The International Organization for Standardization, or ISO (an acronym derived from the French version of the name), is a nongovernmental organization composed of members from the national standards bodies of 167 countries. (Here in the U.S., our national standards body is the American National Standards Institute, or ANSI.) ISO's headquarters are in Geneva, Switzerland. Through its numerous technical committees, subcommittees, and working groups, ISO coordinates the efforts of some 45,000 subject-matter experts worldwide to develop international standards for a wide range of activities, including the manufacture of products, management of processes, and delivery of services. As ISO puts it, a standard "describes the best way of doing something." Depending on the activity, "best" can mean "most efficient," "most cost-effective," "most sustainable," or some other appropriate superlative.
When it comes to observing the Sun — as when watching the partial phases of a solar eclipse — the best way is one that provides a satisfying view while protecting your eyes from injury due to the intense radiation pouring from our daytime star. The relevant international standard for this activity is ISO 12312-2:2015, "Eye and face protection — Sunglasses and related eyewear — Part 2: Filters for direct observation of the Sun." Adopted in 2015 and affirmed in 2020, ISO 12312-2 applies worldwide and supersedes a hodgepodge of earlier national and regional standards.
ISO standards are not laws, though governments sometimes base laws on ISO standards. This means it is not necessarily illegal for a product to fail to meet the requirements of an ISO standard. But it may be inadvisable to buy or use such a product, as failure to meet the requirements of an applicable standard likely means the product is not well suited to its purpose. ISO 12312-2 specifies the properties that a solar viewer should have in order to protect your eyes from injury. For your safety, the AAS Solar Eclipse Task Force recommends that you avoid any solar viewer that doesn't meet the requirements of ISO 12312-2, with changes or exceptions as described later in this article.
On ISO's website, you can purchase a PDF of the ISO 12312-2 standard for 61 Swiss francs (CHF), or about $65 (USD). Much of the explanatory material included in that PDF is also available in the technical report "Solar Eclipse Eye Safety" by a principal author of ISO 12312-2; you can grab that document for free from our Downloads page. Below we describe some of ISO 12312-2's key provisions and explain how the AAS Solar Eclipse Task Force applies the standard when considering whether to add a company to our Suppliers of Safe Solar Fiilters & Viewers page, which we do at no charge.
Applicability of ISO 12312-2
ISO 12312-2 applies only to certain products intended for direct viewing of the everyday Sun, which astronomers call the photosphere and which you can think of as the Sun's bright face. Specifically, the standard applies only to nonmagnifying ("unit power," or 1x), nonfocusing products meant to be used with your eyes but with no optics other than corrective eyeglasses or contact lenses. Furthermore, it applies only to products that can be handheld or worn like (or over) eyeglasses.
ISO 12312-2 does not apply to ordinary sunglasses for general use, which pass thousands of times more light than is safe for looking directly at the Sun; there's a separate standard (ISO 12312-1) for those.
ISO 12312-2 also does not apply to solar filters meant to fit over the aperture (front) of optical devices such as camera lenses, binoculars, or telescopes. But any solar filter made of material that meets the transmittance, uniformity, and quality requirements (see the next section) of ISO 12312-2 should work as an aperture filter too. For more on this, see the last section of this page.
ISO 12312-2 Requirements
ISO 12312-2 sets requirements on the following properties of a safe solar viewer: transmittance (the ratio of transmitted light to incident light) at the ultraviolet, visible, and infrared wavelengths that reach our retinas; uniformity of transmittance; material and surface quality; mounting; and labeling.
Here are the standard's transmittance limits at wavelengths relevant to eye safety (that is, at wavelengths that reach the retina in a normal eye); these come from section 4.1.1:
Luminous transmittance, τv, is the transmittance across the visible spectrum. The standard specifies that the UVA and UVB transmittance be no higher than the measured luminous transmittance, whereas the IR transmittance limit, 3%, is higher. That's because IR radiation is less effective at causing a retinal injury than UV or visible light, as explained in the technical report on solar eclipse eye safety mentioned above.
The standard's maximum allowable luminous transmittance corresponds to that of a shade 12 welding filter, while the minimum corresponds to that of a shade 15 welding filter. Before the advent of cardboard "eclipse glasses" and handheld solar viewers, astronomers routinely used welding filters to view the partial phases of solar eclipses without injuring their eyes. A shade 12 filter is safe, but many observers find the solar image uncomfortably bright. For that reason, shade 13 and 14 filters have been more popular as eclipse viewers, though some observers feel that the solar image in a shade 14 or darker welding filter is too dim.
In a technical article published in the Astronomical Journal in September 2021, Ralph Chou (University of Waterloo, Canada), Stephen Dain (University of New South Wales, Australia), and Rick Fienberg (AAS Solar Eclipse Task Force) evaluated a large sample of commercially available solar viewers collected over several years leading up to the August 2017 "Great American" total solar eclipse. What they found was illuminating (pun intended)!
First and foremost, Chou et al. recommend that the maximum and minimum luminous transmittance in section 4.1.1 of the ISO 12312-2 standard be lowered to 0.0012% and 0.00004%, respectively. It turns out that limits based on welding filters don't translate perfectly to modern purpose-made solar viewers. The latter typically show the Sun as yellow-orange or white, whereas the image in a welding filter is green. Because of our eyes' different sensitivities to different wavelengths, the "comfort zone" for solar viewers shifts slightly to lower transmittances compared with welding filters. Slightly lighter filters (still below the current standard's maximum of 0.0032%) won't hurt you, but for most people they'll produce an image of the Sun that's annoyingly bright. From our perspective on the AAS Solar Eclipse Task Force, any solar filter or viewer shown by a properly accredited test lab (see below) to have a luminous transmittance between 0.00004% and 0.0032% is acceptable; this range goes from the proposed new, lower minimum to the current maximum luminous transmittance of ISO 12312-2 section 4.1.1.
For reference, a filter with a transmittance of 0.001% passes 1 part in 105 (100,000) of the incident light and is said to have an optical density (OD) of 5; that's typical of many solar filters produced for camera lenses, binoculars, or telescopes. Sometimes you'll see such filters described as ND ("neutral density") 5. Chou et al.'s proposed maximum luminous transmittance for solar viewers of 0.0012% corresponds to approximately 1 part in 83,000, an OD of 4.9, or shade 12.5. The corresponding numbers for the proposed new minimum of 0.00004% are 1 part in 2.5 million, an OD of 6.4, or shade 15.9. That's pretty dark, but it still enables a satisfying view of the bright Sun.
Uniformity of Transmittance (§4.1.2)
ISO 12312-2 section 4.1.2 specifies that across a given filter, the luminous transmittance should be uniform to within 10%. As long as no area of the filter exceeds the maximum or falls short of the minimum permitted transmittance, this degree of uniformity seems unnecessarily strict. Tests show that many of the filters that don't meet the requirements of section 4.1.2 because of nonuniformity in transmittance are nevertheless uniform to better than 30% and provide satisfying views of the Sun. Thus we've concluded that relaxing the uniformity requirement to 30% is reasonable; we will propose such a change for the next iteration. In the meantime, the AAS Solar Eclipse Task Force considers acceptable any solar filter or viewer shown by a properly accredited test lab to be uniform in luminous transmittance to no worse than 30%.
Material & Surface Quality (§4.2.1)
ISO 12312-2 section 4.2.1 includes these requirements: "Filters shall be free from defects likely to impair vision in use, such as bubbles, scratches, inclusions, dull spots, pitting, scouring, pocking, scaling, and undulations. Metal coated filter materials shall not exhibit more than one pinhole defect not greater than 200 μm in average diameter within any 5 mm diameter circular zone." Manufacturers seem to have no trouble producing filters that meet these requirements, so we insist that acceptable eclipse glasses and other solar viewers meet the current requirements of ISO 12312-2 section 4.2.1.
Section 4.3.1 of the standard says, "Mountings may be handheld or shaped in the form of spectacles to be worn on the face in front of any corrective (spectacle or contact) lenses worn by the user." This is the wording that renders ISO 12312-2 inapplicable to solar filters mounted in cells meant to fit over the apertures of camera lenses, binoculars, or telescopes. But the same section also says, "Filters may be made with or without a mounting. If mounted, a filter shall be held securely so that it cannot be dislodged by normal handling or by gusts of wind." That sure sounds like it applies to aperture filters and sheets of solar-filter material that amateurs routinely use to make their own aperture filters! We'll attempt to resolve this ambiguity in the next iteration of the standard.
Section 4.3.2 specifies that a solar viewer must be big enough to cover both eyes of an adult and that for eclipse glasses, the nasal cutout must fit a typical adult's nose. But makers of solar viewers keep coming up with variants that weren't anticipated when the standard was written, including child-size eclipse glasses. Chou et al. recommend that the standard be revised to accommodate a wider variety of formats. Accordingly, the AAS Solar Eclipse Task Force isn't too concerned about filter dimensions as long as they seem appropriate to the intended user/wearer.
We do, however, insist that mounted solar viewers (that is, filters in handheld viewers and eclipse glasses) meet the requirements of ISO 12312-2 section 4.3.3 on material quality: "The filter and mounting shall be free from roughness, sharp edges, projections, or other defects which could cause discomfort or injury during use. No part of the filter or mounting which is in contact with the wearer shall be made of materials which are known to cause any skin irritation."
Labeling (ISO 12312-2 §5a-f)
Section 5 of the standard includes the following provisions [with our annotations in square brackets]:
The filter and/or its packaging shall show the following information in the language(s) of the country where the product is to be offered for sale:
a) name and address of manufacturer of the product [where "manufacturer" isn't defined but is generally assumed to be the company that produces or imports the finished product, whether or not that same company makes the solar-filter material used in the product];
b) instructions for use in looking at the Sun or a solar eclipse [which we recommend should be based on the material on our Eye Safety page];
c) warnings that viewing the Sun without an appropriate filter can result in permanent eye injury [except during the total phase of a total solar eclipse, as explained on our Eye Safety page];
d) warnings that filters that are damaged [for example, scratched, punctured, or torn] or separated from their mountings should be discarded;
e) advice on storage, cleaning, and maintenance, as appropriate [see "Is It Safe to Clean Eclipse Glasses and Handheld Solar Viewers?" on our Eye Safety page];
f) obsolescence deadline or period of obsolescence, as appropriate [as noted on our Eye Safety page, modern solar viewers last indefinitely as long as they're stored and maintained in good condition; Chou et al. recommend that this requirement be omitted from the next iteration of ISO 12312-2, and the AAS Solar Eclipse Task Force is ignoring section 5f when examining test reports on eclipse glasses and other solar viewers].
In the next iteration of the standard we're proposing that the labeling or packaging also include a warning that children should use solar viewers only under adult supervision.
ISO creates standards, but it doesn't test products for conformity with those standards. Thus no product or package should suggest that a solar viewer is "ISO certified," "ISO compliant," "ISO registered," or "ISO approved."
Assessment of conformity with ISO 12312-2 is done by laboratories that have the appropriate tools to test solar viewers and whose staff are trained in the appropriate methods for the use of those tools. ISO 12312-2 identifies appropriate tools for testing solar viewers, including, for example, spectrophotometers and microscopes, as well as appropriate test methods, namely, the ones formerly described in ISO 12311:2013, "Personal protective equipment — Test methods for sunglasses and related eyewear" and now incorporated into ISO 18526-2/3/4:2020 "Eye and face protection — Test methods — Part 2: Physical optical properties/Part 3: Physical and mechanical properties/Part 4: Headforms."
How can you tell if a lab is qualified to test products against the requirements of the ISO 12312-2 standard? It will be accredited as such by an organization that is a full member/signatory of the International Laboratory Accreditation Cooperation (ILAC) Mutual Recognition Arrangement (MRA). This means three things:
- The organization maintains conformance with ISO/IEC 17011, a joint standard of ISO and the International Electrotechnical Commission (IEC) that specifies requirements for the competence, consistent operation, and impartiality of accreditation bodies.
- All testing and calibration laboratories to which the organization grants accreditation conform with ISO/IEC 17025, which specifies the general requirements for the competence, impartiality, and consistent operation of laboratories.
- The organization has been peer-reviewed and shown to meet ILAC’s criteria for competence.
Accreditation isn't one-size-fits-all. Labs are accredited to test against the requirements of specific standards. So, for example, a lab that is accredited to test cars and trucks for conformity with ISO 13216: "Road vehicles — Anchorages in vehicles and attachments to anchorages for child restraint systems" may or may not also be accredited to test solar viewers for conformity with ISO 12312-2.
On a related note, as explained on ISO's Name and Logo page, no version of the ISO logo should appear on any product. If a solar viewer has an ISO logo on it, you may conclude only that the manufacturer is unaware of (or ignoring) ISO's policy. It is OK to indicate that a solar viewer meets the requirements of ISO 12312-2, but only if that has been validated through tests performed by a properly accredited laboratory.
How Vendors Get on Our List of Suppliers of Safe Solar Viewers/Filters
We recommend that you obtain solar viewers and/or filters from the sources listed on our Suppliers of Safe Solar Filters & Viewers page. Suppliers don't pay us to get onto this list; they demonstrate to our satisfaction that their products are safe.
Eclipse Glasses & Handheld Viewers
Listed suppliers of eclipse glasses and handheld solar viewers have demonstrated to the satisfaction of the AAS Solar Eclipse Task Force that their products are safe, either by providing us with a test report from an accredited laboratory showing that their products meet the requirements of the ISO 12312-2 standard (with changes or exceptions as described above) or by getting their products from one of the other suppliers on the list. In the first case, we have verified that the testing laboratory is accredited to test for conformity with ISO 12312-2 by an accreditation body that belongs to the ILAC MRA as described above. In the second case, where one supplier tells us that it gets its viewers from another supplier, we verify the relationship by contacting that other supplier.
As noted above, solar filters meant for use with camera lenses, binoculars, or telescopes are not covered by ISO 12312-2; in fact, there is at present no international standard for such filters. Nevertheless, as also noted above, any filter material that meets the transmittance, uniformity, and quality requirements of ISO 12312-2 (with changes or exceptions as described above) should also be safe to use with optics as long as the user follows the instructions on our Solar Filters for Optics: Telescopes, Binoculars & Cameras page. For many of the suppliers on our list, AAS Solar Eclipse Task Force members and other astronomers have extensive experience safely using their products to observe the uneclipsed or partially eclipsed Sun. Suppliers with whom we have no experience convince us that their products are safe by demonstrating conformity with ISO 12312-2 sections 4.1.1, 4.1.2, and 4.2.1 (with changes or exceptions as described above) or by telling us that they get their filter material from one of the suppliers already on our list, in which case we verify the relationship by contacting that supplier.