As reported in April by LEDs' Carrie Meadows, the first standard from the Illuminating Engineering Society and International Ultraviolet Association partnership has been released. LM-92-22 establishes test methods for UV LEDs with peak wavelengths of 200 to 400 nm. This standard will increase the accuracy and consistency of UV LED measurements, ultimately improving germicidal UV (GUV) light sources.
Although this standard and the IES/IUVA partnership that created it may appear to be a reaction to the COVID-19 pandemic, the efforts that led to LM-92 in fact began in 2015, when an ad hoc group of IUVA and IES scientists and UV practitioners met to discuss deficiencies in regulations and standards governing GUV sources.
To improve the situation, the group devised a plan to form several partnerships between IUVA and standards developing organizations (SDOs). These partnerships were intended to quickly generate new standards for GUV source testing, efficacy, and labeling, which would enhance the credibility and understanding of GUV disinfection technologies.
This article details some of LM-92’s innovative measurement techniques, along with the UV LED testing challenges they address. It also discusses what to expect next from the IUVA/SDO partnerships.
U.S. treats GUV sources as pesticides
When the U.S. Congress passed the Federal Insecticide Fungicide and Rodenticide Act (FIFRA) in 1947, chemical pesticides were heavily used in agriculture. The aim was to protect agricultural workers, who were being sickened by improperly formulated and incorrectly applied pesticides. FIFRA was administered by the U.S. Department of Agriculture.
In 1970, the Environmental Protection Agency took over enforcement of FIFRA regulations. EPA updated FIFRA regulations in 1972; however, GUV sources, which had been in use since the early 1900s, were treated as pest control “devices,” defined as “any instrument or contrivance (other than a firearm) which is intended for trapping, destroying, repelling, or mitigating any pest.”1 This classification placed GUV sources in a category with ultrasonic rodent repellants, fly tape, and other mechanical pest deterrents.
Perhaps because the effectiveness of mechanical deterrents is often apparent to the end user (the mouse is caught or it isn’t), FIFRA’s regulations for GUV “devices” are quite limited. According to EPA’s 2020 Compliance Advisory, which clarified the FIFRA regulations that apply to UV sources: “Unlike chemical pesticides, EPA does not routinely review the safety or efficacy of UV light devices2.” The lack of regulations means reduced development and production costs for GUV manufacturers. But because the effectiveness of GUV products is often not immediately apparent to end users, it also allows ineffective products onto the marketplace alongside effective products.
The need for better standards
In 2016, some UV practitioners approached IUVA about the effects of EPA’s FIFRA regulations, as well as the promise of GUV to reduce healthcare-associated infections (HAIs) — a critical issue that had resulted in 72,000 deaths in 20153.
These practitioners knew that better standards had the potential to speed market acceptance of effective GUV products. Several stakeholders had already tried appealing directly to Congress to revise FIFRA, but after a total of 25 letters — and more than one congressional meeting — the consistent congressional response was to let the marketplace work it out.
Congress’s reluctance was ostensibly based upon White House Office of Management and Budget memo Circular A119, first released in 1993 and last updated in 2016. Circular A119 directs agencies to “use voluntary consensus standards in lieu of government-unique standards except where inconsistent with law or otherwise impractical4.”
Rebuffed by Congress, UV practitioners presented their concerns at a panel discussion of industry and government experts at the IUVA Americas event in early 2016. Key National Institute of Standards and Technology and IUVA scientists then met again in February 2018 in to discuss how to improve standards. After a two-day conference on GUV later that year at Yale School of Medicine, IUVA’s Healthcare/UV Working Group was formed.
IUVA tags IES to create optical measurement standards
Shortcomings with contemporary GUV lamp-measurement protocols — mostly “best practice” procedures for low-pressure mercury, pulsed xenon, and deuterium lamps that were obscurely documented in back issues of IUVA newsletters — quickly became apparent. These protocols sometimes did not obtain the correct radiant flux measurement.
NIST’s Optical Radiation Group leader Cameron Miller, along with Ajit Jillavenkatesa, formerly NIST's senior policy advisor, delved into this issue and mapped a path to better UV standards in a 2018 IUVA newsletter article5. They envisioned optical measurement standards covering radiant flux, radiant intensity distribution, and irradiance.
Miller and Jillavenkatesa also came up with a plan to develop these standards using existing SDOs. This eliminated the need for IUVA to become accredited as a standards organization — a process that could take years.
As author of the widely used visible LED testing standards LM-80 and LM-85, IES was selected as the ideal SDO for the optical measurement standards. IUVA and IES signed a memorandum of understanding (MOU) on May 13, 2020.
IES developed LM-92 during COVID lockdown
IES and IUVA scientists and engineers met weekly to generate the first measurement standard covering UV LEDs. While IES had been publishing standards for visible LEDs for more than a decade, UV LEDs presented some unique challenges.
First, UV light — especially short-wavelength UV — is tough on sphere coatings. Barium sulfate with a binding material, the most popular integrating sphere coating, degrades quickly when exposed to UV light. Dust and other contamination also can fluoresce when exposed to UV, distorting measurements. LM-92 includes advice for handling fluorescence and the related phenomenon phosphorescence.
Second, many UV-C LEDs (emitters in the 200- to 280-nm range) exhibit a high forward voltage immediately after turn-on. This voltage decreases in a few milliseconds. LM-92 includes a test for this transient voltage effect in its Annex B. The test plots the forward voltage of two pulsed measurements, one at 25˚C and one at 26˚C on a square-root-of-time scale. By dividing the VF rise by the VF sensitivity factor, the junction temperature (TJ) rise is obtained. If this rise is unreasonably large for the pulsed condition and the expected thermal resistance of the LED, the LED exhibits the transient voltage effect (Fig. 1).
For LEDs with the transient voltage effect, inferring TJ rise using forward voltage is not possible. Since assessing TJ rise is an essential step in long-pulse measurements such as LM-85’s DC and single-pulse methods, these methods cannot be used for many UV-C LEDs. Instead, the LED must be measured in a way that maintains the TJ at the known ambient or case temperature.
Fortunately, while LM-92 was being drafted, IES’s LM-85 standard was also undergoing a major revision to improve its methods and to better handle LED TJ issues. Several new measurement methods were added to LM-85. In these methods. LEDs are powered with short pulses to reduce heating using third-generation source measure instruments. The LM-92 working group borrowed a technique from the revised LM-85 called Differential Continuous Pulse (DCP) as the required testing method for UV LEDs (Fig. 2). The DCP method uses very short pulses (10 and 20 µs) to limit heating and involves the subtraction of two individual measurements to produce a difference measurement. This subtraction removes errors such as current pulse timing and pulse shape errors that are common to both measurements.
The DCP method’s minimal heating keeps the LED junction close to the external ambient temperature, eliminating the need to calculate temperature shift, making DCP a natural pick for UV LEDs.
LM-92 presents the DCP technique in detail, including tables that provide specific requirements for the pulse widths to be used for various current levels. The standard also details techniques for determining the correct forward voltage (necessary when plotting I-V curves) when the transient voltage effect is present.
Future efficacy and labeling standards
LM-92 is the first of the testing standards identified by Miller and Jillavenkatesa. The next, for measuring excimer lamps (also known as far UV-C lamps), is slated to be published this summer. Other measurement testing standards to be developed in 2023 include standards for low-pressure mercury lamps (beyond linear tubes), pulsed xenon lamps, and finished products that are meant to connect to the branch circuit. The last standard includes a technology-neutral method that will enable direct comparison between products based upon dissimilar technologies. An additional standard is planned for the calibration and characterization of UV-C detectors. This document will provide a guide to use and apply UV-C detectors in the field as well as the laboratory.
Along with the need for better standards for measurements, IUVA identified the need for efficacy standards. As presented by Jim Bolton, et al., in a NIST Journal paper, an evaluation of peer-reviewed efficacy research turned up a wide range of reported deactivation dosages for common pathogens 6 — an indication of inconsistency in the laboratory methods used. IUVA negotiated with ASHRAE as the SDO to develop efficacy standards. ASHRAE already maintains two important standards for building air and for surfaces — standards 185.3P and 185.4P. IUVA signed a partnership MOU with ASHRAE on May 12, 2021, almost a year to the day after the IES MOU.
The final part of the IUVA standards plan is product labeling. For labeling standards, IUVA turned to the National Electrical Manufacturers Association. NEMA had already been active in exposing fraudulent GUV product claims, and in 2020, together with Underwriters Laboratory, NEMA issued a position paper on UV-C disinfection. IUVA and NEMA are actively collaborating on a phased approach for development and implementation of a standardized labeling program to convey the test results derived from the new IES and ASHRAE GUV standards7. Interested parties can access LM-92-22 via the IES website.
A multinational group of scientists, engineers, and healthcare professionals recognized the need for better UV standards, and the potential that those standards would unlock for GUV systems and applications. LM-92 and future standards will allow UV-LED manufacturers and GUV-product developers to design better products and to verify the effectiveness of those products for end users8.
1. Environmental Protection Agency, Federal Register Notice 72 FR 54039, 54039–54041.
2. Environmental Protection Agency Compliance Advisory, “EPA Regulations About UV Lights that Claim to Kill or Be Effective Against Viruses and Bacteria” (October 2020).
3. Centers for Disease Control and Prevention, HAIs Data Portal, Data Highlights.
4. White House Office of Management and Budget, Circular A-119, Revised (Feb. 10, 1998).
5. C. Miller and A. Jillavenkatesa, “Pathway to Developing a UV-C Standard – A Guide to International Standards Development,” IUVA News, 20, 4 (Q4 2018).
6. M. Masjoudi, M. Mohseni, and J.R. Bolton, “Sensitivity of Bacteria, Protozoa, Viruses, and Other Microorganisms to Ultraviolet Radiation,” J Res Natl Inst Stan, 126:126021.
7. M. Smith, T. Cowan, "The Need for Consistent Germicidal UV Labeling," UV Solutions, 2022 Quarter 2, 10–11.
DCP in detail
LM-92’s new Differential Continuous Pulse (DCP) measurement method is a twist on a classic electrical engineering noise-reduction technique: differential signal transmission. In differential transmission, a signal is split into two signals of two levels — such as 2S and S — and transmitted via two paths, each equally exposed to noise. At the receiving end, both signals are corrupted by added noise, but when the two received signals — such as 2S+N and S+N — are subtracted, the noise cancels out and the original signal is recovered.
In a DCP measurement, just as in differential transmission, the single continuous pulse measurement is split into two measurements. Both are continuous pulse measurements — that is, a measurement in which the LED is powered with short pulses at a low duty cycle.
The first measurement must be greater than the second for the subtraction to produce a positive result. In LM-92, one measurement is based upon 20-µs pulses (PL) and one uses 10-µs (PS) pulses. Ideally, the current pulses should be perfectly rectangular, but in reality, the leading and trailing edges have slope and distortion — shown in the blue and pink regions in the figure below. This distortion is analogous to the noise that occurs in differential signal transmission.
Since both measurements are subject to the same distortion, when the second measurement is subtracted from the first, the light from the distorted regions of the pulse is removed and what remains is light from a near-perfect rectangular portion of the longer pulse (PL, middle).
*Vektrex supplies test & measurement equipment such as source/measure units and current sources that support evaluation methods described in this article.
Get to know our experts
JEFF HULETT is CTO and founder of test & measurement specialist firm Vektrex, based in San Diego, CA. Over the course of his career, Hulett has acted as chief engineer on products ranging from single-board computers for spaceflight to power supplies for DNA sequencing systems to current sources used in the LED and laser industries for product and system evaluation. Hulett holds several patents related to LED testing and has authored technical articles for publications such as LEDs Magazine and Photonics Spectra. He also serves as chair of the Illuminating Engineering Society’s LM-80 Working Group.
CAMERON MILLER joined the National Institute of Standards and Technology in 1996 and from 2013 to 2021 was the group leader for the Optical Radiation Group. His research areas include all aspects of Photometry & Radiometry and measurement uncertainty. Miller is active in standards organizations and professional societies, such as IES – Testing Procedure Committee, IES – Science Advisory Panel Member, International Ultraviolet Association (IUVA), CIE, ASTM, and ISCC. He is also an NVLAP assessor for the Energy Efficient Lighting Program and the Calibration Program. Miller obtained his PhD in Physical Chemistry from Cornell University (1994).
TROY E. COWAN is the coordinator of the IUVA Healthcare Working Group and principal of Vision Based Consulting. He is an advocate for UV-C’s potential to save lives and the need for standards to demonstrate efficacy to healthcare providers. Cowan has met with several Senate and House Committee staffers and presented to IUVA, SPIE, and ISO on using UV-C to combat HAIs.
This article has been extended with additional images and text/references — abridged version published in the July/August 2022 issue of LEDs Magazine.