Microfade testing (MFT)

Main Catalogs Page > Research and Analysis > Instrumental Analysis >

Contributors: Kirsten Dunne, J. P. Brown, Vincent Beltran, Cindy Connelly Ryan, Rio Lopez, Catherine H. Stephens (wiki editor)

Overview

Welcome to the Wiki page for Microfade Testing (Microfading), which is being monitored and edited by the AIC Microfading Tester International Discussion group.

This page will cover as much information as possible on Microfading, its history, applications, the different models available, and how to build, run and use them. There is also a resources section at the end with links to many different resources.

Please also see the Library on the International Discussion groups page as this includes a user directory.

• Technique: Microfade Testing, MFT, microfading, microfader
• Formal name: Microfade Testing (MFT), Microfading tester

• Microfading is an accelerated light aging technique carried out on a micro-scale. It provides data and information on the relative light sensitivity of the colorants tested. This data can then be used to make informed decisions on lighting and display of the artwork or object that has been tested.  
  

  Dr. Paul Whitmore, developor of the MFT, sitting in front of his MFT
  The technique was developed in the late 1990s by Conservation Scientist Dr Paul Whitmore. Dr Whitmore wanted to develop Microfading as an accessible and approachable tool and as a method of locating dyes and pigments likely to fade rapidly under gallery lighting conditions to enable collection care professionals to make informed display choices.  
• Microfading was developed as an analytical and a risk management technique and tool. The technique assesses the relative light sensitivity of the colourants of an object using a Microfader, which is a piece of equipment that allows the rapid non-destructive testing of the relative light sensitivity of an object or artwork.
• Testing is carried out in-situ on objects / artworks themselves, as opposed to samples taken from them.
• The MFT can also be used as a research instrument to examine color sensitivity of mock-up samples; these accelerated ageing results can be compared change induced by lower intensity techniques (e.g., light box, gallery lighting) to explore issues of reciprocity.
• Note that ‘MFT’ is used to mean both ‘Microfade testing’ (i.e. the act of Microfading), and ‘Microfade tester’ (i.e. the equipment / the Microfader).
• An MFT system can be broken down into four broad segments:

1. Light Source to provide a stable, high-intensity light
2. Focused light delivery system, from lamp to sample and sample to spectrophotometer
   • This includes the delivery of light to the measuring head, e.g. using a specific fiber optic cable
   • The measuring head (focuses the light from the source and collects a portion of the reflected light from the object surface)
   • The delivery of the light from the measuring head to the spectrometer, e.g. again potentially using a specific fiber optic cable (typically a larger diameter fiber than that used for the incident light).  
3. Spectrophotometer to assess the reflected light and light sensitivity of the sample.¹
4. Computer and software

Schematic diagram of the principle of a Microfading Tester (created by JP Brown) For a detailed description of each of these segments please see Chapter 3. MFT fundamentals: Components, operation and Uncertainty, Vincent Laudato Beltran, Microfading tester, Light Sensitivity Assessment and Role in Lighting Policy, Vincent Laudato Beltran, Christel Pesme, Sarah K. Freeman, Mark Benson, Getty Conservation Institute, 2021, p26-38.

Details

There are currently three different models of MFT and the light sources and geometries can vary between them. Please note, we hope to expand the Wiki in due course to include more information on each model

• The ‘Whitmore’ MFT
  

  Original MFT design, often called the “Whitmore MFT”
• The Thomas Retroreflective model
  

  A Thomas MFT (retroreflective) head. The black cube with the fiber optic cables running to it is the retroreflective head, the black cylinder projecting from the cube hold the collimator and objective, and the skinny silver cylinder is a camera.
• The Fotonowy

MFT designed by commercial company Instytut Fotonowy

Currently there are two types of light source in use with an MFT: xenon or LED.

Example of a Whitmore MFT that uses a Xenon lamp. This is the microfader in use at the National Galleries of Scotland, Edinburgh. Image taken and labelled by Bruce Ford.

Results are gathered by exposing the chosen test area(s) to the high-intensity, stable and focused light spot over a designated period of time (a fade test run, usually around 10 minutes but this can vary). The incident spot size is typically 0.5mm or less, depending on the machine and set-up being used. The incident light intensity is usually between 2 and 6 Mlux².  

The MFT was designed to replicate gallery lighting conditions as closely as possible, within the limitations of the available light sources that had the necessary power and output.  Initially, the xenon arc was used due to its roughly flat Spectral Power Distribution (SPD) across the visible spectrum and its high colour temperature. More recently, LEDs have been introduced as light sources MFT systems. Although their SPD’s can vary widely, LEDs are more commonly used in gallery lighting systems and so, if a suitable LED source is chosen, it is felt that this light will more accurately reflect gallery conditions in the absence of sunlight. Filtered xenon sources are thought to be more representative of sunlight-though-glass illumination conditions which obtain in daylit galleries.

The light source is filtered for UV and IR; it is assumed that under gallery conditions UV will be removed from incident light and filtering of IR means there is no heat incident at the test site. This is for the safety of the object and also because localised heating, in addition to raising the temperature at the test site, can create differences in RH at the test site, both of which can affect fading rates.

The MFT uses light in the visible spectrum, i.e. the range of wavelengths, c. 380 or 400 – 700nm, that the human eye requires to view an object or artwork. We understand that UV is likely to affect the colorants of an object negatively. Microfading is about finding those objects that will be affected by the visible spectrum of light that is required to view them.  Capping the upper range at 700nm seems to have arisen from early MFT use and potentially was limited by the spectrometers being used in the early models. However, the range you should measure depends on the ranges of your spectrometer and light source. Ideally, use as much of the range that you can from 360nm-780 nm (the theoretical range a human can see) without generating excessive ‘noise’ in your spectra.

The test area is assessed throughout the fade test run by the spectrometer and data is gathered on the relative rate of change and how the color behaves during this time. The data is then compared to a reference material, currently the Blue wool standards 1-3 (CIE2000, or BW1-4 in CIE1976). The Blue wool (BW) standards consist of 8 dyed wools of increasing sensitivity, the first 3, BW1-3, cover the fugitivity range of light-sensitive colourants that the MFT can distinguish.

The fade test run also gathers spectral data. This is fundamental information, and it can be used to visualise and evaluate the evolution of the spectra at specific wavelengths during the duration of the MFT test run.

Results are specific to the object tested and are not usually to be generalised; color changes are dependent on so many factors including prior exposure, depth of shade, mordants, binders, over layers, substrates, wider environment. This means you cannot generate results from individual objects or surrogates and then apply them to a wider set of the same type of object, unless you are sure they have the exact same history of display, exposure and materials.

As Dr Whitmore said,  

“The accurate prediction of the fading of different colourant systems is an elusive and perhaps unachievable goal. So many aging processes may contribute to change over time that we must hesitate to accept at face value the apparent forecast of light aging, or any accelerated aging tests. Nevertheless, it is difficult to argue with the notion that exhibition policy guided by some information, however, approximate, is preferable to decisions made in the absence of any information at all.”

Having said that, once the MFT user / community has large enough data sets for specific types of objects, analysis of that data and the results of that analysis can potentially inform existing material guidelines with respect to light sensitivity.  

Samples and Methods

Bruce Ford (formerly of National Museum of Australia) conducting an MFT experiment on artwork while using a Whitmore MFT

MFT represents an advance over previous lightfastness assessments, which relied on the evaluation of representative mock-up samples or direct monitoring of an object colour change during display. The use of mock-up samples to inform the light sensitivity of an object is hindered by time-consuming chemical analysis of the colorants, which is not accessible to all collections. Real-time color monitoring is often laborious and also requires the object to undergo actual changes⁴.

The advantage of Microfading is that you do not need to understand the chemical composition or make-up of an object to gather useful data, nor do you need to understand the full exposure history of the object. The microfader can look at the object as it is now and can rapidly provide data to inform risk-based display decisions.

Microfading is typically a non-contact technique; the ‘measuring head’ is usually about 1cm above the object surface. It is based on testing objects in-situ as opposed to testing samples taken from them.

Contact MFT techniques have been used in the past but are not in wide use at the present. An advantage of contact techniques is that they solve the issue of repeatable focusing; this disadvantage is that the measurement requires contact with the object.

The typical MFT light delivery system uses fiber optic cables and lenses to focus the light into the incident spot.

Whitmore systems are set up with a 0/45 degree geometry (supply at 0° to the surface, reflected light collected at 45° to the surface) which mimics typical color spectroradiometers – the intention of this geometry is to minimize the effects of specular reflection, minimizing the effects of any “glossiness” in materials.

Schematic of light from lamp and reflected lamp going to spectrophotometer. This design is slightly different on a retroreflective head (seen below).

Schematic of retroreflective MFT head

Microfading can be described as non- or micro destructive depending on your perspective. You do expose a small area (0.5mm or less) to an intense light beam which could induce fading during a test run. The risk of inducing a visually perceptible localized change is mitigated by the small area illuminated, and by the monitoring of the color change threshold throughout the test run. If the area being tested is close to reaching this threshold (typically a dE of 5 in CIE1976, or 1 Just Notable Difference / Fade), then the test run is terminated prior to the change becoming perceptible to the viewer.  

Whitmore MFT analyzing a gouache painting

Use of Fotonowy MFT on vertical surface

MFT spot (white circle) on a signature, giving indication of spot size

MFT tests should be run at normal ambient conditions, as defined by the environmental needs of the object being tested.  

It is important to run the MFT in a location free of vibrations, as these can affect the readings. It is useful to work on a solid floor where possible, potentially on a balance bench or similar.

Light sources  

Xenon sources

Suitable xenon sources consist of a replaceable xenon arc bulb in a housing which focuses the light from the bulb onto the tip of a fiber optic cable. The bulb is fragile, and achieving good alignment of the bulb and reflector to focus light on the tip of the fiber optic cable can be fiddly.  

Xenon light sources require time to reach thermal equilibrium and stabilize. This can be anything from 20 mins to a few hours. Monitoring the white balance can help indicate when the light has stabilized, or a radiometer or photometer can be used to monitor light intensity to determine when stabilization has occurred.  It is possible to purchase “compensated” xenon light sources which use an internal sensor to monitor light output and change the regulating voltage to maintain consistent power output. These systems still need significant time to warm up, but once at operating temperature they change SPD only slowly.  

Xenon bulbs deteriorate over time – the filament erodes, and the internal surface of the bulb’s envelope blackens, reducing output and changing the SPD of the lamp. The bulbs require periodic replacement at a cost of around $1,000. Lifetimes of the bulbs are typically given by manufacturers as ~2,000 hours (~80 days) of operation.

LED sources

LED light sources are simpler, cheaper, and generally more robust than xenon sources. They consist of an LED source arranged to be delivered into the tip of a fiber optic cable, and a “driver” which controls the current to the LED.  LED sources require a shorter time to warm up (5-10 minutes) than xenon sources. Changing the current from the driver to the LED will cause a ~30 sec period of instability. LEDs will deteriorate over time, but this is usually at a much slower rate than xenon bulbs – long-term stability depends on the specific LED and driver. Operating some LEDs at lower-than-expected currents may cause a shift in spectral power distribution as well as power. Given the variability of SPDs between different LED lamps, the specific SPD of the lamp being used with the MFT should be recorded with a spectrometer and the intensity should be checked with a radiometer or photometer.⁵  

	Xenon Arc lamp	LED bulb
Advantages	Broadband spectral power distribution represents the worst-case scenario. Potentially higher power output than LED sources. Wide wavelength range the possibility of investigating UV fading. Decent stability.	Small and robust, easy to transport. Rapid warm-up. No reflector alignment required. Inexpensive (compared to xenon sources). Notched LED output is more representative of LED-lit galleries. Can be very stable, but depends on specific LED and driver.
Disadvantages	Expensive. Fragile bulb, fiddly to align. Broadband output requires careful filtering. Long warm-up times. Stability can be an issue. Large size is inconvenient for transport.	Generally lower power output than xenon sources. Notched SPD does not represent the worst-case scenario.

Sample sites

Ideally, you would assess the various colors on any object that you wish to test. The target is often to test all colorants observed and identify the most fugitive, however, you can take a variety of risk and significance-based approaches depending on the object, context and available time / resources.

Choice of test sites can be made through a simple visual assessment of the object under good light. Based on your knowledge of the object, you can choose to assume that similar colors on an object have the same chemical composition, or not⁶. However, this approach can be subjective.  

If accessible, scientific testing and analysis of the object’s media can also inform test sites for MFT. However, whilst advantages in helping you prioritize, this is not essential and one of the benefits of Microfading is that you don’t have to understand or know all the media to gain useful information and results.

When choosing sample sites to test, considerations include,

• Prioritization of object / test sites – by potential or perceived risk, condition, likely previous exposure, significance, material knowledge

• Scheduling and other resources needed, e.g. logistics of getting the object to the MFT or vice versa, assistance unframing etc.

• Test site selection – number and location of sites

• Any handling concerns or issues – if a support moves during a fade test this can affect the validity of the data, supports may need to be local weighted or supported from behind for stability.  

Data Collection and Analysis

Before beginning any fade test runs / data collection, on a daily basis the spectrometer should be calibrated. This is done for background (dark) spectrum, which should be ambient conditions (the 0% reflectance signal) and also with a white balance (the 100% reflectance signal).

• Microfading provides data on the relative light sensitivity of a colorant. The relative scale most used is the Blue wool standards 1-3 (or 1-4 if using CIE1976). Often an MFT user starts testing by running these blue wool standards. This can also be a good indication that lamp intensity is correct, or if a Xenon bulb is deteriorating as you will notice a change in the readings obtained for the blue wools. However, please also see the section later in this Wiki, in ‘Considerations, The reference scale being used’ for further notes on Blue Wools. Please note that recording daily conditions of the instrument, including intensity, wattage and integration time, can also help with diagnosing issues if they arise.   For each test run, the user specifies frequency and duration of data collection. These can vary between users, equipment and the purpose of the test. Frequency of data collection can range from every few seconds up to every minute, every 30 to 60 seconds is the most common. Test run duration can vary from 5 to 30 minutes. Durations at the lower end are more typical and time efficient, e.g. 5 to 10 mins. Please note that test run duration can be overridden and the test can be terminated if a Just noticeable difference / the predefined DeltaE / ΔE* threshold is being approached.⁷   Once set up and calibration are achieved, it is a case of choosing test sites and then,  
• positioning the beam on the chosen test site,  
• recording and image of that test site. Some users also annotate a photograph of the object being tested to create a map of test sites,  
• maximising the reflectance, aka focusing the incident beam on the test site
• starting the test run
• monitoring the test run and stopping the test if the predefined DeltaE threshold or duration is reached (typically ΔE*_ab of 5 or ΔE*₀₀ of 3).

At the end of a test run, the data is recorded and saved and can be analysed to provide information on rates of change and to identify the most light sensitive colorant tested. This is usually what then determines the recommended light budget for that artwork.   Data analysis can be done in Excel (using macros helps), or via the GCI’s Spectral Viewer. The automated MFT uses its own custom software that combines both hardware control and data analysis. Data can also be visualised into relevant graphs, such as color difference curves.   The data can also be used as the basis for any reporting that is required. Examples of the type of data and how it can be visualised and represented can be seen below, although this can vary depending on the model and software that you use.

Example of how relative color change rates can be presented, created within Excel.

Example of a graph of colour change curves, CIEDE2000, created within Excel.

MFT color difference curves for different samples as a function of light dose

Spectra collected during MFT run on one sample, can see evolution of curve indicating color change (green is initial, red is final); can also add up/down arrows to indicate direction of change, taken from the Getty Spectral Viewer software.

Spectra collected during MFT with representative colors corresponding to wavelength ranges are shown on plot, taken from the Spectral viewer software.

a*b* diagram, Visible wavelength start and end spectra and photomicrograph example for a test site. Please note white dot on the photomicrograph is the incident beam at less than 10% illumination to allow the exact test site to be recorded.

Spectra collected during MFT with representative colors corresponding to wavelength ranges are shown on plot, taken from the Spectral viewer software.

Considerations & Limitations

Microfading is a reflectance technique and so there can be some limitations, or where the quality of the data obtained can be affected, which should be considered when choosing which objects to test and your test sites.

Surfaces

• Glossy surfaces can cause the data to ‘bounce’, this can be seen as unexpected undulations in the colour fade curves.  

• Transparent and translucent materials can be difficult to test, as they generally do not provide a diffuse reflection.  

• Feathers and fur can be difficult as the nature / scale of the material means that the surface the incident beam is interacting with can be smaller than the diameter of the incident beam. However, there are workarounds, such as pressing the test area under glass. If using glass, remember that reflection from the sample surface will be reduced which means that white standard needs to be obtained through the same glass. Some users choose to re-run their blue wools also under glass slides for more accurate calibration.  

• Fabrics with a nap, e.g. velvet, can have different readings depending on direction of the fibres. The solution can be to take readings with the nap in each direction and produce an average or, again, to press the sample under glass.

• If your surface / substrate moves during a test run, it will impact the readings and will throws errors into the data (this is because it moves in and out of focus when the substrate moves). This can often look like a full spectra move / drop where the amount of change is consistent across the spectra. Where possible and where the object condition allows, it is helpful to locally support or weight an object during testing. This error can also occur if the measuring head moves during a test run.  

Media types

• Pigment particle size and density will influence apparent light sensitivity - lighter washes of the same color of media application on a watercolor, for example, can display smaller rates of change than a mid-density particle distribution of the same paint; ideally always test the mid tone where possible. Dark colors and blacks are difficult to measure accurately because there is little diffuse reflection from these colors.

• MFT results are only a partial predictor of the long-term behavior of chemically unstable materials such as herbarium specimens, color photographs, and pulp paper. These materials are chemically unstable and change color over time, even in dark storage. For example, MFT usually captures only the photo-bleaching of paper under UV-free conditions. However, room-temperature oxidative reactions lead to yellowing, as do reactions initiated by light during exposure but continue during subsequent dark storage (post-actinic processes). The relative rates of the process that cause yellowing depend heavily on factors such as chemical bleaching, sizing agents, lignin content, pH, exogenous and endogenous pollutant levels (including inks), temperature, prior conservation treatments, and so on.

• We are rarely dealing with a single dye or pigment when undertaking a fade run test, a test area will be a combination of the materials with different inherent fading rates. This is one of the reasons that MFT results cannot be generalised and are specific to object being tested.

• Darker media will have a low overall reflectance and thus an inherently worse signal to noise ratio, and so it can be challenging to get satisfactory data / spectra from.  

• For fluorescent materials the light in part has an emitted component, and this is not accounted for in the CIE calculations and therefore this renders the colour difference calculations (dE) suspect. Examples include UV-fluorescent pigments, and paper or textiles containing optical brighteners.  

• It was noted by Jim Druzik, whilst working at The Getty, that as the light sensitivity of an object increases, predictability becomes more problematic. Due to the chemistry involved, this can result in you assuming higher sensitivity than there is, e.g. in particular colourants used in hand tinting late 19^th C photographs. [need to find reference]

• Reversion: within the context of color testing, reversion is the tendency of the color to change back to its original state after the light source is switched off. Examples of where this effect has been seen include Prussian Blue and iron gall ink. In the case of Prussian blue, very rapid fading can be experienced under the MFT, but once the incident light is removed, the fade reverts. There are ways to test for this by switching the light source off, leaving the head in place, leaving for 15-30 minutes and taking another reading. If there is no reversion, the deltaE value should be very similar to the deltaE value at the end of the initial MFT run, if there is reversion, the deltaE value should be substantially lower than the last reading of the run.  

Reciprocity

As with any accelerated technique, does what happens at a high light level over a short duration match what occurs in real time?  

See Barro, L., Sanderson, K., Centeno, S. A., & Saunders, B. (2020). The Exhibition and Characterization of Seven Salted Paper Prints. Journal of the American Institute for Conservation, 59(3–4), 171–185. https://doi.org/10.1080/01971360.2019.1696914 for an example of this.

For a description and discussion of uncertainty in MFT including reciprocity, blue wools, spectral power distribution, color change behaviours and varying predictions, please see Chapter 3. MFT fundamentals: Components, operation and Uncertainty, Vincent Laudato Beltran, Microfading tester, Light Sensitivity Assessment and Role in Lighting Policy, Vincent Laudato Beltran, Christel Pesme, Sarah K. Freeman, Mark Benson, Getty Conservation Institute, 2021, p26-38.

The reference scale being used

Concerns have been raised as to whether blue wools being fabrics cause an issue. Also, some people have observed some differences between BW batches and the microfader can struggle to differentiate BW 3 and 4 when working in CIE2000, which is often why BWs 1-3 only are used. There are also some suggestions that BW4 requires UV to act appropriately and in general, UV is removed from the MFT incident light beam.  

Relative humidity can also affect readings for blue wools; ideally the environment in which you are testing should be the same as the display conditions. Small differences in RH and temperature shouldn’t make a difference, but larger variations might do, e.g. in the range of +/- 5C or c.20%RH. The effect is likely greater for the more fugitive colors.  

SPD, spectral power distribution, of the light source can also impact readings for the blue wools (and all areas being tested), because of wavelength specific fading.  

For a discussion of ISO Blue Wool Fading Standards and their use in microfade testing see: Ford, B. and C Korenberg, 2023. Manufacturing Variations in ISO Blue Wool Fading Standards under Microfading Exposure Conditions. Studies in Conservation, DOI: 10.1080/00393630.2023.2184555 pp.1-12

Colorimetry/Color Space/CIE

• • Color space is an approximation.  
  • Just Notable Fades / Differences are also contentious, as they are subjective and based on an individual viewers ability to detect them. This will vary from person to person depending on lighting, context, object size, texture, pattern, colour, surrounding colour, age of the viewer, their eyesight.  
  • Color difference calculations can’t be compared between the different CIE calculations, e.g. between CIE76 and CIEDE2000. It needs to be clear what color difference measure you are using. You should be using CIEDE2000, or the most recent CIE version. If you are using CIE76, you are looking at geometric distance in CIELAB colour space. CIEDE2000 more closely approximates human visual perception of colour difference. In particular, CIE76 provides a poor representation of human perception of color difference in blues and greens, whereas CIEDE2000 more closely approximates the actual differences seen. This issue with CIE76 is particularly a problem when you are comparing colour differences in blue wool standards vs colors elsewhere in CIELAB color space.  
  • The magnitude of a color difference (dE) is only meaningful over a restricted range of values, in the order of around < or = 5. For example, a dE of 100 is meaningless, whereas a dE of 5 is meaningful. ASTM Standard D2244-05, section 5.2 discusses this. It can be an issue when trying to extrapolate highly fugitive media results. For a summary of color science in relation to Microfading, please see Chapter 2. Color Science in the Context of MFT, Sarah K. Freeman, Microfading tester, Light Sensitivity Assessment and Role in Lighting Policy, Vincent Laudato Beltran, Christel Pesme, Sarah K. Freeman, Mark Benson, Getty Conservation Institute, 2021,p12-25. https://www.getty.edu/conservation/publications_resources/pdf_publications/microfading_tester.html  

Repeatability

Concerns have been raised about repeatability of BW runs, for example, to a degree a scientist would recognise.  

We are not dealing with homogenous media in most cases and variations within a brushstroke / fibre / area of the same media application can lead to variations in results.  

However, it should be remembered that this is a risk management technique and there is little evidence of the degree of error between blue wools meaning that a BW 2 or 1 could be mistaken for each other.

Light sources

• The spectral power distribution of the Xenon lamp is not the same as that used in many gallery spaces, and it contains a higher proportion of the shorter wavelengths.

• A benefit of the Xenon arc lamp is its relatively flat PSD in the visible region, with the hope that this would overlap with the most reactive wavelengths for all light sensitive material. In contrast, the troughs present in LED PSDs may coincide with the most reactive wavelengths for a given material and not identify these materials as light sensitive. This obviously needs to be investigated more systematically.

Many of these issues can be managed by remembering that this is a risk management tool, not an analytical one.

Applications

Microfading data can be gathered on many different material and media types.  

When the results and data are connected with a related gallery / museum lighting policy, which includes recommended light budgets, display recommendations can be made which can both support preservation and collection access.

Recommended visible light levels and quota for artworks on display at the National Gallery of Scotland

Example of the light budget approach taken by the National Galleries of Scotland from 2012. This is based on the approach and table developed by Bruce Ford and Nicki Smith for the National Museum of Australia [insert reference link] and is based on a 500-year acceptable lifespan.

Please note, a more detailed section on acceptable lifespans, Lighting policies and frameworks and the role of Microfading within them is planned for this Wiki. If you have any questions, please see the relevant discussion threads on the MFT-IDG community page, [insert link to specific or just the discussion page? Specific link is currently : Lighting Policy & MFT | Microfading Tester International Discussion Group]

Please also see Chapter 5 of Microfading tester, Light Sensitivity Assessment and Role in Lighting Policy, Vincent Laudato Beltran, Christel Pesme, Sarah K. Freeman, Mark Benson, Getty Conservation Institute, 2021, p60-85.

Budgetary Considerations

The main costs are,

• Initial purchase of the equipment  
• Annual cost of bulbs or parts due to wear and tear (where needed)
• Intermittent upgrading of computer system, if needed
• Staff time and capacity
• Studio space for the equipment

Equipment cost varies between the different types.

The Whitmore / original design is bought in parts and built by the user. The cost of the equipment is between USD $9,300 – $27,000 depending on if you have an LED or Xenon light option. This includes all hardware and computer set up. Please see https://www.microfading.com/equipment.html for more information and current costings.

Also, please see the MFT-IDG discussion thread on "Whitmore MFT: 2021 Parts Lists" lists cost at USD $27,000 (Xenon). Bruce's site (https://www.microfading.com/equipment.html) currently lists the LED version at USD $16,000 and the Xenon version at USD 26,000; both include training

Cost of the GCI Fotonowy (2020) was $36,000; this included 2 6-LED changers ($3100 each) and 1 day of installation/training (USD $3000).

The Retroreflective model can vary depending on which you build, but is in the region of $9000 (not including the optical bench).

After the equipment has been purchased, cost implications are staff time and capacity.  

There are typically no sample preparation costs as you are testing the objects / artworks themselves.

Depending on your set up, you may need a small amount of annual budget for spare / replacement lamps etc. This could be in the region of around US$800 (but this may not be required every year, depending on usage levels).

Case Studies

• Mae West lips sofa – microfading • V&A Blog (vam.ac.uk)

Please also see Appendix 4 in GCI's MFT Guidelines. "Screening Material for Exhibition": https://www.getty.edu/conservation/publications_resources/pdf_publications/microfading_tester.html

and 2010 AIC RATS Postprints: https://www.culturalheritage.org/membership/groups-and-networks/research-and-technical-studies/resources/publications

Suggestions of types of case studies below,

• Individual artworks with interesting outcomes or display solutions
• How you are using an MFT within your organisation
• How you are applying microfader data to collection management
• Lighting policies and integrating MFT approaches and data
• Examples of issues and limitations

Additional Information

The Light Damage Calculator (LDC) Light Damage Calculator - Canada.ca from the Canadian Conservation Institute can serve as a complementary tool to MFT for visualising predicated change. The LDC provides a means of examining the fading of a single colorant, a single colorant under different scenarios and a collection of colored objects.⁸ Please note that the current LDC versions are on the HERIe platform (https://herie.pl/), due to restrictions in CCI's web platform. Two versions are available: the original based on modeled data by Stefan Michalski, and a second based on collected data by Eric Hagan A comparison of MFT data to real time color monitoring will always be helpful.  

This Wiki is under development. The topics below are planned for future addition and expansion; Principles of Microfading including more details on:

• Filtration of the light source
• Head
• Detector
• Optical paths
• Ray optics
• Blue Wools
• Color space
• Color difference and CIE L*a*b* color space - CIE76 vs CIE96 vs CIE2000
• Colour chemistry JNF / JNDs
• Risks associated with using this technique
  

More detailed information on the different MFT models, including for each,

• History
• Set up i.e. the component parts and what they tend to look like
• How to build / source
• Where to work e.g. space preparation / space needed, portability
• Operation
• Calibration
• Budget required
  • Approximate cost to purchase equipment for this technique
  • Annual cost to maintain or run
  • Staff time required
• Time / duration of tests
• Test site size
  More details on approaches to testing
• Approaches
• Choosing objects
• Choosing test sites
• Destructive vs non-destructive
• Test run times
• Object preparation and needs
• Risks and how managed
  Interpreting results
  Reporting; what to include
  

Applications to Collections Management and Lighting Policy

• Lighting policy
• Acceptable lifespan
• Light budgets

References, Resources, Databases, Publications

• Barro, L., Sanderson, K., Centeno, S. A., & Saunders, B. (2020). The Exhibition and Characterization of Seven Salted Paper Prints. Journal of the American Institute for Conservation, 59(3–4), 171–185.
  mft_guidelines_references - lists all references from the GCI's MFT Guidelines publication
• Presentations from conference on anoxia and microfading
• Predicting the Long-Term Light Stability of Color Photographic Prints: Comparing Macro and Micro-fade Testing Results
• microfading.com
• In situ Accelerated Light Ageing with Portable Microfade Spectrometry - Nottingham Trent University
• Presentations from conference on anoxia and microfading - Conservation DistList (culturalheritage.org)
• A quick guide to microfading: How, what and why? - The National Archives blog
• Advancing Microfading Tester Practice - Getty Conservation Institute (GCI)
• Microfading Tester: Light Sensitivity Assessment and Role in Lighting Policy - GCI

Back to AIC Wiki Main Page

Back to Research and Analysis page