For some laser measurement tasks, the calorimeter is the preferred device. For others, it’s the pyroelectric joulemeter.
The range of commercially available lasers continues to grow rapidly, with growing scope of pulse widths, average power, continuous wave power repetition rates and wavelengths. Similarly the range of available energy and power meters continues to grow. Consequently, selecting a meter to accurately measure the power or pulse energy can be a daunting prospect. The relative merits and applicability of the two most common types of meters, disc calorimeters, and pyroelectric detectors, form the basis of this article.
Figure 1. Schematic cross-section of a typical disc calorimeter. The inset shows how a thermopile converts heat flow to current flow.
The principle of calorimeter laser detection of radiation is simple. Incident light is absorbed either by a surface (opaque) absorber, or a volume (semi- transparent) absorber. The absorbed energy is converted to heat which is measured by a thermoelectric sensor. Most of these meters are called disc calorimeters because they use a disc shaped absorber and are based on the original device developed at the National Bureau of Standards (now the National Institute of Standards and Technology).
In a disc calorimeter, heat is typically measured by a thermopile, which is essentially a Peltier cooler used in reverse. A Peltier, or thermoelectric cooler is a solid-state reversible heat pump based on the thermocouple junctions. It consists of a series of p-n semiconductor junctions (thermocouples) that pump heat when current flows through them. Conversely, as a thermopile, a thermoelectric cooler will generate a potential (charge separation) when the two sides of the device are at different temperatures. The measured current is proportional to the heat flow. In a typical disc calorimeter the thermopile is sandwiched between the absorbing (i.e. heated) disc and a large heat sink as shown in Figure 1.
One of the main advantages to the thermopile over alternative methods is spatially uniform response. As shown in Figure 1, the p-n and n-p junctions are arranged to be thermally in parallel but electrically in series. Consequently, it does not matter which part of the disc is heated or through which part of the thermopile the heat leaks to the heat sink: total heat flow is recorded. This has obvious advantages for measuring non-uniform laser beams or for the typical case where the laser beam is much smaller than the absorbing disc.
The calorimeter can accurately measure laser power of the energy in a single laser pulse. The disc gains heat through the absorption of laser radiation, and loses heat not only through the heat sink but also by convection and reradiation from the front surface of the disc. The response of this system is linear as long as the disc does not get so hot that reradiation and convection losses increase non-linearly. This, along with damage thresholds, sets an upper limit on the maximum power and energy ratings for accurate measurement. In practice, damage thresholds are determined by localized heating of the absorber by the laser. This is dependent on whether a surface or volume absorber is utilized, the absorber's heat capacity, and also the thermal conductance of the thermopile.
Surface absorber or volume absorber? Each has its advantages and disadvantages. Surface absorbers are thermally conductive substrates with a broadband coating, usually comprising finely divided metal or carbon. This type of detector has a very flat spectral response. Because they are good heat conductors, surface absorbers can withstand relatively high power densities and are therefore preferred for continuous-wave power measurement. However, because absorption occurs in a very thin layer, high peak power pulses produce thermal damage. For this reason, surface absorbers should not be used with Q-switched lasers.
Volume absorbers use discs of a dielectric material sufficiently transparent to all absorption through their volume instead of concentrating damaging heat at the surface. As such, absorption causes the laser energy to diminish exponentially as it passes into the disc. Although volume absorbers are not as spectrally flat as surface absorbers, they can withstand much higher peak powers without sustaining damage. Because they are relatively poor steady-state heat conductors, volume absorbers damage more easily than surface absorbers during cw power measurement.
Compared to calorimeters, pyroelectric detection is a relatively new method of quantitatively measuring incident laser energy. In this type of detection, special properties of ferroelectric crystals are used to sense changes in temperature due to absorption of laser radiation.
Ferroelectric crystals are a class of materials that can be prepared in a state of permanent polarization. This is somewhat analogous to the permanent magnetism that can be induced in ferromagnetic materials. In ferromagnetic materials, the degree of polarization is temperature dependent. For certain materials such as lithium tantalate and a new class of ceramics, the temperature dependency, d(pol)/dt, is a steep function at room temperature.
When one of these materials absorbs laser radiation, the incident energy is converted to heat. The corresponding temperature rise causes a change in the magnitude of electrical polarization of the crystal. This causes a charge separation (voltage) across the crystal, which is sensed as a current between two electrodes on the faces of the crystal (Figure 2). The current continues to flow until both sides reach the same voltage The current is given by:
I = pAdT/dt, where p is the pyroelectric coefficient and A is the area of the electrode.
Figure 2. In a pyroelectric detector, the polarization of a ferroelectric crystal is changed by the laser-induced temperature rise.
It is important to note that pyroelectric detectors have no steady state response; they only respond to changes in absorbed energy (heat).
The rise time of the pyroelectric voltage and the recovery time, or thermal time constant, of the detector (time to return to room temperature) are important parameters. These determine the range of pulse repetition rates and pulse durations a particular detector can measure accurately. Pyroelectric energy meters meters can be accurate, but only within a specified range of repetition rates and pulse widths. For example, if a particular detector has a fast recovery time Ð if it dissipates heat quickly Ð then during long laser pulses, it will start to achieve a steady-state condition. Even though more energy is being absorbed, the temperature and current will not increase proportionately to the total absorbed energy (Figure 3).
Figure 3. The effect of thermal recovery time on the relationship between the radiation pulse shape for a pyroelectric detector.
Conversely, if the detector has a slow recovery time relative to the pulse repetition rate, the signal will be an accumulation of the heat from several pulses. Again, accurate pulse energy measurement will not be possible.
The pyroelectric element is usually coated in order to enhance the response. "Black" coatings which absorb almost all of the incident radiation have a very flat spectral response. Partially reflecting coatings such as a thin layer of chromium have a faster response suited have a faster response suited to higher repetition rates.
Speed of response. The major differences between the two types of detectors are their relative speed of response and the fact that calorimeters respond to total incident flux intensity whereas pyroelectric detectors respond to changes in incident flux.
Calorimeters are best suited to measure average power. They usually have an overall response time between 1 and 3 seconds. For most applications of cw or psuedo-cw lasers, this represents the best compromise between the need to smooth (average) the laser and detector noise in order to get a steady reading of laser power, and the need for speed of response fast enough to use in optimizing (tweaking) the laser output.
In comparison, pyroelectric detectors are relatively fast but can not be used to measure cw laser output directly, since they only respond to changes in incident energy.
They are the best choice to measure the output of pulsed lasers with repetition rates up to several hundred kHz. The maximum pulse duration for accurate measurement varies from detector to detector (for example, from 0.45 ms to 4 ms). They can be used to measure a chopped or amplitude-modulated cw laser, but in order to obtain an absolute measurement the duty cycle of the modulation must be accurately known.
Typical calorimeters and even pyroelectric detectors are simply not designed to measure the pulse shape of most lasers Ð the time constants are designed for a compromise of modest speed and low noise, and are just too long for this type of measurement. A photodiode should be used for this purpose. Photodiodes can be used in conjunction with either calorimeters or pyroelectric detectors.
Calibration/NIST Traceability. Almost all commercial meters are traceable to a NIST standard. The transfer standard maintained by each manufacturer is usually a calorimeter. As a general rule, commercial calorimeters are more accurate than equivalent joulemeters and also tend to better maintain accuracy. Another advantage of calorimeters is the ability to use an electrical resistive heater as an internal calibration source (see Figure 1). Applying a fixed voltage to the heater applies a precisely known power to the calorimeter and the calibration in output volts per input watts is determined.
Other important aspects must be taken into consideration when selecting a meter. If a meter is deficient in only one of these areas, it may be rendered useless for a given application; all factors are potentially critical. Although a meter is only a small fraction of the cost of a laser system, it is the device that will be used to optimize the laser performance.
Spectral Range. The light-absorbing materials in both calorimeters and pyroelectric detectors are chosen for a number of different properties. One of these factors is spectral flatness. This is n important consideration for the many lasers that operate at multiple wavelengths. The absorption of some of these materials varies only a few percent over a spectral range of several microns. Figure 4 shows the spectral response of a "black" and a chromium coating used in Scientech’s pyroelectric detectors.
Dynamic Range. At first glance, a large dynamic range of response may not seem an important feature. For example, if the meter is to be used with a single 10 W laser, surely a 30 W scale should suffice. Surprisingly, this is not the case. Most applications require that a meter serve double duty as a calibrated measuring tool as well as a dynamic diagnostic to allow optimization of laser output. For lasers consisting of an oscillator and one or more amplifiers, many critical adjustments to the laser alignment will be made with a very low output power. Even for single stage lasers, part of the alignment procedure is often carried out with the laser power barely above threshold. A large dynamic range and an auto-range function can be very useful in this regard, since the combination allows the user to obtain high resolution at whatever power level is being used.
Sensitivity/Minimum Response. Every meter has a minimum detectable power, energy or change in energy which it can detect. This is determined by the noise of the meter. In detection of light, there are a number of potential sources of noise including statistical noise in the detector itself, amplifier noise, and shot noise, to name a few. The minimum detectable response is defined as that which is equivalent to the overall detection noise, i.e., the amount of laser radiation that produces a signal to noise ratio of 1:1.
For pyroelectric joulemeters, this quantity is defined as the noise equivalent energy — the change in incident laser pulse energy that gives a signal/noise ratio of unity. For calorimeters it is defined as the minimum resolution — the change of laser power which will give a signal to noise ratio of 1.
To measure low power/energy (<30 mW-mJ) accurately usually requires the calorimeter be shielded from ambient temperature fluctuations. Specially designed enclosures with optical baffling are available for just this purpose.
Damage Threshold. Although all meters have a specified maximum power and energy level rating, laser damage to the detector is still the most common cause of performance degradation. The important parameter that determines damage is energy per unit time per unit area. If the pulse time is short, or the laser beam is tightly focused, the probability of damage is greatly increased. As discussed earlier, detectors that rely on surface absorption have a much lower damage threshold to high peak pulse power than those that employ volume absorption.
The two principle reasons for laser damage are hot spots and failure of the operator to correctly extrapolate the damage rating for the instrument to a particular situation. For example, a pyroelectric detector is rated for a certain peak fluence, the operator may forget that this specification scales with pulse width.
Figure 5 shows how the pulse width shifts the maximum allowable energy density for Scientech pyroelectric detectors. Many types of lasers can produce output beams with small spot sizes of very high intensity in their beam profiles, especially if their optics are dirty or misaligned. The energy/power density in these hot spots is much higher than the average value for the beam and is often sufficient to damage a detector.
Interchangeable Controller. Almost all commercial meters consist of two separate parts: the sensor which generates a voltage or current, and the controller (or meter), which converts this signal into power (watts) or energy (joules), displays the result and often performs minor computations.
Many laboratories operate more than one type of laser. Often there is no single detector that will adequately measure the performance of all of them. In order to avoid the necessity to purchase several complete systems, certain manufacturers offer interchangeable detectors with the same meter. For example, a single controller is capable of operating with a range of different types of calorimeters as well as pyroelectric detectors. The controller can identify the type of detector by the "intelligent" circuitry on the calorimeter and by the output setting on the pyroelectric detector.
Multiple Pulse Analysis. Most of today’s pyroelectric measurement systems have sufficient "on-board intelligence" in the controller to capture a number of pulses (up to 1000) and statistically analyze them. This can be a very useful feature in applications where it is necessary to monitor and/or eliminate pulse-to-pulse fluctuations in laser output energy.
For measuring average power of cw or pulsed lasers, a disc calorimeter is the most suitable choice. To determine the particular meter and detector from the wide array of commercially available, consider the various application-specific factors discussed previously, such as wavelength range, damage threshold and dynamic range.
For measuring pulse energy of lasers with repetition rates of up to 4 kHz, use a pyroelectric detector. Once again, the exact model of meter is chosen by considering the various factors specific to your measurement needs.
Finally, if your require absolute measurements, consider the manufacturer’s capability for providing NIST traceable calibrations. Your experimental data or your laser performance can be limited by the capabilities of your meter. Accordingly, reliable, traceable calibration is a significant component of the meter’s performance.