser makers covers how to describe these grades. For naming successors to the key holders, see succession for diagnostic equipment makers.

When you do hire, test for the right thing. Hand candidates a real drifting unit from your casebook and ask them to think aloud. Their reasoning reveals more than any instrument list on a CV.

Sit with your two most trusted calibration technicians this week and ask a simple question. Which model would you least like a newcomer to touch, and why? Their answers are the first chapter of the casebook and the first pairing to arrange. See the hub covering instrument, implant and disposables makers for how this links to engineer growth and tender peaks. If you want help with grades, a learning plan or hiring a second holder, outline your workshop for us.