Jun 2, 2017

LED testing and thermal issues

Unfortunately, no, there isn’t a reliable way to measure LED light performance right now. The National Institute of Standards and Technology (NIST) recently tested 118 identical lamps in laboratories around the globe using their Measurement Assurance Program. The results showed an undesirable difference of 4% in luminous flux measurements even though the lamps were all the same.  

Thermal issues represent the biggest problem when measuring device performance, because LEDs are extremely sensitive to temperature changes. A typical white LED yield is reduced to 0.25% for a 1°C rise, representative of a temperature fluctuation slightly more than the often-used 100Ω platinum RTD temperature sensor.

Why care?

The inability to accurately measure and compare LEDs has direct consequences on product manufacturing. Device performance can’t even be compared within the same company because manufacturers can’t know if two different production lines are creating interchangeable parts, a vital part of today’s industry.  Although manufacturers widen performance allowances to compensate for the uncertainty, the problems don’t go away and the lighting designers must inherit them. 

Labs can precisely regulate temperatures by using thermoelectric coolers which work fine under steady-state conditions; however, tests that use pulsed current create dynamic heat causing LED temperatures to increase during the pulse. Timing and light measurements are critical in these types of tests because light output is in constant decline.     

Another costly outcome results from variations in relative measurements. With any measurement variability, noise and trends become indistinguishable. Testing costs escalate because times must be increased for critical LM-80 qualification tests, thus forcing labs to perform many additional tests on more samples. 

What can be done?

One way to improve the process is by using the Illuminating Engineering Society of North America’s (IESNA’s) LM-85 Laboratory Method for Electrical and Photometric Measurements of High-Power LEDs. Accuracy is improved because the LED’s semiconductor junction temperature is tied to the LM-85 measurements, a vital step for ensuring reproducible measurements when performed in various laboratories. Additionally, the LM-85 continuous pulse method shows a rise in LED temperature improving flux measurement accuracy.

Instruments made to measure LEDs

Today’s instruments are typically borrowed from other industries or from previous technologies, such as those used to test incandescent lighting. The industry needs more advanced testing methods that can assure measurements happen at exactly the same point in time.

Current instruments concentrate on steady-state accuracy, many times at the expense of timing accuracy. Pulsed current sources and spectrometers with hardware-based activation represent instruments with good timing and would allow testers the ability to know exactly when measurements are made.

Because of the prosperousness associated with fast-developing LED technology, the industry has had to bear the costs and technical uncertainty related to the 4% inaccuracy. As innovation and growth continue, the present methods will become increasingly difficult and unreliable. Businesses and related organizations need to get ahead of the problem now by developing and using better testing methods and more accurate instruments.

Hisco carries a large inventory of advanced and innovative LED products including Loctite LED light curing systems and devices. 

Leave a Reply

Your email address will not be published. Required fields are marked *