Patent Application
20150316411
The disclosure relates to LED photometric testing, and more particularly to the use of calibration standards in such testing.
In-line testing of light-emitting diodes (LEDs), particularly white high-brightness LEDs used for general illumination, is generally performed with an integrating sphere and a photospectrometer. Accurate measurement of color coordinates and total light output requires calibration and periodic recalibrations traceable to an absolute standard certified by NIST or another national lab. Depending on the construction of the test apparatus (tester), initial and periodic calibrations may be required for each different type and package of LED (which can number into the dozens or even hundreds of different configurations).
Current calibration techniques involve a cumbersome two-step process where a “golden” set of devices are calibrated in a laboratory tester against a NIST standard. These devices are then run through a factory tester as “transfer standards.” Calibration offsets are calculated and programmed into the factory tester. The calibration must be done periodically (typically monthly, but may be otherwise) and is also performed when the tester is reconfigured to run a different device or package.
It typically takes a couple of hours to complete a calibration, resulting in significant loss of tester availability. The transfer standards are typically “home-made” and do not fit the testers well. In addition, the transfer standards usually do not stay with the testers because typically only one or two sets of such transfer standards are made and shared among all testers in a company. The multiple steps (including the frequent moves of the standards among testers) add measurement uncertainties and introduce more opportunity for human or mechanical errors. Temperature variations can also be a source of additional error.
In a light-emitting diode, a junction between p-type and n-type semiconductor forms the diode. The accuracy of an LED calibration standard is only guaranteed when the junction temperature, Tj, is held at its specified value. This is typically accomplished with a temperature regulator, which may be a heater/cooler comprising, for example, a resistive or thermoelectric device, a large heat sink, a fan, and an electronic controller. The temperature regulator causes the calibration standard to be significantly larger than the LED alone (see
Controlling the temperature of an LED calibration standard using a temperature regulator is also slow and inefficient because the diode junction being controlled is heated or cooled from outside the diode. In addition, the combined cost of the added temperature regulating device, the electronics for controlling it, and the heat sink can be approximately 100 times the cost of the LED alone.
The present disclosure provides a method for measuring optical radiation from a light-emitting diode (LED). The method comprises providing a pulse-width modulated (PWM) signal to the LED. The PWM signal has current pulses with a duty factor, a pulse width, an amplitude, and a frequency. One or more of the duty factor, the pulse width, the amplitude, and/or the frequency of the current pulses are adjusted to provide a forward voltage, Vf, corresponding to a target junction temperature, Tj, of the LED. The method comprises measuring, when the target junction temperature is obtained, the optical radiation of the LED during a current pulse. The method may further comprise measuring the Tj of the LED; calculating a target Vf based on a predetermined relationship of the change of Vf to the change of Tj. The method may be performed repeatedly for each of a plurality of LEDs.
In another embodiment, an LED test apparatus is provided. The apparatus comprises a stage configured to hold one or more LED for testing. The apparatus also comprises a signal generator for providing a power signal to an LED in the stage. The signal generator is configured to provide a PWM signal wherein the signal generator can adjust a duty factor, a pulse width, an amplitude, and/or a frequency of the PWM signal such that the LED operates at a target Tj. A photospectrometer is provided for measuring an optical radiation of the LED in the stage. In some embodiments, the LED test apparatus further comprises an integrating sphere having a test port, and wherein the stage is configured such that an LED in the stage provides optical radiation into the integrating sphere by way of the test port. In some embodiments, one or more LED calibration standards are affixed to the stage of an LED test apparatus.
In another embodiment, a method for calibrating an LED test apparatus is provided. The method comprising providing a PWM signal to an LED calibration standard. The PWM signal has current pulses with a duty factor, a pulse width, an amplitude, and a frequency.
One or more of the duty factor, the pulse width, the amplitude, and/or the frequency of the current pulses are adjusted to provide a Vf, which corresponds with a target Tj of the LED calibration standard. When the target junction temperature is obtained, the optical radiation of the LED calibration standard is measured during a current pulse. Optical measurement offset value(s) for the LED test apparatus are calculated according to the measured optical radiation of the LED calibration standard.
In another embodiment of the present disclosure, a method for heating an LED to a target Tj is provided. PWM signal is provided to the LED. The PWM signal has current pulses with a duty factor, a pulse width, an amplitude, and a frequency. The method comprises adjusting the duty factor, the pulse width, the amplitude, and/or the frequency of the current pulses to provide a Vf which corresponds with the target Tj of the LED.
In another embodiment, an LED test apparatus is provided. The Apparatus has a stage configured to hold one or more LED for testing. At least one LED calibration standard is affixed to the stage. The apparatus has a signal generator for providing a power signal to an LED in the stage. The signal generator is configured to adjust a duty factor, a pulse width, an amplitude, and/or a frequency of the power signal such that the LED operates at a target Tj. The apparatus has a photospectrometer for measuring an optical radiation of the LED, when a LED is in the stage, or the LED calibration standard affixed to the stage.
For a fuller understanding of the nature and objects of the disclosure, reference should be made to the following detailed description taken in conjunction with the accompanying drawings, in which:
In one aspect, the present disclosure can be embodied as a method 100 for measuring optical radiation from a light-emitting diode (LED) calibration standard. In order to provide thermal regulation of the LED calibration standard, a pulse-width modulated (PWM) signal is provided 103 to drive the LED calibration standard. As is known in the art, a PWM signal is made up of a plurality of current pulses (typically square wave pulses of high and low values, but may have other waveforms). It should be noted that the term PWM is used for convenience throughout the application to refer to a signal comprising a plurality of pulses, and should not be interpreted as limiting the disclosure only to embodiments where the pulse width of the signal is modulated. As will be apparent in light of the present disclosure, in some embodiments, other characteristics of the signal are modulated while the pulse width is constant. Characteristics of the current pulses can be varied. For example, the pulse width of a current pulse can be made longer or shorter in duration, such as in
According to embodiments of the present disclosure, the junction temperature, Tj, of an LED driven by a PWM signal can be controlled by varying the characteristics of the PWM signal. While LED current pulses have been used for determining Tj, the present disclosure allows for active heating of the LED to bring Tj to target. The PWM signal may be adjusted, for example, to provide a forward voltage across the junction of the diode, Vf, which corresponds with a target junction temperature, Tj, of the LED. This is found to be useful for self-heating (i.e., intrinsic heating) of the diode junction in an LED calibration standard—thereby reducing or eliminating the need for external heaters in such standards. Furthermore, the size of the heat sink can be reduced or eliminated, possibly using only the LED substrate or a printed circuit board (PCB) on which the LED is mounted as the heat sink. Without an external heater and heat sink, an LED calibration standard can be significantly reduced in size and, in some cases, may be the same size as an LED device-under-test. This allows for in-situ calibration standards on production equipment as further described below. Such fast control and stabilization of the junction temperature of LEDs may also allow for higher throughput testing of production LEDs.
As such, the method 100 includes the step of adjusting 106 the duty factor, the pulse width, amplitude, and/or the frequency of the PWM signal pulses to provide a Vf corresponding to a desired Tj of the LED calibration standard. The target Tj is often the designed real-world operating Tj of the LED and is higher than the test ambient temperature (or the heat sink temperature). A typical value for target Tj (depending on the type of the LED) is between 40° C. and 85° C., but higher and lower target junction temperatures are possible. The target Tj may be a range around a target temperature. For example, a target Tj may have a tolerance off 1%, 2%, 3%, 5%, 10%, or other values around the temperature. As Tj approaches its target value, the properties (pulse width, duty factor, frequency, and amplitude) of the current pulses may be further adjusted 106 to minimize the time required for Tj to settle at its target value (as shown in the figure above, for example). In some embodiments, a feedback loop can be used to adjust the pulse properties according to the Tj.
When the target Tj is obtained (e.g., Tj has settled within its tolerance from the target Tj), the optical radiation of the LED calibration standard is measured 109. The optical radiation measurement 109 may include, for example, spectral flux, luminous flux, radiant flux, color coordinates, correlated color temperature, etc. A typical optical radiation measurement takes 2˜20 ms to complete. In some embodiments, it is preferred to make this measurement 109 after the initial steep rise of Tj. For example, in the graph of
A measurement using an exemplary method according to an embodiment of the present disclosure was made. The PWM signal over the first 5000 ms is shown in
In embodiments of the present disclosure, the amplitude of the current pulse during which the optical radiation measurement is made, or when calculating the target Vf, must be the same as the value for which the Vf vs. Tj function was established. At other times, the current pulses can have a different amplitude (from near zero to more than the measurement current) and/or a different duty factor (from near zero to 100%, i.e., CW) in order to reach and maintain the target Tj effectively, as long as the drive signal is within the safe parameters for the LED.
In some embodiments, one or more pulse parameters may be held constant while varying other parameters to obtain a Vf for a desired Tj. For example, in some embodiments of the method 100, repetitive current pulses can be run with specified amplitude through the LED. The properties of repetitive pulses (mainly the duty factor, but may also include absolute pulse width, and frequency) can be adjusted to obtain a specific forward voltage Vf corresponding to a desired Tj for the specified amplitude of current.
The mechanism for controlling Tj is the balance between heating by such adjustable LED current pulses and heat dissipation into ambient air and/or the minimized heat sink. Tj can be driven to the desired value and maintained at such value by adjusting the LED current pulses (e.g., pulse width, duty factor, amplitude, and frequency) to obtain a desired Vf. This adjustment can be made using a static algorithm or in real-time using an adaptive algorithm. In an example of a static algorithm, a target Vf can be set based on prior knowledge of the absolute Vf vs. Tj function for a given current for each type of LED, or for a specific individual LED.
In an example of a real-time adjustment, the target Vf can be set by first measuring actual Tj (equal to the LED substrate or heat sink temperature which can be obtained using a separate temperature sensor, for example, a calibrated thermocouple, prior to turning on the LED) and Vf at the beginning of the first pulse. The target Vf may then be calculated based on a known “change of Vf” vs. “change of Ti” function in real-time for of an LED. Such real- time approaches are usually more accurate than static algorithms. Other techniquest for determining target Vf (i.e., Vf corresponding to a target Tj) can be used.
The accuracy of the presently disclosed pulsed LED standard is comparable to that of its CW counterpart, because it is determined by the accuracy of LED current and Tj at the moment of measurement, which is obtained through precision timing for pulsed standards. In addition, because Tj can be controlled adaptively, a single hardware design and algorithm can accommodate a wide range of LED types, testing tools, and operation conditions. No separate heating mechanism (other than self-heating as disclosed herein), e.g., from heat sink, hot air, or any radiation heating, is required, although such mechanisms can be used in combination the presently disclosed technique. While prior heating techniques using external mechanisms require times on the order of minutes to tens of minutes for heating, the presently disclosed method is advantageous and the target Tj may be reached as fast as a few seconds or less. The faster heating capability enables faster and more frequent calibrations. In some cases, the heating time may be fast enough for production testing of LED's at their real-world operating temperatures. Ambient or heat sink temperature (or other environmental conditions) would generally not affect the final Tj or the optical radiation measurement results, because Tj is adjusted in real-time to an absolute target value above and independent of ambient temperature.
During the self-heating phase (with current pulses) before optical radiation measurement, in order to reduce the time required to reach target Tj, the amplitude of the pulses may be different from that to be used for calculating target Vf and for optical radiation measurement. The waveform of the current pulses in during this time may be different from square-wave, for example, the waveform may be triangular, having smooth tops and bottoms, or modulated by <100% (never reaching zero), to reduce noise and errors caused by high harmonics. Such pulses may also be interspersed with regular measurement pulses such as described above. This method of heating may be used alone, or in combination with other heating methods, for example, hot air, radiation, or resistive heating, which may result in more accurate temperature control and/or faster settling.
In another aspect of the present disclosure, an LED test apparatus 10 is provided. The apparatus 10 may be used, for example, to test the function of LEDs during the manufacturing process. The apparatus 10 comprises a stage 12, where an LED device-under-test (DUT) 90 is located during testing. The stage 12 may be a platform, a conveyor belt, a rotating turret, or any other device suitable for positioning the DUT 90 in the apparatus 10. In some embodiments, such as where the stage 12 is a conveyor belt, the apparatus 10 may be configured such that the DUT 90 is in continuous motion during the test.
The stage 12 is further configured receive an in-situ LED calibration standard 95 in place of a DUT 90. In some embodiments, the in-situ calibration standard has a form factor that is the same as a DUT. In other embodiments, the form factor of the standard need not be the same, but is compatible with the DUT such that the apparatus 10, including the stage 12 does not require reconfiguration in order to process the standard.
The apparatus 10 comprises a signal generator 20 for providing a power signal to a device-under-test 90 or a calibration standard 95 in the stage 12. In this way, the DUT 90 or calibration standard 95 can be energized for measurement of the optical radiation. The signal generator 20 is configured to provide a PWM signal for heating the calibration standard 95 before measurement. For example, the signal generator 20 may include a controller 22 which is programmed to perform any of the methods described above for heating an LED to a target Tj. In a particular embodiment, the controller 22 is programmed to cause the signal generator to provide a PWM signal to the calibration standard 95 and to adjust parameters of the PWM signal pulses (duty factor, pulse width, amplitude, frequency, etc.) to obtain a target Tj of the LED calibration standard 95. The signal generator 20 may be configured to provide a PWM signal to a DUT 90 to heat the DUT 90 to a desired Tj. In this way, the DUT 90 can be tested at a selected operating temperature.
The controller 22 may further be programmed to determine a Tj based on the Vf provided by the signal generator 20. As such, the controller 22 may be further programmed to provide closed-loop control of the Tj.
Although described as a controller, it is to be appreciated that the controller 22 may be implemented in practice by any combination of hardware, software, and firmware. Also, its functions as described herein may be performed by one unit, or divided up among different components, each of which may be implemented in turn by any combination of hardware, software and firmware. Program code or instructions for the controller 22 to implement the various methods and functions described herein may be stored in processor readable storage media, such as memory.
The apparatus 10 further comprises a photospectrometer 30 for measuring an optical radiation of a device in the stage 12. The optical radiation measurement may include, for example, spectral flux, luminous flux, radiant flux, color coordinates, correlated color temperature, etc. In some embodiments, the stage 12 is moveable such that a DUT 90 or an LED calibration standard 95 can be moved into a testing position for measurement by a spectrometer.
The photospectrometer 30 may be in coordination with the signal generator 20 such that the optical radiation measurement is performed after the device being measured (whether a DUT 90 or a calibration standard 95) has been brought up to a desired Tj. For example, in some embodiments, the photospectrometer 30 is in communication with controller 22 and may receive a measurement actuation signal from the controller 22. In this way, when the device settles within the tolerance range of the target Tj, the controller 22 provides a signal to the spectrometer 30 at the start of the next PWM pulse such that the spectrometer may measure the device at a time during such pulse. Other configurations and timings may be used. For example, spectrometer 30 may be signaled by a controller that is not controller 22, or the spectrometer 30 may provide a signal to other components, etc. A typical optical radiation measurement takes 2˜20 ms to complete. In some embodiments, the spectrometer 30 may be configured to measure the optical radiation of a device after an initial steep rise of Tj during a PWM measurement pulse. In some embodiments, the start time of the optical radiation measurement within the current pulse can be set dynamically at a specific target Tj (for example, using Vf to determine Tj). In other embodiments, the measurement may also be set to start when Tj is slightly below the target value. In this way, the average value of Tj during the entire optical radiation measurement time is equal to its target value.
The LED test apparatus 10 may include an integrating sphere 40. Such integrating spheres are known for use in optical measurements. When an integrating sphere 40 is used, the stage 12 is configured such that a device in the stage 12 provides its optical radiation into a hollow cavity 42 of the integrating sphere 40. For example, the stage 12 may position the device at an test port 44 of the integrating sphere 40. As such, the optical radiation is scattered in a diffuse way and may be measured by the photospectrometer 30 which is configured in a measurement port 46 of the integrating sphere 40. In another embodiment, where the stage 12 is a conveyor belt, the integrating sphere 40 may have an inlet port 47 and an outlet port 48 such that the conveyor belt may transport a device through the cavity 42 of the integrating sphere 40.
Apparatus of the present disclosure advantageously provides that LED calibration standards are presented to the spectrometer (for example, by way of the integrating sphere) in the same way as the devices-under-test. In this way, geometric calibration offsets can be reduced or eliminated. In this regard, it may be advantageous, in some embodiments, to configure the integrating sphere to collect 100% of the light from both the calibration source 95 as well as the DUTs 90. Transfer standards can be made from product LEDs (i.e., can be made from the same LEDs as the devices-under-test), which allows for calibrating that is specific to each model of LED product. This would significantly reduce measurement errors caused by using a “generic” calibration standard for different products with different optical properties (particularly spectrum, beam profile, light output, and self-absorption of light).
Embodiments of the present disclosure may use calibration sources which are relatively broadband so as to cover the full range of wavelengths emitted by various LED DUTs. Alternatively, multiple calibration sources could be provided, each optimized to match various LED DUTs based on lumen output, drive current, angular light distribution (beam profile), spectrum (or CCT), etc. In some embodiments having multiple in-situ transfer standards 95, a first standard may be calibrated off the apparatus in order to establish absolute calibration and traceability (this is not expected to be done frequently and can become a maintenance step). The other in-situ standards of such an apparatus can then be calibrated to the first standard.
In some embodiments, the stage 12 is configured such that the LED calibration standards 95 (i.e., the in-situ transfer standards) are removable from the stage, allowing them to be calibrated or recalibrated on a separate measurement system. For example, the standards may be calibrated using a laboratory tester calibrated to NIST, to establish and maintain traceability and accuracy. Similarly, the stage 12 may be configured such that the LED calibration source module may be exchanged with a different calibration source, allowing for the use of a calibration source which is the same as (or similar to) the device-under-test.
In some embodiments, such as the exemplary embodiment of a tester 60 of
An apparatus 60 may further comprise an alignment camera 80 for proper positioning of a device within the test port 74 of the integrating sphere 70. The alignment camera 80 may be positioned to align the stage 62 based on a marker 63 on the stage 62, which is away from the DUT 90. In other embodiments, the alignment camera is positioned to align the stage 62 based on the alignment of the DUT 90 in the test port 74. In the apparatus 60 depicted in
The integrating sphere may also have a calibration port in which a traditional LED calibration standard can be placed. For example, the apparatus 60 of
In some embodiments, apparatus of the present disclosure will not include an LED device-under-test or a calibration standard until the apparatus is placed into service. However, in other embodiments, one or more LEDs and/or one or more LED calibration standards may make up a part of the apparatus. For example, in some embodiments, an apparatus comprises a stage having an LED calibration standard affixed thereto.
Although the present disclosure has been described with respect to one or more particular embodiments, it will be understood that other embodiments of the present disclosure may be made without departing from the spirit and scope of the present disclosure. Hence, the present disclosure is deemed limited only by the appended claims and the reasonable interpretation thereof.
This application claims priority to U.S. Provisional Application No. 61/988,087, filed on May 2, 2014, now pending, and U.S. Provisional Application No. 62/065,749, filed on Oct. 19, 2014, now pending, the disclosures of which are incorporated herein by reference.
This invention was made with government support under DE-EE0005877 awarded by the U.S. Department of Energy. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 61988087 | May 2014 | US | |
| 62065749 | Oct 2014 | US |
