One missing line in the fluorescence list costs you later

1. Four words in the tender, exposed on site

The specification says fluorescence attachment included, and every bidder prices it. After installation the user arrives with a double-stained slide: green channel bright, red channel black. The configuration list has one filter cube.

Adding a cube is not a simple purchase order. Bands must match, the size must fit the turret, lead time must be scheduled, and the acceptance form waits. A half-day acceptance turns into a coordination meeting where nobody is pleased. It happens several times a year.

2. One cube serves one dye set

A filter cube is an exciter, a dichroic and an emitter as a matched set with fixed bands. The one for FITC will not show Texas Red. How many you need depends on how many dyes you run.

Multi-band cubes do exist and cover two dyes in one slot, saving turret space, at the cost of bleed-through and lost brightness. On weak samples that trade is not always worth it.

More cubes is not automatically better either. Bands you never use cost money and crowd the turret, which makes switching riskier. Build the list from the dyes you actually run.

3. The light source is a consumable

A mercury lamp is billed twice: a few hundred hours of life, and gradual decay. Late in life brightness drops, exposure stretches, and weak signal drowns in background.

LED sources last longer and need no warm-up, but the spectrum must line up with the cubes. Their cost spreads over a decade, not over one purchase. An excitation peak that misses the filter wastes the entire channel.

Ask four things on the sheet: source type, rated life, price per replacement, and whether a service visit is required.

4. Objectives and camera are on the bill too

Transmission and numerical aperture of a fluorescence objective decide how much light reaches the camera. A general-purpose objective still produces an image, but exposure stretches, photobleaching accelerates, and the sample dims before the run ends.

Judge the camera by sensitivity, not pixel count. Under weak signal, sensitivity buys short exposure and low light dose, which living samples need. Gain can rescue a frame, but noise comes with it, so do not treat gain as the normal setting.

5. Stray light and slides are easy to forget

Fluorescence suffers from stray light: a room that is not dark, a missing shield, a monitor left on. Wrong slide thickness and autofluorescent mounting medium lift the whole background.

None of this costs extra, and it belongs in acceptance: lights off, a blank slide, one frame, is the background clean. That tells you more than the datasheet. Check whether the stand has a mounting point for a light shield; adding one later is not always possible.

A fluorescence configuration follows the experiment, not the phrase included if available. Describe the dyes, the specimen type and whether you image continuously, and the list can be right the first time.