240.437.4615 | Contact Us

Round Robin Test of the IUVA UV Lamp Efficiency Testing Protocol (Revised)

James R. Bolton1 and Michael J. Santelli2

1. 628 Cheriton Cres. NW, Edmonton, AB, Canada T6R 2M5 (corresponding author)
2. Light Sources, 37 Robinson Road, Orange, CT 06477


Over the past several years, the Manufacturers’ Council of the International Ultraviolet Association (IUVA) has developed a protocol for testing of the UV efficiency of low-pressure and low-pressure high-output UV lamps. The first version of this protocol was adopted in 2008 (Lawal et al., 2008). A round-robin test of this protocol was conducted in 2009, but the results were not satisfactory. In reviewing the 2008 protocol, there was no flaw in the requirements but rather the practice. Such factors as adherence to protocol, reflectance and measurement methods led to high sources of error.

Hence, in 2013, The Manufacturers’ Council adopted a revised protocol (see Appendix A on page 15) that further refined the requirements of the 2008 protocol. Before finalizing this new protocol, it was decided to conduct a second Round Robin. This paper presents the results of this new Round Robin.

The Round Robin was initiated by testing at Light Sources Inc. (Orange, CT) of 15 Light Sources 320 W (model GPHHA1554T6L/4P) low-pressure high-output (spot amalgam) UV lamps (arc length 1.475 m) according to the revised Protocol using an electronic power supply (Philips Electronic Ballast – Model TUV 325W XPT) and a recently calibrated radiometer (International Light model ILT 1700 with an SED240 detector with an NS254/NS254 filter and a W diffuser). After the tests were completed, the lamps, power supply and radiometer were shipped sequentially to nine other manufacturers where similar tests were conducted.


The testing procedures followed the protocol given in Appendix A. Some centers used method 1 to minimize reflection, and some used method 2, as shown in Appendix B.


Table 1 gives the average lamp efficiencies under four conditions as obtained from the tests conducted by the participating manufacturers, which were (in random order): Hanovia, Trojan, Xylem, Light Sources, Calgon Carbon, Heraeus, Philips, Foshan-Comwin, LightTech and Ozonia. Table 2 gives the average lamp voltages, lamp powers, optical powers, lamp current and the ambient and surface temperatures. Detailed tables of results for individual lamps are in Supplementary Files.

The four conditions were:

  1. Steady-state efficiency after a period of at least 10 min
    • Input to the lamp
    • Input to the ballast
  2. Efficiency at time of peak output (usually about 2-5 min)
    • Input to the lamp
    • Input to the ballast


The following observations can be made:

  1. The results are generally highly reproducible among the various test centers.
  2. There was a 2-3% drop in efficiency between measurements across the lamp vs. those from the wall. This indicates that the power supply consumes about 7-12% of the input power with the rest going to the lamp.
  3. One would expect that the efficiency at peak output should be more reproducible among the various labs; however, there is not much difference in the reproducibility between the data for steady state and the data for peak output.
  4. There does not appear to be any significant correlation between lamp efficiencies and the ambient temperature, although the temperature range is quite small.
  5. One test center (C) failed to make reliable measurements of the output across the lamp.
  6. Some centers (e.g., C and H) reported electrical values that were well outside the norm. Perhaps there was a problem with the calibration of their power analyzers.
  7. The distance from the center of the lamp to the radiometer detector varied from 2.88 to 4.00 m. There did not appear to be any correlation in the results vs. this distance. This confirms the recommendation that this distance be at least twice the arc length.
  8. When the lamps, ballasts and radiometer were returned to Light Sources and again tested in October 2014, the results were almost the same as when the lamps were originally tested in September 2013.


These results provide strong verification that the IUVA Test Procedure is effective and valid.


The authors wish to express their thanks to the participating manufacturers and specifically to:
Bruno Ferran – Ozonia North America
Casey Theys and Christina Crowley – Calgon Carbon
Alin Cojocaru – Trojan
Peter Vigh – LightTech
Egbert Vossen – Philips
Martin Kessler – Heraeus
Uwe Kanigowski – Xylem
Mark Aston – Hanovia
Vincent Liao – Foshan Comwin

The authors wish to especially thank Light Sources who provided the UV lamps, power supplies and radiometer for the tests.


Lawal, O., B. Dussert, C. Howarth, K. Platzer, M. Sasges, J. Muller, E. Whitby, R. Stowe, V. Adam, D. Witham, S. Engel, P. Posy, A. van de Pol. 2008. Proposed Method for measurement of the output of monochromatic (254 nm) low pressure UV lamps, IUVA News, 10(1), 14–17.