Measuring Coolscan Dynamic Range With a Stouffer Step Wedge
Australian photographers who still shoot 35mm film, or who have inherited generations of Kodachromes from the 1970s, often want a hard number for how much latitude their Coolscan can actually capture. The marketing materials from Nikon in the late 1990s quoted figures around 3.6 to 4.2 D for the 4000 ED and the 8000 ED, but those numbers were measured under ideal conditions with freshly calibrated hardware. A scanner twenty-five years old, sitting in a Brisbane studio or a Perth garage, may not hit those original specifications. Testing with a calibrated Stouffer transmission step wedge gives an objective, repeatable measurement that anyone can perform on a kitchen table, without needing to ship the scanner anywhere.
The test is quick, inexpensive, and survives the postal system, which matters in a country where shipping a precision optical target from a Sydney supplier to a Hobart archivist can take the better part of a week. This article walks through what a step wedge is, how to use one with a Coolscan 4000, 5000, 8000, or 9000 ED, and how to interpret the resulting density curve for archival purposes.
What Dynamic Range Means in a Film Scanner
Dynamic range in a scanner describes the range of densities it can distinguish between pure black on the densest part of a negative and pure white in the brightest highlight. For 35mm colour negatives this often spans from roughly 0.1 D in the densest shadows of a Portra 400 frame to around 3.4 D in the most opaque highlight of a heavily overexposed transparency. A scanner needs to capture both extremes without crushing the shadows into a flat noise patch or blowing the highlights into clipped white.
The Coolscan range uses a fixed-gain analogue front end feeding a 14-bit or 16-bit analogue-to-digital converter, depending on the model. The 4000 ED and 5000 ED use a 14-bit A/D, while the 8000 ED and 9000 ED push 16 bits. That extra two bits in the higher-end models is what gives them the quoted dynamic range advantage, and it is also why an archivist scanning a tray of old Velvia slides in Melbourne will often notice the difference between an 8000 and a 4000 in the deep shadow detail under a forest canopy.
It is worth noting that the scanner's dynamic range and the scanner's usable dynamic range are not always the same thing. Noise in the analogue front end, ageing photomultiplier tubes in the LED-cooled models, and stray light bouncing inside the optics all erode the practical range. A step wedge measures the practical range, which is what actually matters when you are deciding whether to rescan a faded Ektachrome or accept the result you have.
The Stouffer Step Wedge as a Test Target
The Stouffer step wedge is a strip of 35mm-format film, or a larger sheet, printed with a series of discrete density patches arranged from nearly clear to nearly opaque. The most common model for Coolscan testing is the Stouffer TP-21, which provides twenty-one patches from 0.05 D to 3.05 D in 0.15 D increments. There are denser variants, and a finer-resolution model for users who want to look at the knee of the response curve.
| Step Wedge Model | Density Range | Step Increment | Typical Use |
|---|---|---|---|
| Stouffer TP-21 | 0.05–3.05 D | 0.15 D | Standard Coolscan range check |
| Stouffer TP-31 | 0.05–4.55 D | 0.15 D | Extended range verification |
| Stouffer TP-101 | 0.05–5.05 D | 0.05 D | Detailed response curve analysis |
| Stouffer R-2 Reflection Scale | 0.05–1.95 D | 0.10 D | Reflection scanner verification |
The wedges are manufactured to a calibrated tolerance, typically within 0.02 D of the labelled value, and they hold that calibration for years if stored away from heat and humidity. That makes them a much more trustworthy reference than a homemade greyscale target printed on inkjet paper, which is convenient but rarely accurate enough to draw conclusions from.
For Australian users, the practical consideration is sourcing. Stouffer Graphics is based in the United States, and a small step wedge plus international shipping can run into the AUD $80 to $120 range once Australia Post charges and GST are factored in. Local photo specialty shops in Sydney's Broadway or Melbourne's Nicholas Building occasionally stock them, but most archivists end up ordering directly.
Setting Up the Test on a Coolscan
Before scanning, clean everything. The Coolscan's glass carrier plates attract dust, and the inside of the LED housing can develop a faint haze over the years. A rocket blower, a microfibre cloth, and a careful pass across the film channel make the difference between a clean step wedge scan and one with mysterious bright spots in the middle patches.
Place the step wedge in a clean 35mm film holder with the density patches facing the light source. In the Nikon Scan software or the alternative VueScan driver, set the scan type to positive, disable ICE, GEM, and DEE, and set the output bit depth to 16 bits per channel. Disable any colour management during the test so the raw density values can be read without an embedded profile shifting the numbers. Scan at the scanner's native optical resolution, 2700 dpi for the 4000 ED or 4000 dpi for the 8000 and 9000 ED, and save the result as an uncompressed 16-bit TIFF.
Measure each patch in a tool like ImageJ, GIMP, or Photoshop by sampling the average pixel value over the centre of each step. Convert those pixel values back to density using the standard formula, density equals log base 10 of 255 divided by the pixel value for an 8-bit scan, or the equivalent for a 16-bit scan. Plotting density against the labelled step density produces the scanner's response curve, which can then be compared against the manufacturer specification sheet.
Reading the Step Wedge Curve
A perfectly linear scanner would produce a straight line, but no real Coolscan does. The curve flattens at the bright end, where the photomultiplier saturates, and at the dark end, where read noise and stray light add a floor. The difference between the highest density patch that still shows separation from the next darker patch and the lowest density patch that still shows separation from the next lighter patch gives the practical dynamic range.
For the 4000 ED, a healthy unit typically reads somewhere between 3.2 and 3.6 D of usable range. The 5000 ED, with its 14-bit A/D and improved optics, often reaches 3.5 to 3.8 D. The 8000 ED and 9000 ED, with their 16-bit converters and multi-line CCDs, can reach 3.8 to 4.2 D in good condition. If a unit measures more than 0.3 D below its model rating, the optical path probably needs attention: cleaning, realignment, or, in the worst case, a new LED or PMT.
Watch for the knee at the bright end. If the curve goes vertical rather than bending smoothly, the scanner is clipping detail in the densest slide regions, which matters for Kodachromes and Ektachromes from the 1960s and 1970s that have faded unevenly. A small Australian collection of family slides from a Sydney estate sale, for instance, may have a mix of well-preserved and sun-damaged frames, and the scanner's ability to hold highlight detail in the sun-damaged ones is exactly what this test will reveal.
Choosing an Archival File Format
The TIFF produced during the test is the right starting point for an archive, but most Australian archivists will also want a long-term preservation copy and a working copy. The three formats most commonly considered are TIFF, DNG, and JPEG, and each has a different balance of fidelity, file size, and future-proofing. A 16-bit uncompressed TIFF is the safest master copy, a lossless DNG is a reasonable alternative if file size or embedded metadata matters, and a high-quality JPEG is fine for working copies that will be shared online or printed at small sizes.
For an archivist scanning a deceased relative's 4000-slide Kodachrome collection, the choice usually comes down to storage budget and how often the files will need to be re-migrated. Cloud storage in Sydney data centres is reasonably priced, but a full set of 4000 dpi 16-bit TIFFs from a 36-exposure roll can run to 4 GB, which adds up quickly across a collection of several hundred rolls. The archival file format guide covers the trade-offs in more detail and walks through the metadata fields worth preserving for a multi-generational archive.
Whichever format is chosen, the bit depth should never drop below the scanner's native output. Saving a 16-bit scan as an 8-bit JPEG during the archival pass throws away precisely the highlight and shadow detail the dynamic range test was designed to verify.
Australian Climate and Long-Term Storage
A scanner that measures beautifully today can drift over the next decade if it lives in an environment that stresses the optics and electronics. The Queensland coast, including Brisbane and the Gold Coast, runs humidity well above 70 percent for much of the year, which encourages fungal growth on internal mirrors and on stored film. Hobart and the Tasmanian highlands run cooler but also damper for much of the year, while Adelaide and Perth present a dry heat that bakes the rubber drive belts and ages the plastic film carriers faster. Melbourne sits somewhere in the middle, with sudden temperature swings that stress the mechanical film transport more than they stress the optics.
Storing the Coolscan itself in a climate-controlled room, ideally between 18 and 22 degrees Celsius with humidity around 45 to 55 percent, extends its working life considerably. Silica gel packs in the film storage drawers and a small dehumidifier in the scanning room go a long way. The Australian National Film and Sound Archive in Canberra maintains its working scanners under similar conditions, and its published guidance is a useful reference for any home archivist trying to set up a comparable workspace.
Once a collection has been scanned and the original slides or negatives are boxed up for cold storage, the scanner itself becomes less critical, but the digital files need to be checked periodically. Re-reading a sample of the TIFFs every two or three years and confirming that the bit depth and file integrity have not been corrupted by a failing hard drive is the kind of boring, unglamorous task that prevents a thirty-year family archive from quietly disappearing into read errors. A checksum log stored alongside the files makes the verification a five-minute job rather than a weekend.
The next step is straightforward: order a Stouffer TP-21 from a local Australian reseller or directly from the manufacturer, set aside an afternoon, and run the scan described above on your own Coolscan to establish a baseline measurement before the optical path ages any further.