detector basics - mark allen

Germany and other countries: LASER COMPONENTS GmbH, Phone: +49 8142 2864 0, Fax: +49 8142 2864 11, [email protected]: LASER COMPONENTS S.A.S., Phone: +33 1 3959 5225, Fax: +33 1 3959 5350, [email protected]

Detector Basics

The radiation sensitive key components of InfraTec‘s detectors are single crystalline lithium tantalate (LiTaO3) elements formed as a very thin plate capacitor. Lithium tantalate is a pyroelectric crystal whose ends become oppositely charged when heated. Although this unique effect was already known in the ancient world and was given the name pyroelectric in 1824 by Brewster, even though the broad application in infrared detectors was introduced in the early 1970s. Nowadays due to its simple but robust construction and its performance the pyroelectric detector is one of the most widely-used thermal infrared detectors. In figure 1 the individual stages of the transformation from infrared radiation to an electrical signal is

represented. Via a window or IR filter with a transmission rate F of the radiation arrives at the pyroelectric

element. The radiation flux S is absorbed and causes a change in temperature TP in the pyroelectric element. The thermal to electrical conversion is due to the pyroelectric effect by which the temperature

change TP alters the charge density ΔQP on the electrodes. An electrical conversion often follows in which, for example, an electrical signal ΔuS is created by a preamplifier or impedance converter.

Fig 1: Conversion stages of the pyroelectric infrared detectors

QP uSFST P

Electrical conversion

Thermal conversion

Thermal to electrical conversion

Pyroelectric Library


Detector Basics

1 Thermal Conversion

Within this chain thermal conversion is the basis for a high responsivity and a high signal-to-noise ratio (often

abbreviated SNR or S/N), through which a high temperature change TP is the objective. Figure 2 represents a simplified thermal model and in figure 3 the equivalent electrical circuit is depicted. The

radiation sensitive element is characterized by the absorption rate , the heat capacity HP and the thermal conductance GT to its surroundings which is represented by a heat sink with a given temperature TA.

Fig 2: Simplified thermal model Fig 3: Equivalent electrical circuit

Using the thermal time constant T

PT G

H (1)

the temperature difference results in 22PT

SFP

HGT

(2)

or for sinusoidal agitation in the steady state 21

1~

~

TT

SFP G

T

(3)

For significant temperature differences to occur the product F has to be as near to 100 % as possible. This can especially be achieved by the use of an absorption layer. The heat capacity value HP has to be low. For this the thickness tP of the radiation-sensitive element has to be very low. Compromises are necessary as the required reduction in the thermal conductance GT is opposed by the increase of the thermal time

constant T.

AS HP

GTTP

t P

TA

F S

heat sink

F

S

1/j

HP

1/G

T

T

P


Detector Basics

2 Thermal to Electrical Conversion

The thermal to electrical conversion is due to the pyroelectric effect and is proportional to the temperature rate and the detector‘s surface area:

dt

TpAi P

SP

(4)

For sinusoidal agitation and considering equation (3) the result for the rms value (root mean square) of the pyroelectric short circuit current iP is as follows:

21

1~

~

TT

SFSP G

pAi

(5)

Figure 4 represents the frequency dependence on the temperature change and the short circuit current of a typical pyroelectric detector at an incident radiation flux of 1 µW. The frequency dependence on the temperature change portrays the typical low pass characteristics. The corner frequency fT results from the thermal time constant according to equation (6):

TTf 2

1 (6)

and has the value of 1 Hz. Below the corner frequency the temperature change attains a saturation value of 513 µK. Above the corner frequency, the pyroelectric current, however, attains a saturation value of approximately 2.2 pA.

1

10

100

1000

0,001 0,01 0,1 1 10 100 1000Frequency [Hz]

ΔT

P [

µK

]

10

100

1000

10000

i P [

fA]

Temperature change

Pyroelectric current

F==1

S=1µW

GT=1.95mW/KHP=310µWs/K

T=159ms

tP=25µmAS=4mm²

cP=3.1J/cm³/Kp=17nC/cm²/K

Fig 4: Frequency dependence on temperature change and short circuit current of a pyroelectric element


Detector Basics

3 Electrical Conversion

3.1 Responsivity

The extremely low current, supplied by a high-impedance source has to be converted by a preamplifier with a high-impedance input. There are two alternatives available: voltage mode and current mode. The voltage mode can be implemented using a voltage follower and the current mode using an inverting operational amplifier (OpAmp) as seen in figure 5.

+

-

Current Mode

-

+

-

+

Voltage Mode

-

1/jCP

RG uS

1/jCPuS

1/jCfb

Rfb

+

Fig 5: Alternative preamplifier circuits

The signal voltage uS and the responsivity RV for both modes can be defined respectively using the same equation:

2/122/12 1

1

1

11~~

ETTSSFS R

GpAu

(7)

2/122/12 1

1

1

11~~

ETTSF

S

SV R

GpA

uR

(8)

wherePGE

G

CR

RR

is valid for the voltage mode (9)

andfbfbE

fb

CR

RR

for the current mode. (10)


Detector Basics

High-megohm resistors may be necessary for both current and voltage mode to achieve a high signal voltage and responsivity but the feedback capacitance Cfb is kept considerably lower than the capacitance of

the pyroelectric chip CP. Therefore the electrical time constant E is considerably lower for the current mode and the signal voltage above the electrical corner frequency is considerably higher than for the voltage mode. Figure 6 illustrates the frequency dependence of both modes for typical detectors based on the results represented in figure 4.

0,01

0,1

1

10

100

0,001 0,01 0,1 1 10 100 1000Frequency [Hz]

uS [

mV

]

10

100

1000

10000

100000

RV [

V/W

]Voltage mode

Current mode

(E VM)² = 1

S=1µW

CP=62pFCfb=0.68pF

RG=Rfb=24G

E VM=1.5s

E CM=16ms

(T)² = 1

(E CM)² = 1

Fig 6: Comparison of the frequency dependencies of signal voltage/responsivity for voltage and current mode

3.2 Noise and specific detectivity

The noise sources of the pyroelectric chip limit the detectable radiation flux or the signal-to-noise ratio. Noise sources are: tan noise of the pyroelectric element

Temperature noise

Input noise voltage of the preamplifier

Input noise current of the preamplifier

Johnson noise of the high-megohm resistor


Detector Basics

As a measure of the signal-to-noise ratio the specific detectivity is frequently used in conjunction with infrared detectors:

N

VS

u

RAD ~

2/1* (11)

Where Nu~ is the effective noise value which is related to a noise bandwidth of 1 Hz at the preamplifier

output (voltage noise density). In table 1 the individual noise sources as well as the appendant noise voltage

components are summarized. The quadratic superposition of the components results in the noise voltage

density Nu~ . The frequency dependence and the resulting noise voltage density for a typical pyroelectric

detector LME-302, with the already utilized parameters, is portrayed in figure 7.

Noise sources Components of the noise density NXu~

tannoise of the pyroelectric element V

E

PPND AR

CkTu2

12

2/1

1)tan4(~

(12)

Temperature noise 2/12 )4(~T

VNT GkT

Ru

(13)

Voltage noise of the preamplifier VNNV Aeu ~ (14)

Current noise of the preamplifier V

E

NNI AR

iu2

121

~

(15)

Johnson noise of the high-megohm resistor V

E

NR AR

R

kTu

21

2

21

1

4~

(16)

Table 1: Five essential noise sources (AV: voltage gain)

In an ideal pyroelectric detector the heat exchange due to radiation between the pyroelectric chip and its

surroundings acts as the only unavoidable noise source. This temperature noise NTu~ determines the

theoretically highest possible specific detectivity of a pyroelectric detector operated at room temperature:

WHzcmD 10max 108.1 (17)

In a typical pyroelectric detector the other noise sources are considerably higher, shown in figure 7. The

Johnson noise of the high-megohm resistor dominates at low frequencies (< 10 Hz). At medium frequencies

(100 Hz), the tanδ noise of the pyroelectric element dominates the resultant noise density. At high

frequencies (> 1000 Hz), the voltage noise of the preamplifier specifies the resultant voltage noise density.


Detector Basics

1

10

100

1000

10000

100000

0,1 1 10 100 1000 10000Frequency [Hz]

No

ise

[V/H

z1/2]

tanδ Noise

Temperature Noise

Voltage Noise

Current Noise

Johnson Noise

Resultant Noise

Fig 7: Frequency response of the resulting voltage noise density and of the various components of the voltage noise density of a typical pyroelectric detector (LME-302)


Detector Basics

4 Voltage Mode

4.1 General information

Due to its simplicity the voltage mode is the most commonly used operating mode for pyroelectric detectors. The following restrictions have to be considered regarding the layout of the amplifier and signal conditioning unit: The signal voltage of a pyroelectric voltage mode detector usually includes very low-frequency parts

(mHz) caused by 1/f characteristics.

The cut-on frequency of the amplifier‘s high pass should not be too low.

The gate resistor (load resistor) should have a resistance of at least 10 GOhm for high performance.

The best solution for the protection of high-impedance components against humidity, which would cause

current leakage, is the integration of these inside transistor style housing. Pyroelectric detectors should

not be used without integrated impedance preamplifiers in high performance applications.

The output signal of voltage mode detectors corresponds to the time-integral of the IR radiation.

This behavior suppresses fluctuations effectively. Sinusoidal signals, however, are phase-shifted by 90°

by this electrical lowpass filter (f > fT).

4.2 Circuit diagram

In the simplest case the preamplifier is formed as a JFET source follower. The gate resistor and the JFET are integrated into the detector housing. The resistor RS in the source line is placed outside the detector housing (see figure 8). The high signal-to-noise ratio and the low temperature dependence, as well as the simplicity of the circuitry, are the reason for this widespread use.

uncompensated parallel compensated serial compensated

Fig 8: Basic circuits for the voltage mode

The gain for these circuits results from the transconductance gfs of the JFET in the operating point and

the source resistance RS:

11

Sfs

Sfsv Rg

RgA (18)

The demand for a high source resistance or a small drain current can be deduced from this.


Detector Basics

See below for an example equation for a gain of at least 0.8 (IDSS = saturation drain current):

1.0DSS

D

I

I (19)

However the demand for a low output resistance limits the increase of the source resistance necessary for a gain near to the value of 1. The source resistance should not be over 100 kOhm at drain voltages up to 15 V. A constant current source can be used as an alternative as this possesses a very high inner resistance. Next to a gain value of approximately 1 the temperature dependence of the transconductance is simultaneously suppressed and therefore the temperature stability of the gain is improved. See figure 9 for suggestions concerning the operation of the source follower. The JFET used by InfraTec represents a IDSS with a characteristic value of 1 mA. The recommended drain current values for the operation of the detectors are between 10 µA and 100 µA.

Remarks

(1)

The simplest circuit

High drain current and gain dependency

on operating point

Low dynamic

Asymmetrical tuning

(2)

Lower current and gain dependency on

operating point as in (1)

A greater tuning range than in (1),

symmetrical tuning possible

Negative supply voltage

(3)

Higher dynamic inner resistance

Greater tuning range than in (1),



Lower temperature drift

2 active and 2 passive components

(4)

Higher dynamic inner resistance

Low noise

Greater tuning range than in (1),



Greater temperature drift than in (1)

Fig 9: Alternative low-noise drain current supplies for voltage mode detectors

Output

Output

Output

Output


Detector Basics

For the design of the drain current supply circuitry please note the following: The noise optimum for the JFET used in the InfraTec detector lies at 20 µA.

Pyroelectric detectors generate a DC offset in temperature ramps, which defines the large signal

behavior and can lead to significant changes in the gain of the source follower. Uncompensated

standard detectors portray a positive offset shift. In comparison compensated detectors approximately

portray a ten-fold lower shift, which, dependent on the symmetry between the active and the

compensating element, can be positive or negative.

This occurring effect, taking place exclusively in the temperature ramps, can be minimized at the

expense of a higher noise, using a lower electrical time constant (available for all types on demand).

The integrated current sources available, for example the LM 134 from NSC, worsen the signal-to-noise

ratio or are expensive (REF200 from Burr-Brown/TI).

4.3 Wiring suggestions

The electronic components shown in figures 10 to 15 which are connected to the pyroelectric detector considerably determine noise and large-signal response. However, low cost OpAmps can be used due to the high signal level of pyroelectric detectors in comparison to thermopiles. The best results are achieved using low-noise amplifiers, which have been developed for high quality audio applications.

Wiring for a low voltage supply preamplifier for the NDIR (nondispersive infrared) gas analyzing

sensor

Fig 10: Using TO18 detectors with electrically isolated housing. (3.3 V Lithium battery supply; (0.4 … 33) Hz; gain 60)


Detector Basics

TG

ain

[d

B]

-10

0

10

20

30

40 a b

Frequency (Hz)

10m 100m 1 10 100 1k 10k

Ph

ase

[d

eg

]

-80-60-40-20

020406080

Fig 11: Gain and phase of preamplifier as per figure 10 vs. frequency

T

Time (s)16.0 17.0 18.0 19.0 20.0

Uo

ut

(V)

1.0

1.5

2.0

2.5

3.0Ip=2pA; fp=0,5Hz

T

Time (s)8.0 8.5 9.0 9.5 10.0

Uo

ut

(V)

1.0

1.5

2.0

2.5

3.0Ip=2pA; fp=1Hz

T

Time (s)9.00 9.25 9.50 9.75 10.00

Uo

ut

(V)

1.0

1.5

2.0

2.5

3.0Ip=2pA; fp=2Hz

Fig 12: Uout of detector and preamplifier as per figure 10 at different frequencies a)0.5 Hz b)1 Hz c)2 Hz (simulated)

Wiring for a high precision preamplifier for the NDIR gas analyzing sensor

Fig 13: Using dual channel detectors. (±5 V or higher; (0.3 … 22) Hz; gain 100)

a) b) c)


Detector Basics

TG

ain

[d

B]

-10

0

10

20

30

40

50 a b

Frequency (Hz)

10m 100m 1 10 100 1k 10k

Ph

ase

[d

eg

]

-80-60-40-20

020406080

Fig 14: Gain and phase of preamplifier as per figure 13 vs. frequency

T

Time (s)16.0 17.0 18.0 19.0 20.0

Uo

ut

(V)

0.0

500.0m

1.0

1.5

2.0

Ip=2pA; fp=0,5Hz

T

Time (s)8.0 8.5 9.0 9.5 10.0

Uo

ut

(V)

0.0

500.0m

1.0

1.5

2.0

Ip=2pA; fp=1Hz

T

Time (s)9.00 9.25 9.50 9.75 10.00

Uo

ut

(V)

0.0

500.0m

1.0

1.5

2.0

Ip=2pA; fp=2Hz

Fig 15: Uout of detector and preamplifier as per figure 13 at different frequencies a)0.5 Hz b)1 Hz c)2 Hz (simulated)

a) b) c)


Detector Basics

5 Current Mode

5.1 General information

Current mode pyroelectric detectors are not as widely available as voltage mode ones, the most probable reason for this being that elementary pyroelectric detectors are mass-produced for light switches and motion detectors. Due to the complexity of the preamplifier circuit its use was limited to very few applications. At InfraTec we can supply a wide spectrum of current mode detectors which makes it less complicated to include detectors for gas and fire detection.

5.2 Circuit diagram

Transistor style housing containing only the

pyroelectric element

Transistor style housing containing the pyroelectric

element, JFET and feedback resistor (an additional

feedback capacitor of several picofarad is also

possible). The integrated feedback capacitor

prevents so-called gain peaking.

Transistor style housing containing the pyroelectric

element (also available with thermal compensation

as LME-335 or LME-345) and a complete current

voltage converter with low input bias current

OpAmp.

Fig 16: Three alternative pyroelectric detectors suitable for current mode


Detector Basics

5.3 Wiring suggestions

The following examples are supposed to inspire one to consider the current mode as a reasonable alternative to the classic voltage mode based on the most modern available components.

Fig 17: Circuit for current mode ((0.2 – 25) Hz; 1 V/nA) of the pyroelectric detectors LME-301 and LME-501

Fig 18: Circuit for current mode (typical 31,000 V/W) of the pyroelectric detectors LME-300 and LME-500

OpAmps with a low voltage noise should be used. Disadvantages of the circuitry depicted in figures 17 and 18 include: EMC (Electromagnetic Compatibility) problems due to parasitic capacities

Permanent voltage offset across the pyroelectric element due to VGS of the JFET

IGSS of the JFET determines the level and temperature dependence of the current noise

These disadvantages can be avoided by integration of the OpAmp into the detector housing.


Detector Basics

Fig 19: Circuit for current mode (typical 64,000 V/W) of the the pyroelectric detectors LIM-262

Modern low power OpAmps with low current and voltage noise ensure the same signal-to-noise ratio as in a simple JFET source follower, however due to the considerably lower electrical time constant a considerably higher responsivity can be achieved. The advantages of the integrated current mode detector are that: Although there is a very high responsivity (RV > 100,000 V/W) there is also a high stability.

There is a very low output offset together with a very low offset temperature drift.

There is very low current noise together with a lower temperature drift of the specific detectivity at low

frequencies (f < 100 Hz).

There is a low electrical time constant and therefore a short warm-up phase and fast recovery time.

There is no signal loss when using the parallel compensation in thermal compensated detectors.


Detector Basics

6 Infrared Sources

The application of a suitable IR source depends on the actual case of operation. Numerous sources with the corresponding IR spectra are available. In the chapter ’Spectral Emission of IR Sources’ some spectra are described. For simple applications incandescent lamps can be used. For more demanding uses normally hot plates are applied. The operation of LED’s is limited due to the very small spectral emission range. The radiation of the light sources always needs to be chopped. Common frequencies are between 0.1 an 100 Hz. One possibility to realize this is the mechanical interruption of the radiation with a chopper wheel. An alternative is the electronic triggering of the source (for example: lamp 6 V/115 mA T1 1160-08 (MGG); hot plate 6.5 V/135 mA MIRL17-900 (Intex) or IR-LED). For the realization different circuits can be used. An incandescent lamp typically is chopped with a few Hz (max. up to 10 Hz), the control circuit always needs to be adapted to the used lamp. The maximum chopper frequency for a hot plate source is in the range of up to 100 Hz. The electronic drive of the IR source MIRL17-900 with constant voltage driver is shown in figure 20 (+V = 9 V DC).

Fig 20: Example: Source MIRL17-900 (Intex) with constant voltage driver

An electronic drive independent of the supply voltage [+V = (8 ... 18) V DC] can be realized with a constant current driver. Figure 21 gives an example for this circuit with additional adjustment of the lamp currents for lighting and dark lighting.

Fig 21: Example: Source MIRL17-900 (Intex) with constant current driver


Detector Basics

7 Low Noise Power Supply

Batteries (Alkaline, Silver oxide-cell, Lithium cell) are a good choice for the detector supply as they have a very low ground noise. Generally linear dropout controllers can be used for the supply of both positive and negative voltage. Typical fixed controller IC’s are to be seen in figure 22 and 23.

Fig 22: L78L05 fixed +5V and L79L05 fixed –5V Fig 23: LT1761ES5-X (X=fixed voltage) and LT1964ES5-5 fixed (-5.0V)

For variable supply voltages the controllers LM317LZ (pos. voltage) / LM337L (neg. voltage) are a good choice as shown in figure 24. To achieve an adjustable output with declined proprietary current consumption LT1761ES5-BYP (1.3 ... 20) V and LT1964ES5-BYP (-1.3 … -20) V can be used. We recommend to employ ceramic capacitors respectively solid tantalum capacitors with a small equivalent series resistance (ESR) and a low temperature coefficient.

Fig 24: Variable power supply LT1761ES5-BYP for positive voltage


Advanced Features of InfraTec Pyroelectric Detectors

1 Pyroelectric Detectors with JFET source follower or integrated CMOS-OpAmp - A Comparison

In 2003 InfraTec added pyroelectric detectors with integrated CMOS Operational Amplifiers (OpAmp) to our established family of detectors with integrated Junction Field Effect Transistors (JFET). Advancements in analog Silicon based technologies have allowed us to replace the classic JFET design with a more complex amplification circuit for nearly all applications. Well established JFET detector designs like the LIE-302, LIE-316 and LIM-222 have been enhanced with CMOS OpAmps as in the LME-341, LME-345 and LIM-262. LME-316 with JFET (thermal compensated) LME-335 with CMOS OpAmp (thermal compensated)

-

+

Ground=Case

Source

Drain

+

-

-

+ +

-

V-

Ground=Case

V+

Out

Voltage mode Current mode

Fig 1: Examples of pyroelectric detectors with JFET (left side) and OpAmp (right side)

Detectors with JFET and OpAmp are distinguished by more than orders of magnitude in signal voltage. The fundamental difference you find is in the analysis of the pyroelectric signal. As a result you will get different frequency responses of signal and noise (response and typical parameters in figure 2).

0,01

0,1

1

10

100

0,001 0,01 0,1 1 10 100 1000Frequency [Hz]

uS [

mV

]

10

100

1000

10000

100000

RV [

V/W

]Voltage mode

Current mode

(E VM)² = 1

S=1µW

CP=62pFCfb=0.68pF

RG=Rfb=24G

E VM=1.5s

E CM=16ms

(T)² = 1

(E CM)² = 1

Fig 2: Frequency response of signal voltage uS and voltage responsivity RV of a pyroelectric detector with 2 mm x 2 mm active sensing element



In voltage mode the pyroelectric current, created in the single crystalline LiTaO3 chip, charges the electric capacity. The resulting voltage is displayed by a simple source follower (JFET, gate resistor and external source resistor). In current mode the generated pyroelectric current is transformed by a Current-Voltage-Converter (OpAmp with feedback components, also named Trans-Impedance-Amplifier TIA). The frequency dependent conversion factor I/U is determined by the complex feedback components and is typically in the range of

(10 ... 200) pA/V. While the thermal time constant T (typically 150 ms) as a measure of the thermal coupling of the pyroelectric element to its surrounding is effective in both operation modes, the electric time constant

E is determined by different components. In voltage mode E is calculated as a product of pyroelectric chip

capacity CP and gate resistor RG (typically 1.5 s). In current mode E is only determined by the feedback components Rfb and Cfb (typically 16 ms). Main differences between pyroelectric detectors with JFET and CMOS-OpAmp At common modulation frequencies between 1 Hz and 10 Hz in gas analysis and flame detection the

detector will operate above the thermal and electrical time constant (1/f behavior of signal). The maximal

responsivity is located beyond the normal modulation frequency range. Low-frequency disturbances up

to some millihertz will be transmitted. Detectors need settling times up to some 10 seconds.

Detectors in current mode are mostly operated between both time constants and resultant cut-on and

cut-off frequency. Here the signal voltage is on its highest level and stable over a broad frequency range,

possibly over some hundred hertz. Low-frequency disturbances are one magnitude away from the cut-on

frequency and will therefore by suppressed 10 times more compared to the voltage mode. The

measuring signals are already stable after a few seconds.

Due to the virtual short circuit of the pyroelectric element in current mode, an antiparallel connected

compensation element does not lead to a reduction of signal and detectivity. Furthermore an incomplete

illuminated pyroelectric element in current mode does not cause a loss of both signal and detectivity in

contrast to the voltage mode.

Why we are using CMOS-Operational amplifiers? CMOS technology combines technological and customer demands for a low supply voltage, low power consumption, Rail-to-Rail performance at output and low chip costs. Additionally the completely isolating gate (SiO2) in the operational amplifier shows a better performance during operation at high temperatures as opposed to the JFET design. The current mode which earlier was only possible to apply in combination with very expensive OpAmps like OPA128 or AD549, can now be applied in applications for gas analysis and flame detection which were previously dominated both technologically and price wise by the JFET. Comparison of the modulated output signal for detectors with JFET and OpAmp The electrical time constant defines the form of the output signal in current and voltage mode. Identical time constants lead to the same signal form in both modes. In current mode we can work with a nearly arbitrary electrical time constant, which is an essential advantage. Therefore short time constants are preferred due to the resulting short settling time.



Figures 3 to 6 show typical signal characteristics in order with decreasing electrical time constant.

Fig 3: LME-302 Voltage mode at 1 Hz, 5 Hz and 20 Hz, thermal time constant 150 ms, electrical time constant 5s

Fig 4: LME-335 Current mode at 1Hz, 5 Hz and 20 Hz, thermal time constant 150 ms, electrical time constant 20 ms

Fig 5: LME-341 Current mode at 1 Hz, 5 Hz and 20 Hz, thermal time constant 150 ms, electrical time constant 5 ms

Fig 6: LME-351 Current mode at 1 Hz, 5 Hz and 20 Hz, thermal time constant 150 ms, electrical time constant 1ms



Feedback Resistor Influence to Responsivity (signal), Detectivity (signal-to-noise ratio) and Stability of the operating point for a detector with integrated OpAmp In accordance with voltage mode detectors the Ohm rating of the integrated resistor leads to opposite detector properties: A large resistor results in a high signal and an increased detectivity since the noise only increases with

the square root of the resistor value. In contrast an amplifier stage after the detector would increase

signal and noise by the same ratio.

A small resistor increases the stability of the DC operating point, therefore a thermal compensation is

often not necessary for R<10 GOhm.

1

10

100

1000

0,1 1 10 100Frequency [Hz]

Res

po

nsi

vity

[kV

/W]

LME-335LME-345LME-351

1,0E+07

1,0E+08

1,0E+09

0,1 1 10 100Frequency [Hz]

Sp

ec. D

etec

tivi

ty [

cmH

z1/2/W

]

LME-335LME-345LME-351

Fig 7: Frequency response of responsivity (signal) and specific detectivity (signal-to-noise ratio) as a function of the feedback resistor; LME-335 (100 GOhm), LME-345 (24 GOhm), LME-351 (5 GOhm)

Power supply for InfraTec CMOS OpAmp detectors We use mainly a split power supply (±2.2 ... ±8) V for our detectors. Principally a single supply (4.5 ... 16) V is also possible, for this version the TO39-housing of our single detectors is located on Reference potential (mostly U/2). For customized detectors we can also integrate different operational amplifiers, which can be operated either with a very small supply voltage (single supply +2.2 V) and isolated detector housing or with very high supply voltages and a high dynamic range (split supply ±13 V). InfraTec can assure the availability of detectors with JFET source follower and CMOS-OpAmp for many years to come. The technical advantages will however accelerate the trend to use the current mode OpAmp detectors.



2 Microphonic Effect in Pyroelectric Detectors

2.1 Basics

All pyroelectric crystals are inherently piezoelectric. When a pyroelectric detector is mechanically excited through shock or vibration, an unwanted signal is produced. This behavior is called a microphonic effect or vibration response. The interaction of mechanical and electrical variables in piezoelectric crystals can be expressed in an open-circuit operation by a simplified equation for the electrical field strength E and its dependence on stress T as shown in table 1. Depending on the orientation of the stress, two basic effects can be distinguished: the Transverse Effect (along the chip edges) and the Longitudinal Effect (through the thickness).

Transverse Effect Longitudinal Effect

Model

ET

ET

||

Open-circuit voltage abhd

uT

vib~

2330

31

ahd

uT

vib~

2

2

330

33

Open-circuit voltage of a 30 µm thick LiTaO3 chip, 3 mm sq. element size, acceleration 9.81 m/s² (b=3 mm, h=30 µm, a=1 g)

Vuvib 25 Vuvib 8.0

Table 1: Comparison of transverse and longitudinal effect on a square thin chip

If the stress T is applied into the plane of the chip and transverse to the electric field strength E one can use the transversal model. The stress is a linear function in the X direction and exhibits its maximum value at the bearing point and zero-crossing at the right border. The mean stress is half of the maximum stress TXX. A lithium tantalate element with a square electrode of (3 x 3) mm and thickness of 30 µm would produce an open-circuit vibration voltage of about 25 µV at 1 g (9.81 m/s²). If the acceleration acts longitudinal to the electrical field strength E and out of the plane of the chip the open-circuit vibration voltage per g is about 0.8 µV for a 30 µm thick lithium tantalate chip. This vibration voltage represents a limit for the reduction of the microphonic effect for a single element. Any further reduction could be achieved only by adding an anti-parallel/anti-serial compensating element.



2.2 Reduced microphonic voltage by ‘in-chip-compensation’

The simplest solution for reducing the stress caused by the transverse effect is the simultaneous clamping of the chip, both at the left and right borders. This results in tensile and compressive stress in the halves of the element and would counterbalance each other. A central fastening leads also to the compensation of stress. As shown in figure 8 the open-circuit vibration voltages are minimized by both the fastenings.

Fig 8: Stress distribution on a double-sided (left), and center (right) clamped pyroelectric detector chip (3 x 3 x 0.03) mm³ with acceleration in the X direction

The analytical description of the microphonic effect shows that the transverse effects are minimized by an outer symmetrical mounting, or a central mounting and that the longitudinal effect is much lower. However, stress overshoot around the mounting points is produced, when the acceleration is applied out-of-plane. In this case the simplified analytical description of the longitudinal effect does not provide proper results, since the chip structure and chip mounting varies significantly from the ideal configuration. In contrast to the ideal configuration, the arrangement and thickness of the electrodes is different. Furthermore, an additional absorption layer displaces the neutral stress line out of the center line. The influence of an acceleration out-of-plane could only be described accurately by a numerical analysis, for example by the simulation software ANSYS Multiphysics. As shown in figure 9, the base for the numerical analysis is a chip holder. The chip holder consists of a base plate with a central column surrounded by four symmetrically arranged columns. The pyroelectric chip is assembled on the chip holder by adhesive bonding.

base plate

adhesive bonding

centrical column

pyroelectric chip

outer column

Fig 9: Model of the assembly of chip holder and pyroelectric chip



In the plane of the pyroelectric chip the open-circuit vibration voltage is compensated by the symmetrical arrangement of all columns. If we look at the out-of-chip plane, the open-circuit vibration voltage is compensated by the arrangement of the outer surrounding columns in such a manner that convex and concave warpings are induced by mechanical excitation in the normal direction. These warpings produce tensile and compressive stress at one surface and the reversed at the opposite side of the pyroelectric chip as illustrated in figure 10. The positive and negative piezoelectric charges generated by the stress are compensated by the electrodes, which cover the top and the bottom sides. To achieve a stress compensation, the chip deformation can be optimized by modifying the arrangements of the mountings (see figure 11).

Fig 10: Deformation and stress of a quarter of a pyroelectric chip assembled on a chip holder using different support positions along the chip diagonal

0,0

1,0

2,0

3,0

0 0,5 1 1,5 2Radius edge to bearing [mm]

Acc

eler

atio

n r

esp

on

se [

µV

/g]

Fig 11: Acceleration response of the chip shown in figure 10 as a function of the mounting point position

The stress distribution inside of a pyroelectric chip is also affected by the chip coating, technological parameters, as well as the elastic modulus of the adhesive by which the pyroelectric chip is bonded onto the chip holder. Nevertheless the statistics of sample measurements confirmed a good fitting of the test results and simulated values.



2.3 Microphonic effect at Voltage and Current mode operation

The operational mode of a pyroelectric detector and the frequency range affects the result of piezoelectricity at detector’s output (see figure 12).

Voltage mode The Open-circuit operation of a pyroelectric chip as described in the earlier ‘Detector Basics’ section is given typically in voltage mode for frequencies higher than 0.5 Hz. The criteria for this is the electrical break point set by the product of chip capacity CP (30 ... 120) pF and gate resistor RG (5 ... 100) GOhm. In this frequency range the vibration or so called microphonic voltage at the detector output is identical with the open-circuit vibration voltage (see ‘Detector Basics’):

vibVMvib uu (1)

A typical signal of a voltage mode detector is in the order of millivolts. A 100 g acceleration (e.g. a shock) can produce a similar but disturbing signal of the mechanical vibration if its frequency fits the amplifier pass band.

Out

+5V

+RSRG

+

+

Gnd/Case

V-

Out-

V+

Cfb

Rfb

Fig 12: Voltage mode operation (left) and current mode operation (right) of a pyroelectric detector

Current mode State-of-the-art pyroelectric detectors make use of the current mode (see figure 12) more and more. From the open-circuit vibration voltage uvib the short-circuit current ivib is derived, which increases in a linear manner with the frequency. The short-circuit current flows in a preamplifier and generates a signal voltage u’ at the output:

2/12)(1

1

E

fbvibvibCM Riu

(2)

As in voltage mode operation the microphonic voltage at detector’s output is constant for frequencies well above the electrical cut-off frequency. Because the electrical time constant in current mode is defined by the feed back components Cfb and Rfb (this time constant is clearly shorter) this frequency range starts only from some 10 Hz. In this frequency range the vibration voltage at the detector output is created by the open-circuit



vibration voltage but amplified by the quotient of the capacitance of the pyroelectric chip and of the feedback capacitance:

1fbPvibCMvib CCuu (3)

Comparison

1,0E-04

1,0E-03

1,0E-02

1,0E-01

1,0E+00

1,0E+01

1,0E-01 1,0E+00 1,0E+01 1,0E+02 1,0E+03 1,0E+04Frequency [Hz]

sig

nal

/vib

rati

on

no

ise

volt

age

u'[

mV

]

signal voltage VM

vibration noise voltage VM

signal voltage CM

vibration noise voltage CM

S = 1µW

Cfb = 2.2 pF

CP = 114 pF

E CM = 11 ms

E VM =0.6 s

T = 159 ms

R = 5 GAS= 9 mm²

dP = 30 µm

a = 9.81 m/s²

(E VM)² = 1

(T)² = 1

(E CM)² = 1

Fig 13: Comparison of the signal voltage u’S and of the vibration interference voltage u’vib z (longitudinal effect) in voltage mode and current mode

Figure 13 shows the frequency correlations with straight lines for the vibration voltage and dashed lines for signal voltage. The results are indicated with circles and squares for the current mode and the voltage mode respectively. Even if the vibration voltage behaves differently for current and voltage mode the ratio of signal to vibration voltage at a specific frequency (means the distance between the two red or the two blue lines in the diagram) is the same for current and voltage mode. There is no advantage from point of vibration responsivity by using current or voltage mode.



2.4 Introduction of the term ‘Microphonic-Equivalent Power (MEP)’

InfraTec introduced the Microphonic-Equivalent Power (MEP) for a simplified discussion:

V

vib

R

RMEP

(4)

with

a

uR vib

vib ~

(5)

The MEP is defined as the quotient of the vibration responsivity and voltage responsivity and indicates the incident radiation flux, required to generate an equivalent root-mean-square (rms) signal voltage for a given vibration. It is expressed in the units of W/g (gravitational acceleration = 9.81 m/s²). Table 2 summarizes vibration responsivity, voltage responsivity and MEP. Of course the MEP is more or less independent from the operation mode of the preamplifier. In the case of a standard detector, the vibration responsivity and hence the MEP strongly depend on the spatial direction of the applied vibration. In the case of microphonic reduced detectors (‘low micro’ type) with the novel chip holder, it is clearly demonstrated that the MEP value can be reduced to about (3 … 2) nW/g @ 10 Hz in all three spatial directions. Detector Vibration Responsivity

Rvib (10 Hz, 25 °C) in µV/g

Voltage ResponsivityRv (500 K, 10 Hz, 25 °C) in V/W without window

Microphonic Equivalent Power MEP (10 Hz, 25 °C)

in nW/g

x y z x y z

LIE-502 (VM) 16 1,6 3,5 160 100 10 22

LIE-500 (CM) 550 55 120 5.500 100 100 22

LME-502 (VM) 0,5 0,5 0,4 160 3 3 2

LME-500 (CM) 65 65 50 22.000 3 3 2

Table 2: Vibration responsivity, voltage responsivity and microphonic-equivalent power of standard and ‘low micro’ detectors

Conclusion There are three ways to reduce the influence of the piezoelectric behavior on pyroelectric detectors in customer’s sensor modules: Suppress mechanical vibrations as good as possible by pulse damper (smooth rubber, flexible cables).

Please note that the elongation [mm] at a constant acceleration [m/s²] is frequency dependent. A

sinusoidal acceleration of 1 g = 9.81 m/s² is the result of a peak-to-peak elongation of:

70 cm at 1 Hz 7 mm at 10 Hz 70 µm at 100 Hz 0.7 µm at 1 kHz

In a compact sensor module mechanical damping can only be realized for frequencies higher than

100 Hz.



Limit the electrical pass band of the amplifier stages especially at the high frequency side by a steep low

pass.

Compensation of the microphonic voltage by sophisticated mounting of the pyroelectric chip.

InfraTec offers a variety of pyroelectric detectors of voltage mode (VM) or current mode (CM) operation with a reduced vibration response (so called ‘low micro’) based on InfraTec’s patented chip mounting technology. The reduction of the microphonic voltage is in the order of a twentieth (5 %) of a conventional pyroelectric detector. Figure 14 shows their typical frequency response. Please note that the differing vibration voltage of the CMOS OpAmp detector series LME-351, -341 and -335 caused on a differing gain. LME-335 offers the highest responsivity (90,000 V/W @ 10 Hz).

0,1

1

10

100

1000

10 100 1000Frequency [Hz]

Vib

rati

on

Res

po

nsi

vity

[µ

V/g

]

LME-335LME-341LME-351LIE-502LME-502

Fig 14: Test results of microphonic effect

LME-502 (VM, (3 x 3) mm², ‘low micro’), LIE-502 (VM, (3 x 3) mm², conventional),

LME-351 (CM, (2 x 2) mm², ‘low micro’, 5 GΩ//0.2 pF), LME-341 (CM, (2 x 2) mm², ‘low micro’, 24 GΩ//0.2 pF),

InfraTec’s ‘low micro’ technology is available for single element detectors such as LME-316 (VM) or LME-345 (CM) and multi color detectors such as LMM-244 with an element size of (2 x 2) or (3 x 3) mm². These detectors can be identified at the part description by a ‘M’ in the second digit (LME instead of LIE or LMM instead of LIM).



3 Beamsplitter Detectors - even for the narrowest signal beams

The principle of the multi color detector with integrated beamsplitter is shown in the following picture. The IR radiation entering through the aperture stop is divided by a beamsplitter in two or four parts (4 channel pictured). Each of the partial beams goes through an IR filter and then hits a pyroelectric detector chip. This design works well with single narrow-beam sources, or in situations where contamination (dust, insects) in the light path could be an issue.

Infrared Filter

Beam Splitter Array

Detector Chip

Radiation

Infrared Filter

Beam Splitter Array

Detector Chip

Radiation

Fig 15: Principle of reflective beamsplittering Fig 16: 3-D Design LIM-054

3.1 General design

The beamsplitters are made of gold plated microstructures for the two and four channel devices to achieve a homogeneous distribution of the radiance. The filters are mounted at a certain angle to obtain a normal incidence of the radiation. This configuration avoids drifts in the filter transmission curves to shorter wavelengths and the influence of the opposite filter (reflections). In addition to four channel beamsplitter detectors using four-sided micro pyramids, InfraTec has also developed two channel detectors based on micro V-grooves. In the following figure SEM images of two and four channel beamsplitters are shown. The V-groves pitch is 100 µm and the pyramids are 50 µm, with the tilt angle of the filters and detectors set at 30°.



Fig 17: Micro groove (2-channel) Fig 18: Micro pyramids (4-channel)

3.2 Comparison to other multi channel detectors

The beamsplitter detector has a single aperture stop compared to other multi channel detectors. It is possible to use a gas cell with a smaller diameter reducing the gas volume. A smaller gas volume reduces the size of the sensor module and accelerates the gas exchange. Also, the signal ratio of all channels is independent from aging, mechanical shift or pollution processes among one another. The multi color detectors should be used for the analysis of gas mixtures with few known gases. Typical examples for a successful application are anesthetic gas monitors and the pulmonary function testing. Variable color detectors allow a more flexible operation of the analyzer enabling the detection of adjoining or overlapping absorption bands. So far the measurement of single gases like ethanol and carbon dioxide as well as gas mixtures of methane, propane and anesthetic gases have been tested. In the following table characteristics of the multi and variable color detectors are summarized. Detector Specification Multi Color Multi Color Variable Color

Principal Individual Windows Beamsplitter Tunable Fabry-Perot Filter

Filtering Parallel Parallel Serial

Area to be illuminated Ø 9.5 mm Ø 2.5 mm Ø 1.9 mm

Spectral Range (2 … 25) µm (2 … 25) µm

(3.0 … 4.1) µm and

(3.9 … 4.8) µm

Current Mode Yes Yes Yes

Voltage Mode Yes Yes No

Thermal Compensation Yes No Yes



4 Basics and Application of Variable Color Products

The key element of InfraTec’s variable color products is a silicon micro machined tunable narrow bandpass filter, which is fully integrated inside the detector housing. Applying a control voltage to the filter allows it to freely select the wavelength within a certain spectral range or to sequentially measure a continuous spectrum. This design is very different from detectors with fixed filter characteristics and enables the customer to realize a low resolving and low cost spectrometer. The variable color product group includes the LFP-3041L-337 and LFP-3950L-337 which differ in the wavelength range they each cover. The pyroelectric detector used is similar to the standard LME-337 device.

4.1 Fabry-Perot filter (FPF)

The filter-detector assembly (see figure 19) is based on the well-known Fabry-Perot Interferometer (FPI). Two flat and partially transmitting mirrors with reflectance R are arranged in parallel at a distance d, forming an optical gap. Multiple-beam interference is created inside the gap and thus only radiation can be transmitted, which satisfies the resonance condition according to equation (1). One of the mirrors is suspended by springs so that the distance d can be decreased by applying a control voltage. As the resonance condition changes so does the wavelength of the transmitted radiation.

m

dnm

cos2 (1)

n refractive index inside the gap

β angle of incidence (β=0 in fig 19)

m interference order

d optical gap

Fig 19: Configuration and operation principle of the FPF with an integrated pyroelectric detector

The transmittance spectrum T() of the FPF (see figure 20) is described by the Airy-Function. Given that n=1 (air inside the gap) and β=0 (vertical incidence), this results in:



1234

Tmax

T50HPBW

T ()

FSR

CWL

)2(sin1 2

max

dF

TT

(2)

with the F-value:

21

4

R

RF

(3)

Fig 20: Transmittance spectrum of a FPF

The characteristic parameters of a FPF can be derived from the previous equations. In our case, the first interference order is used (m=1), while the higher orders are blocked by means of an additional bandpass filter. The center wavelength (CWL) of the filter corresponds to the resonant wavelength in theory. In practice the CWL is measured as the mean value of the two half-power-points (T50). The half-power bandwidth (HPBW) is the decisive factor for the spectral resolution:

R

RdHPBW

1

2

(4)

The spectral distance of two adjacent interference peaks limits the maximum usable tuning range. This is referred to as the free spectral range (FSR). The ratio of the tuning range to the bandwidth (in the frequency or wavenumber domain) is given as the

Finesse RF~

, which is the figure of merit of a FPF:

21

~ F

R

R

HPBW

FSRFR

(5)

4.2 Optical Considerations

In the real world there are some practical constraints, so it’s necessary to complete the previous statements and equations: The mirrors of the FPI are made from dielectric layer stacks (Bragg reflectors). This limits the width of the reflective band and thus the usable spectral tuning range to about 1.3 µm. Bragg reflectors have a distinct phase shift, which actually has to be considered in the equations stated above. This causes an increase of the mechanical adjustment travel Δd:

dkCWL 2 (k<1 phase correction) (6)



An inclined but collimated beam results in a negative drift of the CWL (see figure 21 left). The most common case is an uncollimated beam with a certain angle of divergence and intensity profile. The resulting transmittance spectrum can be seen as the superposition of collimated ray-beams with different angles of incidence and intensities. The superimposed spectrum has a broader HPBW and the CWL at slightly lower wavelengths (see figure 21 right).

Fig 21: Influence of angle shift and divergence angle on bandwidth and peak transmittance of a FPF

Beam divergence can be minimized by using a light source with collimated output or by means of an additional prefixed aperture (see figure 22 left). If the desire is to maximize the optical throughput, then focusing optics can be used, but larger divergence angles will be a side effect (see figure 22 right).

High Resolution High SNR

FP-Detector

Aperture

IR-sourcecollimated output

Sample cell

FP-DetectorFocusing optic

IR-sourceSample cell

Fig 22: Possible optimizations for the optical design of a microspectrometer with FPF detector left set up: corresponds with an illumination by a parallel beam; right set up: corresponds with a high angle of incidence (AOI)

The performance of such a system strongly depends on the optical conditions. A compromise between spectral resolution and signal-to-noise ratio (SNR) for the particular application needs to be found. This principle is in fact valid for all spectrometric applications. Figure 23 shows the correlation of the achievable SNR with a given spectral resolution, measured with two tested measurement set ups according to figure 22. Please note that a parallel beam ø1 mm offers the highest spectral resolution but only 3 % of the intensity and thus the resulting low detector signal voltage compared to an illumination using f/1.4 optics



100

1.000

10.000

40 50 60 70 80R

SN

R

Fig 23: Measurements of the SNR vs. spectral resolution LFP-3041-337 with a modulated IR source at Hz, left end point: Illumination at high angle of incidence (AOI) using f/1.4 optics right end point: Illumination with a parallel beam ø 1 mm

Tuning the CWL of the FPF results in a variation of the HPBW and the peak transmission within certain limits, too. The additionally implemented broad band pass and the pyroelectric detector element also show some spectral characteristics. The spectral response of the detector is therefore a superposition of different fractions, but has to be considered as a whole in the application. It is stated as the relative spectral response, referring to a ‘black’ reference detector with a flat spectral response (see figure 24).

0V11,7V

18,1V22,2V25,2V

27,5V

29,2V

30,7V

0,0

0,2

0,4

0,6

0,8

1,0

2800 3200 3600 4000 4400 [nm]

rela

tive

sp

ectr

al r

esp

on

se

Fig 24: Relative spectral response of a FPF detector LFP-3041L-337 at several tuning voltages

„High SNR“ f/1.4 optics

„High Resolution“ parallel beam Ø1mm



4.3 Filter operation

The filter is activated electrostatically. The driving electrode (Vc+) is arranged at the fixed reflector carrier, the movable reflector carrier acts as an electrode with the fixed reference potential Vcref (see figure 26). Applying a tuning voltage Vc = Vc+ - Vcref results in an electrostatic force Fel decreasing the electrode gap del.

2

20

2 el

celel d

VAF

(7)

With this the drive capacity is increased from ≈50 pF in passive state (Vc=0 V) to ≈65 pF at maximum modulation. Additionally a parasitic parallel capacity of 1 nF needs to be considered. In the case of steady state, a typical non-linear characteristic curve is received (see figure 25).

3,0

3,2

3,4

3,6

3,8

4,0

4,2

4,4

0 5 10 15 20 25 30 35Vc [V]

CW

L [

µm

]

Fig 25: Typical steady-state control characteristic for LFP-3041L-337

The polarity of the control voltage needs to be maintained, even as in equation (7) this doesn’t seem to be necessary. The circuit points Shield, Substrate and Vcref should be on the same stabilized, low-impedance potential. Spikes, a ripple voltage and other interfering signals at these circuit points can cause cross talk to the pyroelectric detector components by parasitic capacitances.

Fig 26: Example of circuitry for detector and filter operation (±18 V bipolar supply)



The movable reflector of the FPI is a mass-spring-system, which is progressively damped by the air cushion in the gap. The non linear fraction in equation (7) effects a decreasing stiffness of the whole system with increasing modulation, i.e. smaller gap. This behavior leads to several effects: The filter exhibits an acceleration sensitivity but vibrations will be damped due to the mechanical low pass characteristic. In steady state a dependence of the CWL on the position related to the gravitation field is observed (see figure 27). Both effects are dependent on the actual mirror position.

10

15

20

25

30

35

40

3 3,2 3,4 3,6 3,8 4 4,2 4,4CWL [µm]

CW

L [

nm

]

lower limit

upper limit

LFP-3041L-337CWL shift at turn around in the gravitation field

10

15

20

25

30

35

40

3,9 4,1 4,3 4,5 4,7 4,9 5,1CWL [µm]

CW

L [

nm

]

lower limit

upper limit

LFP-3950L-337CWL shift at turn around

in the gravitation field

Fig 27: Position dependency of the CWL, typical values and tolerance zones of LFP-3041L-337 (left) und LFP-3950L-337 (right)

The filter shows a stability limit at the so-called pull-in point. This should never be exceeded during

operation, otherwise the filter could be damaged.

As a guideline for steady state can be given: Don’t exceed the control voltage for maximum modulation

(e.g. CWL=3000 nm for LFP3041L-337) for more than 0.5 volts. This is an individual value for each

device.

The transient response of the filter is non-linear. For shorter wavelengths the system reacts slower,

because the total stiffness of the system is low and the air damping is high. The stated settling time (see

figure 28) is defined as the necessary time to achieve the final value of the CWL with a tolerance of

±1 nm for a control voltage step.



Time response of LFP-3041L-337

0

50

100

150

200

250

300

350

400

3000 3200 3400 3600 3800 4000 4200 4400

Wavelength [nm](Example: For a jump from 4200nm to 3400nm the settling time will be 170ms)

sett

ling

tim

e [m

s](f

inal

val

ue

of

the

CW

L w

ith

in ±

1nm

to

lera

nce

)

reaching 3000nm downwards

reaching 3400nm downwards

reaching 3800nm upwards

reaching 4200nm upwards

Fig 28: Transient response for LFP-3041L-337, typical values

The FP filter also shows a significant temperature dependency. As a temperature change mainly results

in a mechanical detuning of the filter, the correlation between HPBW and CWL remains unchanged.

LFP-3041-337Tolerance field TC of CWL

2,2

2,6

3,0

3,4

3,8

3,0 3,5 4,0CWL [µm]

TC

CW

L [

nm

/K]

LFP-3950-337Tolerance field TC of CWL

2,2

2,6

3

3,4

3,8

3,9 4,3 4,7 5,1CWL [µm]

TC

CW

L [

nm

/K]

Fig 29: Temperature coefficients of the CWL, typical values and tolerances for LFP-3041L-337 (left) and FP-3950L-337 (right)



4.4 Operation modes and measurement methods

The capabilities of LFP (so called variable color) detectors are numerous. Depending on the measurement task and operation mode, different advantages compared to conventional single or multi channel detectors with fixed NBP filters can be found. Hereafter three different operation modes will be explained in detail:

Sequence of channels In the most simple case several fixed detector channels shall be substituted by a tunable detector. The filter is sequentially adjusted to the individual spectral channels. Beside the obvious advantage of the flexibility and expandability in the channel choice (in the region (3 … 5) µm) additional advantages may be achieved: Simple multi channel detectors have separated apertures, which yield to the well known issues regarding

non-uniform illumination, long-term stability, source drifting, pollution, etc. The variable color detectors

don’t show these problems due to their principal design and singular light path.

Detectors with an internal beamsplitter also have a common aperture, but each channel is getting only a

fraction of the whole radiant power. Applying the sequential measurement we can always use the whole

incident radiant power. For four different channels and comparable conditions regarding aperture size

and filter bandwidth theoretically a duplication of the SNR can be reached.

Step scan The method described above can still be expanded in such a way that continuous spectra can be obtained. The required acquisition time for the mapping of a spectrum depends on the following facts: 1. Number of measuring points (wavelength range, step size):

To get a continuous spectrum it must be scanned at minimum with a step size which corresponds to the half filter bandwidth (sampling theorem). Moderate oversampling can be useful. The reasonable step size is in the range (10 … 50) nm.

2. Recordings of the measuring points (modulation frequency, integration time): Recordings of the measuring points (modulation frequency, integration time): These parameters define the SNR. Beside the detector properties and the applied analysis methods, the radiant power, modulation depth of the IR source and the design of the measuring section are crucial.



Settling time of the filter: The actual settling time of the filter depends on the wavelength as described earlier. It should therefore be implemented variably to achieve an optimum of speed.

Transmission characterizationof Polystyrene foil

0

20

40

60

80

3000 3200 3400 3600 3800 4000[nm]

Tra

nsm

issi

on

[%

] LFP-3041L-337

FTIR 8/cm

Fig 30: Measurement example for the step scan mode (Polystyrene foil) LFP-3041L-337: spectral resolution

R=65; SNR1000:1; 100 data points; acquisition time 10 s

Continuous scan By using a pyroelectric detector only modulated radiation can be analyzed. Normally this is realized by mechanical chopping or electrical modulation of the IR source. If the filter is however continuously scanned the spectral information can be used directly for the modulation. The filter is actuated dynamically in this case. This particular operation mode has principally the potential to accelerate the recordings of spectra remarkably. The earlier mentioned non-linear effects during dynamic operation however need to be considered separately. In most cases it is not correct to consider the filter as a linear system with low pass behavior even in a limited operation range. Two basic approaches are possible for a dynamic operation: Presetting of voltage characteristics Vc(t) for filter modulation (sinusoidal, ramp or similar…) and

Detection of the resulting characteristics of the CWL by an adequate calibration for example with

wavelength standards (e. g. NBP filters) or

Evaluation of the detector signal with dedicated software algorithms (chemometric techniques)

Presetting of a designated progression of the CWL(t) and determination of the compatible voltage

characteristic Vc(t) for example with wavelength standards



85%

90%

95%

100%

105%

0 100 200 300 400 500 600 700scan time [ms]LFP-3041L-337

No

rmal

ized

FP

F d

etec

tor

sig

nal

40

60

80

100

120

3100 3300 3500 3700 [nm]FTIR

Tra

nsm

issi

on

[%

]F

TIR

LFP-3041L-337

FTIR 4/cm

Transmission characterization of 2.2 Vol% Methane

Fig 31: Measurement example for the continuous scan mode with dynamic filter tuning (Methane)

Figure 31 gives an example for the dynamic operation. The IR source is working in DC operation, while the filter goes through the desired wavelength range. Except for the DC-portion, the whole spectral information is contained in the generated detector signal. The actuation and analysis has to include both the dynamic properties of the filter and the detector.

4.5 Summary

With the extension of our product range by variable color detectors additional technologies are available for our customers. All types of our multispectral detectors are complementing one another: Conventional dual and quad channel detectors can be used in competitive volume applications

Our dual and quad channel beamsplitter detectors with one aperture are used as long term stable and

very accurate measuring modules for different spectral channels

Variable color detectors with a high SNR allow a more flexible operation of the analyzer enabling for

example the detection of adjoining or overlapping absorption bands. They are also of interest for

applications, where more than 4 spectral channels shall be scanned within a short time frame.

Datasheet - Detector Accessory

Evaluation kit for Fabry-Perot detectorsPurpose:

The kit supports customer needs for an initial test of InfraTec s Fabry-Perot detectors (FP) without having to develop test circuitry andsoftware themselves. It allows easy control of detector and IR sourcewith customized software to analyze the detector signals. With the helpof this kit a quick and easy configuration of a simple FP spectrometer ispossible.

LFP-3041L-337 or LFP-3950L-337

FIGURE

CONTENT AND FEATURES OVERVIEW

It includes a basic board and a FP detector board with USB Interface, USB cable, ribbon cable, CD-ROM with testand measurement software, USB driver and a manual. The following features are integrated:

· Activation of FP detector and analysis of signals· Activation of IR source· Extensive configuration options including mappings of complete spectra by use of the Fabry-Perot

evaluation workbench software

A version of this kit with one mounted LFP detector (LFP-3041L-337 or LFP-3950L-337) and a MIRL 17-900 IRsource is also available on request.

SPECIFICATION REQUIREMENTS

· Power supply: 12 V DC· Current consumption: 150 mA (without IR source)· Integrated detector supply: ±5 V· Integrated FP-control voltage generation: 0-35 V; 12 bit· Control of IR source up to 800 mA DC to 100 Hz

(square, 2-98 % duty cycle)· Signal acquisition with 12 bit, 1 kHz, analysis with

FFT-technique· RoHS conformity

· PC with Windows 2000, XP or Vista· USB 2.0 Interface· Power supply 12 V DC (300 mA Minimum)· FP detector and IR source


Datasheet - Detector Accessory

Evaluation kit for Fabry-Perot detectors

PERFORMANCE DETAILS

User interface of evaluation software; Transmittancespectrum of a polystyrene foil measured in scan mode

Signal voltage of LFP detector @10 Hz modulationfrequency and constant control voltage

Electronics and software of the Fabry-Perot evalua-tion kit are designed to give the user full access toperformance and tuning capabilities of InfraTec snew variable color products.

The basic operation principles are very similar asnormally applied for conventional single or multi-color pyroelectric detectors: Modulation frequency,duty cycle and driving current of the IR source canbe configured by software. The detector signal canbe displayed real-time and recorded with 12 bitresolution and 1 kHz sampling frequency. Evalua-tion of the RMS amplitude is implemented by a FFTalgorithm.

In addition to that, the electronics can providevoltages up to 35 V with 12 bit resolution to controlthe tunable filter. To easily create calibration files forthe tuning, characteristics of individual detectors arealso supported by the software. Latency for filtersettling between measurements at different wave-lengths can be configured, too.

In the basic operation mode the filter can bemanually set to any wavelength within the tuningrange. Measurements can continuously be recordedinto a log file.

The more advanced Sequence mode is intended tocompare the tunable detector with conventionalmultispectral solutions like filter wheels or multicolordetectors. The filter is periodically adjusted to a setof wavelengths, which can be predefined by theuser.

The Scan mode is used to obtain complete spectrawith a designated spectral range and step size.Measurements can be displayed and saved as rawsignal spectra or as transmittance spectra if apreviously measured background spectrum wasdefined as reference.

The software also supports the new innovativecontinuous sweep mode. This particular operationmode has principally the potential to accelerate therecordings of spectra remarkably. Here the IRsource is DC-driven, but the filter is continuouslyscanned and so the spectral information can beused directly for the modulation. With the assis-tance of the software waveform, tuning speed andwavelength range can be adapted to find the mostsuitable operation. Recordings can be made withthe integrated logging function, so external softwarecan be used for further signal analysis.

Please find more details in the evaluation kit manualand in our Fabry-Perot detector application note.

Sweep mode measurement of a polystyrene foil;detector signal and control voltage

© InfraTec GmbH, May 1st 2009


Expansion of the InfraTec product family New angular resolving pyroelectric flame detector PIA-903-X002

© InfraTec 10 (All the stated product names and trademarks remain in property of their respective owners.) Page 1

As a leading supplier of tailor-made pyroelectric detectors InfraTec has shown extensive expertise in producing a wide range of single and multi-color pyroelectric detectors for gas analysis and flame detection. With our standard pyroelectric detectors (e.g. single detector LME-551 or quad detector LMM-244) it is possible to detect a flame and its composition in a distance of more than 50 meters, but until now the position of the flame inside the field of view (FOV) was not resolvable. As an upgrade to standard tri-band flame detectors, the new PIA-903 detector with 3 x 3 element array and integrated optics is able to provide additional spatial information about the observed scene for a rough angular location of the flame. Calculating the signal ratios of all channels offers an interpolation of the angular position inside of a 100° conic FOV and in combination of two or more flame sensor units a 3D coordinate of a fire pocket. Instead of the Lithium Tantalate (LiTaO3) pyroelectric material that InfraTec is generally using for the high performance measuring application, here a PZT thin film technology is employed to build the active elements. It allows very small chip sizes which enable us to fully illuminate 9 (3 x 3) individually addressable elements by a 1 mm diameter beam. A built-in lens produces an intra-focal image of the flame, which allows the interpolation between the individual pixel signals and results in a remarkable angular resolution of 5° or even better. Together with a 9 channel transimpedance amplifier (current mode operation) inside of the housing a high Responsivity of 8000 V/W at 10 Hz is realized. Even for a pixel size of only (500 x 500) µm a high D* of >1.5E108 cm√Hz/W (500 K, 10 Hz, without window and optics) can be guaranteed. Although D* is three times lower as for our LiTaO3 detectors, it is still clearly higher compared to low cost detectors for motion detection or Thermopile arrays.

-45° +45°

-45°

+45

°

0°

0°

Y

X

Fig. 1: Monitoring a production area Fig. 2 : Optical scheme Fig. 3: 3x3 Sensor array

In Fig. 1 the principle arrangement of the detector is illustrated. The position of the light spot on the sensor, a blurred image of the flame (Fig. 6), can be interpolated with a very simple formula, which is normally applied with analog quadrant detectors but can be easily adopted to a 3x3 sensor array. In conjunction with the integrated lens the spot position is directly related to the incidence angle of the detected flame. Fig. 4, 5 & 6 illustrate the principle of angular determination by interpolation simplified to the one-dimensional case (3 sensing elements). The calculated virtual X coordinate is strictly monotonic within the range of ± 50°, in which the resolution is mainly limited by the signal to noise ratio of the measurement.


Expansion of the InfraTec product family New angular resolving pyroelectric flame detector PIA-903-X002

© InfraTec 10 (All the stated product names and trademarks remain in property of their respective owners.) Page 2

Fig. 4: Measured angular response

Fig. 5: Interpolated x position

Fig. 6: Images of the flame with varying incidence angle in

one direction (azimuth angle)

If this principle is applied to the two-dimensional case the x and y coordinates of the spot position over both angle directions become nearly flat planes (Fig. 7) within a range of about 100°, which indicates a good linearity. An angle resolution of about 5° or better might be achieved, which corresponds to a virtual resolution of about 20x20 pixels.

CBAAAACCCX 321321

CBACBACBAY 333111

.

Fig. 7: Interpolated x position plotted over both angle

directions

Fig. 8 : Determination of the flame position by means

of two PIA-903 detectors

If two PIA-903 detectors are positioned at different locations in one area it is possible to define the 3D coordinate of the flame. This might be helpful for a fast reaction in large production facilities or other places. The principle is illustrated in Fig. 8. InfraTec continues developing new pyroelectric detectors to better support customer needs. The PIA-903 detector is the first model of a new InfraTec product line. It was designed to fill a very useful spatial “guard” function in any kind of flame detector. From this many other applications are possible. Any further ideas, remarks or questions are sincerely appreciated by InfraTec.

www.lasercomponents.com 12/10 / V2 / HW / infratec/multicolor/ pia-903-app-note.pdf



Temperature Behavior

1 Introduction

Detectors that measure radiation by means of the temperature change of an absorbing material are classified as thermal detectors. Thermal detectors respond to any wavelength radiation that is absorbed, and when an appropriate absorbing material or ‘blackening’ is applied to the detector element surface, they can be made to respond over a wide range of wavelengths. Thermal detectors can be classified by their output into a number of categories. One of the most commonly used in infrared (IR) applications, is the pyroelectric detector type that derives its output from changes in the electrical polarization of certain crystals with temperature.

Commercially available single crystal pyroelectric materials have been developed which generate reasonable electric charges for small temperature changes, and at the same time they are stable, uniform and durable. It is this development that has made practical the large scale production and application of cost effective, high performance pyroelectric infrared detectors into a wide variety of commercial, industrial and military applications. As the differing and diverse environmental conditions can adversely affect pyroelectric detector stability and performance, it is extremely important for IR system designers to understand how system requirements can best be translated to optimize detector specifications, and to acquire some basic knowledge of detector characteristics and construction.

As for any IR system, including those using pyroelectric detectors, system designers must determine the magnitude, as well as the spatial characteristics, spectral content and frequency of incoming radiation. An optical system that will deliver this radiation to the detector must then be designed. Making this basic information available to the detector manufacturer will generate the best suitable detector configuration regarding cost and performance. Pyroelectric detectors can be used over a wide ambient temperature range without cooling or controlling the detector temperature. As they are typically manufactured from generally available single crystal materials, they are cost effective, and can readily be made into many detector configurations and sizes. In addition, detectors do not require special power supplies and detector signal processing and noise limited operation can be achieved readily with commercially available operational amplifiers (OpAmps) and relatively simple circuits.

Pyroelectric detectors will only respond to changes in incoming radiation levels. In addition, because they are constructed from piezoelectric materials, they will all respond to mechanical vibration to some extent although any resulting microphonic noise can be minimized by proper detector design and assembly techniques. Since pyroelectric detectors are often being used to detect small temperature changes, their performance can be affected by external temperature changes such as those caused by air movements. It is, therefore, particularly important for the system designer to understand how these changes affect detector performance.

This application note is offered to provide some insight into how temperature affects the various components that comprise a typical pyroelectric detector and how the effect of ambient temperature variations on detector performance can be minimized. Ambient temperature variations cause drift of detector signal and noise, and fluctuations or instability in detector output. These two effects should always be considered separately since they are caused by factors that require different corrective measures in detector design.



2 Temperature dependent Components within the Pyroelectric Detector

Figure 1 shows the basic commercial single element pyroelectric detector offered by InfraTec GmbH. In the figure there are the temperature dependent active and passive electrical and optical components indicated.

IR Filter

JFET

Feedback resistorand capacitance

Gateresistor

Compensation element

Active element

usual JFET circuitryJFET = Junction Field Effect Transistor

state-of-the-artCMOS OpAmp circuitryOpAmp = Operational Amplifier

OperationalAmplifier

Fig 1: Typical InfraTec detector construction

2.1 Temperature dependence of the pyroelectric chip

To varying extents, components labeled in figure 1 contribute to both drift and fluctuation, although detector instability due to varying temperatures at fixed operating conditions can generally be attributed mainly to the physical qualities of the detector chip itself. In a typical application, the active element of the pyroelectric detector is coated with an appropriate ‘black’ coating to enhance the absorption of chopped or modulated incoming radiation, in the wavelength range of interest defined by an IR filter, thus warming the pyroelectric crystal. Depending on the operating mode of the detector, the resulting current or voltage generated above the capacitance of the detector chip (typically 50 pF) is then converted into useful signal. The electrical charges (current) generated even by minimal increases in crystal temperature are on the order of pico- to nanoamperes. It is important to note that an increase in detector case/package temperature can produce ‘false’ signals. Electrostatic peaks of up to 140 V, for example, have been measured from highly insulated detectors with case temperature increases as small as 10 °C.

2.1.1 Temperature drift/temperature coefficient

By definition, the Curie point is that temperature above and beyond which pyroelectric crystals cease to exhibit pyroelectric effects. The pyroelectric coefficient of these materials, which is a measure of eventual detector sensitivity, actually increases with increasing temperature up to the Curie point. In the case of Lithium Tantalate (LiTaO3), from which almost all of InfraTec’s pyroelectric detectors are made, the Curie temperature is about 600 °C. In most practical applications where operating ambient temperatures are substantially lower (<+120 °C), the intrinsic temperature coefficient (TC) of the pyroelectric current is approximately +1500 ppm/K (+0.15 %/K).



This is particularly important to know in system applications where the ambient temperature is changing, and this characteristic will manifest itself as detector ‘drift’. Since the compensation element in a thermally compensated detector does not contribute to signal, this TC also applies to compensated detectors. Standard applications never analyze the pyroelectric current from the active element itself but the signal voltage generated by the pyroelectric current (voltage responsivity) is analyzed. Therefore the resulting TC of the pyroelectric chip depends on the operating mode of the detector either the element in the open-circuit-operation (voltage mode detectors, JFET circuitry) or as short circuit (current mode detectors, OpAmp circuitry). Only the TC of the pyroelectric current has an effect if the short circuit mode is applied but the TC will be overlaid by the relative permittivity of the Lithium Tantalate using the open-circuit operation.

2.1.2 Effects of temperature fluctuation on detector stability

As stated earlier, if not accounted for in the pyroelectric detector design, a change in the detector case temperature alone will usually generate a very low frequency signal, which can be of substantial magnitude. The commonly used voltage mode detectors with JFET circuitry are particularly susceptible to these temperature changes. Signals generated can cause detrimental effects to a chosen preamplifier by pushing the limits of its operational range. The longer the thermal and electrical time constants of these detectors, the more sensitive they are to these changes. The state-of-the-art CMOS OpAmp detectors present in temperature ramps a DC offset adjustment in the same dimension but the wanted signal is 100-times higher. That means that the ratio DC offset adjustment / signal will be lower than 1 % of the JFET circuitry and hence the dynamics range will not be limited anymore. Since thermal and electrical time constants are determined by detector construction, they can be reduced by design to attenuate, but not completely eliminate, the effects of expected case temperature changes in uncompensated detectors.

0

500

1000

1500

2000

2500

20 25 30 35 40 45 50 55 60Case temperature [°C]

Off

set

volt

age

[mV

]

0,0

0,5

1,0

1,5

Tem

per

atu

re r

amp

[K

/min

]80GOhm gate resistor

10GOhm gate resistor

temperature ramp

Offset voltage of an uncompensated voltage mode detector at various gate resistors during a temperature ramp

Fig 2: Offset voltage vs. temperature at various gate resistors (uncompensated voltage mode detector, JFET circuitry)



Both can be tailored to meet specific operating conditions although it should be noted that varying the electrical time constant of a detector is not as complex as varying the thermal time constant of a detector. In the InfraTec product line there are only the extended detectors in series LIE-312 and LIE-332 offered with a thermal time constant other than standard. Figure 2 shows the typical effect of temperature variation on detector offset voltage for uncompensated detectors with different gate resistors. Please note that the noise is inversely proportional to the sq. root of the intended resistor decrease, e.g. a 9-times higher stability will cause a decrease of detector detectivity down to 33 %. To overcome the problems associated with detector instability due to case temperature changes, an additional optically inactive detector chip, known as the compensation element, may be added inside the package. The detector is then called a compensated detector. The compensation element will have negative (or the opposite) polarity of that of the active element. As case temperature varies, charges generated at the active and compensation element will essentially have the effect of canceling each other. Figure 3 showing the offset voltage vs. temperature of thermally compensated detectors, clearly presents the effectiveness of this arrangement.

0

1000

2000

3000

4000

5000

20 25 30 35 40 45 50 55 60Case temperature [°C]

Off

set

volt

age

[mV

]

0

1

2

3

4

5

Tem

per

atu

re r

amp

[K

/min

]

uncompensated

compensated

temperature ramp

Offset voltage of a compensated and an uncompensated voltage mode detector during a temperature ramp

Fig 3: Offset voltage vs. temperature (compensated and uncompensated voltage mode detectors, JFET circuitry)

Thermally compensated pyroelectric detectors can be classified into two types: a) serial or b) parallel compensated, depending on the electrical connection of the active and the compensation element (see figure 4). The compensation element is completely optically shielded from incoming radiation, hence although optically inactive but still an electrically effective capacitor. The influence of the compensation element regarding signal and detectivity is different between the common JFET circuitry and the state-of-the-art CMOS OpAmp circuitry.



The net effect of compensation for usual JFET circuitry is reduction of detectivity to approximately 60 % for both compensation types, although signal and noise will be affected differently for each type. The approximate values of signal and noise for each type of compensated detector are shown as a percentage of those of an uncompensated detector in the figure above. Due to the virtual short of compensation and pyroelectric element in the state-of-the-art OpAmp circuitry the parallel compensation neither causes signal nor detectivity losses.

Serial Compensation Parallel Compensation

+

-

+

-

Serial Compensation (Only with InfraTec’s LIM-314) JFET circuitry: 100 % signal; 170 % noise OpAmp circuitry: not reasonable Parallel Compensation JFET circuitry: 50 % signal; 70 % noise OpAmp circuitry: 100 % signal; 100 % noise

Fig 4: Serial and parallel compensation to increase the stability of the DC operating point of pyroelectric detectors in temperature ramps

The decision to use serial or parallel compensation is determined for JFET voltage mode detectors by the operating condition from the detector. Parallel-compensated detectors are more stable at strong (more than 2 K/min) or long time constant temperature ramps (longer than 1 hour). On the other hand the user may prefer the double signal of the serial-compensated detector. We always recommend testing parallel-compensation first. Due to the availability of the state-of-the-art OpAmp detectors it is no longer necessary to compromise between signal magnitude and stability.

2.2 Temperature dependence of gate resistor as well as feedback components

High value resistors used as gate resistor in JFET circuitry and feedback resistor in OpAmp circuitry are characterized by high negative temperature coefficients (i.e. resistance decreases with increasing temperature). (82 and 100) GOhm resistors that are routinely used with certain types of pyroelectric detectors, for example, typically have TC’s of (-1500 ... -2000) ppm/K. NPO capacitors as well as printed strip lines are used in the feedback capacitance of the OpAmp detectors and hence there is no measurable temperature influence on the detector parameters.

2.3 Temperature dependence of integrated JFET and CMOS operation amplifier

The gate leakage current, as well as the current noise of integrated JFETs, increase substantially with increases in temperature. Particularly at the higher temperatures, this increase is exponential. The gate leakage current of a typical FET is normally less than 1 pA at +25 °C, but can increase to well over 100 pA at +125 °C.



CMOS OpAmps use a insulated input which enables an ultra low input bias current and low noise. The OpAmp circuitry is best suited at high temperatures up to +100 °C (see figure 5).

0

1

10

100

1000

1 10 100 1000Frequency [Hz]

No

ise

den

sity

[µ

V/H

z½]

JFET 85°C

CMOS 25°C

CMOS 85°C

JFET 25°C

Noise of LME-351 (CMOS) and LME-300 (JFET+external OPA277) detector design(both with identical feedback resistor of 5GOhm)

Fig 5: Different behavior of JFET and CMOS input transimpedance amplifiers at temperature increase

Another temperature effect can be created by the amplification drift of JFET or OpAmp. Increases in temperature will reduce the JFET common-source forward transconductance and increase the pinch-off voltage. Due to the high open-loop amplification and strong negative feedback of the OpAmps a drift of the OpAmp circuitry amplification determined by the temperature is not observed.

2.4 Temperature dependence of IR filters and windows

InfraTec pyroelectric detectors are coated with a black absorption coating whose spectral characteristics have been shown not to be temperature dependent in the storage and operating range. Please note that the maximum storage and operating temperature for detectors with metal black layer (applies only to LIE-312 and LIE-332 used for IR-spectrometer) is only 60 °C. Above this temperature the high absorption foamy structure of the metal black layer will irreversibly sinter, which will cause a reduced absorption capability. To varying degrees, however, the spectral transmission of all windows and optical elements / filters fitted as part of detector package will vary with temperature. For uncoated IR optical materials (e.g. CaF2), temperature variation will mainly affect their transmission and absorption edge (hence their useful range). A signal drift up to -400 ppm arises e.g. for CaF2 windows.



The spectral thermal shift and changes in transmission characteristics of coated filters, however, is determined by the materials and coating design used to manufacture such filters, and therefore vary by the type of filter (e.g. narrow band pass - NBP, wide band pass - WBP), wavelength region and filter manufacturer. Depending on the application InfraTec uses IR filters manufactured with a low TC design (typical temperature shift +0.22 nm/K, typical angle shift @ 15° -27 nm, see figure 6) or a low angular shift design (typical temperature shift +0.40 nm/K, typical angle shift @ 15° -12 nm, see figure 7). All integrated narrow bandpass filters in InfraTec pyroelectric detectors for the 3 to 5 micron region have been chosen for their stability in design and are typically deposited on silicon substrates.

Temperature shift of NBP-filters "CO2 standard"

0

20

40

60

80

100

4000 4100 4200 4300 4400 4500

Wavelength [nm]

Tra

nsm

issi

on

[%

]

15°C

55°C

CO2 standard [D]NBP 4.26µm/180nm

Temp.shift <0.28nm/K

Fig 6: Temperature shift of NBP filter with low TC design

Temperature shift of NBP-filters "CO2 high AOI"

0

20

40

60

80

100

4000 4100 4200 4300 4400 4500

Wavelength [nm]

Tra

nsm

issi

on

[%

]

15°C

55°C

CO2 high AOI [Z]NBP 4.27µm/170nm

Temp.shift <0.45nm/K

Fig 7: Temperature shift of NBP filter with low angular shift design

As a general rule, as temperature increases, IR bandpass filters will shift to longer wavelengths with some loss in band transmission. In addition, an angular shift results in a CWL shift to shorter wavelengths generally.



3 The Effects of Temperature Variation on overall Detector Performance

Accounting for the temperature dependence of the key components that constitute a complete detector assembly, some typical data can be offered on detector performance parameters that may be of particular interest to IR system designers.

3.1 Detector offset voltage (at a stabilized temperature – no temperature ramp)

Due to the increase in pinch-off voltage and gate leakage current, higher temperatures will always increase the offset voltage at the JFET circuitry because the negative TC of the gate resistor does not compensate this increase. Figure 8 shows the offset voltage vs. temperature of a typical, but representative, InfraTec detector with JFET circuitry. The DC offset voltage of the OpAmp circuitry is determined by the offset voltage of the OpAmp itself (typical in the range of ±1 mV) together with input bias current and feedback resistor. The bias current is typically +3 pA at 85 °C. Figure 9 shows the offset voltage vs. temperature of a typical, but representative, InfraTec detector with OpAmp circuitry.

0,75

0,80

0,85

0,90

0,95

1,00

1,05

1,10

20 30 40 50 60 70 80 90Temperature [°C]

Off

set

volt

age

[V]

Ch1 NBP 3.40µm/120nm [G] @ 4Hz

Ch2 NBP 3.95µm/90nm [H] @ 4Hz

Offset voltage vs. temperature of LIM-222-GHdual channel detector 47kOhm source resistor

Fig 8: Offset voltage vs. temperature (JFET circuitry); after thermal transient period

0,1

1

10

100

20 30 40 50 60 70 80 90Temperature [°C]

Off

set

volt

age

[mV

]

Offset voltage vs. temperature of CMOS OpAmp detectors[5GOhm feedback resistor]

Fig 9: Offset voltage vs. temperature (OpAmp circuitry); after thermal transient period



3.2 Detector noise

An increase in gate leakage (JFET) or bias (OpAmp) current and the decrease in gate (JFET) or feedback (OpAmp) resistor value with an increase in temperature, results in higher detector noise. This increase in noise is particularly prominent at lower frequencies; see figure 10 below. As already discussed in 2.3 the noise arising from the JFET is considerably greater at higher temperatures compared to the OpAmp.

100%

1000%

10000%

20 30 40 50 60 70 80 90Temperature [°C]

Noise density (10Hz)JFETNoise density (10Hz)CMOS OpAmp

Noise vs. Temperature for a JFET and a CMOS input

Fig 10: Noise of a JFET and a CMOS OpAmp circuitry at various temperatures

3.3 Signal voltage (responsivity)

InfraTec pyroelectric detector signals are nearly linear with temperature over the most common temperature range for many applications (see figure 11 for JFET circuitry). Please note for TC evaluation that a common CaF2 window will create an additional TC of approx. -400 ppm/K.

10

20

30

40

50

20 30 40 50 60 70 80Temperature [°C]

Sig

nal

vo

ltag

e [m

V]

47KOhm, 4Hz

47KOhm, 10Hz

TC of 4Hz signal about -490 ppm/KCaF2 window included


Signal vs. temperature of LME-302-61 (500K black body)

Fig 11: Signal voltage vs. temperature at JFET circuitry (LME-302-61: voltage mode with CaF2 window)



Detectors with JFET circuitry and without windows have typical TC´s in the range of (-200 ... 200) ppm/K, because the TC´s of the single components are almost compensated. A current source instead of the source resistor increases the signal voltage by (10 … 15) % but an influence on the temperature dependency cannot be detected. The negative TC of the detector as shown in figure 11 is mainly caused by the CaF2 window. The temperature dependency of detectors with OpAmp circuitry is more significantly influenced by the modulation frequency. Above the electrical breakpoint the resulting TC becomes more and more positive and whereas below the breakpoint more and more negative. The detector LME-351-61 has a comparatively high electrical breakpoint of 160 Hz due to the feedback resistor of 5 GOhm. The modulation frequencies shown in figure 12 are much smaller than 160 Hz and cause without CaF2 window a TC of the detector between +300 ppm/K (@ 60 Hz and -400 ppm/K @ 4.6 Hz). Including the CaF2 window with a TC of -400 ppm/K the detector has a TC of -100 ppm/K @ 60 Hz and -800 ppm/K @ 4.6 Hz as shown in figure 12.

240

290

340

390

440

20 30 40 50 60 70 80Temperature [°C]

Sig

nal

vo

ltag

e [m

V]

4.6Hz46Hz60Hz


TC of 4.6Hz signal about -800 ppm/KCaF2 window included


Signal vs. temperature of LME-351-61 (500K black body)Electrical time constant 1ms

Modulation frequency << limit frequency

Fig 12: Responsivity vs. temperature at OpAmp circuitry (LME-351-61: current mode with CaF2 window)

The detector LIE-245-10 has a noticeable greater electrical time constant of 16 ms (24 GOhm//0.68 pF). The electrical breakpoint is between 4.6 Hz and 10 Hz. TC behavior changes around these limits as shown in figure 13.

520

540

560

580

600

620

640

660

680

700

720

20 30 40 50 60 70 80Temperature [°C]

Sig

nal

vo

ltag

e [m

V]

4.6Hz

7Hz

10Hz

TC of 4.6Hz signal about -390 ppm/KSi ARC window included

TC of 7Hz signal about +110 ppm/KSi ARC window includedTC of 10Hz signal about +350 ppm/K

Si ARC window included

Signal vs. temperature of LIE-245-10 (500K black body)Electrical time constant 16ms

Modulation frequency around limit frequency

Fig 13: Responsivity vs. temperature at OpAmp circuitry (LIE-245-10: current mode with Si ARC window)



4 Summary

With this application note, InfraTec has attempted to provide IR system designers with some useful data regarding the effect of ambient temperature variation on pyroelectric detector performance due to the complex interaction of components that form just part of the detector assembly. As pyroelectric detectors are designed and incorporated into a specific application, however, designers must also consider how other system components (external to the detector) such as sources, optics etc. can also alter the expected detector performance. The interrelation of the temperature coefficients for single components of a detector are also influenced by the signal conditioning method of the software used. Data from a simple set-up is included here to demonstrate the importance of understanding how system parameters interact. The temperature behavior of responsivity is presented in three figures (also applies to the signal voltage measured directly at the detector).

94%

96%

98%

100%

102%

104%

20 30 40 50 60 70 80Temperature [°C]

Res

po

nsi

vity

RV/R

V 2

5°C

Resulting TCratio Ch1 / Ch2@ f>1Hz

Average TC ofresponsivity@ f>1Hz

TC RV: -280ppm/K

TC RV: -810ppm/K

Temperature coefficient (TC) of responsivityLIM-222-GH (JFET), 500K black body

Fig 14: Dual channel detector LIM-222-GH (TO39 housing; small chip size; thermal compensation; JFET; voltage mode; ch1: NBP 3.40 µm/120 nm HC; ch2: NBP 3.95 µm/90 nm Ref.), IR source 500 K black body.

This figure illustrates a configuration which generates a slightly negative average TC of -810 ppm/K (channel 1 = -950 ppm/K, channel 2 = -670 ppm/K), at an operating temperature of 60 °C which results in a 3 % signal loss compared to room temperature (black curve). Applying the ratio between the signal of the gas channel (channel 1) and the reference channel (channel 2) the impact of aging and pollution in the optical path of the gas being evaluated can be compensated (red curve). According to the Theory of Errors the resulting TC emerges from the ratio for TCch1 - TCch2 (in figure 14 (-950 ppm/K) - (-670 ppm/K) = -280 ppm/K). In principle, positive and negative values can emerge for the resulting TC through this ratio.



98%

100%

102%

104%

20 30 40 50 60 70 80Temperature [°C]

Res

po

nsi

vity

RV/R

V 2

5°C

Average TC ofresponsivity@ f>20Hz

Average TC ofresponsivity@ f<5Hz

Resulting TCratio Ch1 / Ch2@ f>20Hz

Resulting TCratio Ch1 / Ch2@ f<5Hz

TC RV: +580ppm/K

TC RV: +140ppm/K

TC RV: +20ppm/K

TC RV: -220ppm/K

Temperature coefficient (TC) of responsivityLIM-262-GH (OpAmp), 500K black body

Fig 15: Dual channel detector LIM-262-GH (TO39 housing; small chip size; thermal compensation; OpAmp; current mode; feedback 100 GOhm; ch1: NBP 3.40 µm/120 nm HC; ch2: NBP 3.95 µm/90 nm Ref), identical filter placement and IR source as with the LIM-222-GH in figure 14.

The CMOS OpAmp equipped LIM-262-GH has been tested under the same arrangement and conditions as the LIM-222-GH. The current mode detector’s typically positive TC of responsivity was measured. The TC is in the range of (+140 ... +580) ppm/K, depending on the modulation frequency. Again in this case, the resulting TC is calculated from the Channel 1/Channel 2 ratio as a difference of TCch1 - TCch2. Depending on the TC value of the single channel, positive and negative values can emerge, based on the calculated difference.

90%

92%

94%

96%

98%

100%

102%

104%

20 30 40 50 60 70 80Temperature [°C]

Res

po

nsi

vity

RV/R

V 2

5°C

Resulting TCratio Ch1 / Ch2

Average TC ofresponsivity

TC RV: -290ppm/K

TC RV: -1250ppm/K

Temperature coefficient (TC) of responsivityLIM-222-GH (JFET), T1 lamp 6V/115mA

Fig 16: Dual channel detector LIM-222-GH identical to figure 14 except for IR source which is a T1 lamp 6 V/115 mA



The temperature dependency of an identical dual channel detector LIM-222-GH when illuminated with a T1-type miniature light bulb 6 V/115 mA is presented in figure 16. This thin-walled light bulb is often used in gas analyzers. The modulation frequency is limited to 4 Hz by thermal time constant of the lamp filament. Only the single channels change with respect to the 500 K black body illumination (-1250 ppm/K instead of -810 ppm/K). Beside the lamp glass transmission, additional factors like modulation of the lamp and positioning accuracy of the optical channel influence the TC. The resulting TC (channel 1/channel 2) = TCch1 - TCch2 is very comparable to the one with the 500 K black body irradiation (see figure 14). The additional temperature drift caused by the light bulb is being eliminated by the ratio. From concept to production, InfraTec constantly strives to produce pyroelectric detectors with the highest quality and best performance on the market today for any application. All standard InfraTec pyroelectric detector packages are hermetically sealed and backfilled with very dry nitrogen for stability and long term reliability. Given the importance of translating system requirements into detector specifications, we take particular pride in our responsiveness and close technical interface with our customers. We invite your inquiries.


JFET and Operational Amplifier Characteristics

Pyroelectric detectors of InfraTec use for the first signal processing stage Junction Field Effect Transistors (JFET) and CMOS Operational Amplifiers (OpAmp) built-in in the detector housing. Specifications are adapted for high-impedance pyroelectric elements.

1 Standard JFET - for single and multi color detectors

Features Very low voltage and current noise

High input impedance

Full performance from low-voltage power supply, down to 2.5 V

Low leakage for improved system accuracy

Absolute maximum ratings Gate-Source / Gate-Drain voltage: -50 V

Power dissipation: 50 mW

Specifications (TA = 25 °C unless otherwise noted)

Limits Parameter Symbol Test Condition

Min MaxUnit

Static

Gate-Source Breakdown Voltage -V(BR)GSS IG = -1 µA, VDS = 0 V 50 -

Gate-Source Cutoff Voltage -VGS(off) ID = 0.1 µA, VDS = 15 V 0.4 1.5V

Saturation Drain Current IDSS VDS = 15 V, VGS = 0 V 0.5 1.5 mA

VDS = 0 V, VGS = -30 V, TA = 25 °C - 2 pAGate Reverse Current -IGSS

VDS = 0 V, VGS = -30 V, TA = 150 °C - 10 nA

Gate Operating Current -IGSS ID = 0.1 mA, VDG = 15 V - 2

Drain Cutoff Current ID(off) VDS = 15 V, VGS = -5 V - 50pA

Gate-Source Forward Voltage VGS(F) IG = 1 mA, VDS = 0 V - 0.8 V

Dynamic

Common-Source Forward Transconductance

gfs 0.7 2.1 mS

Common-Source Output Transconductance

gos

VDS = 15 V, VGS = 0 V, f = 1 kHz

- 12 µS

Drain-Source On-Resistance rDS(on) VDS = 0 V, VGS = 0 V, f = 1 kHz - 2000 Common-Source Input Capacitance

Ciss - 7

Common-Source Reverse Transfer Capacitance

Crss

VDS = 15 V, VGS = 0 V, f = 1 MHz

- 3

pF

Equivalent Input Noise Voltage en VDS = 10 V, VGS = 0 V, f = 1 kHz - 8 nV/Hz



2 Special JFET - design for single and multi color detectors (on request)

Benefits Reliable operation at high temperature or increased ionizing radiation

Disadvantage related to standard JFET Lower gain; higher output impedance

Features Ultra high input impedance

Ultra low leakage for improved system accuracy

Absolute maximum ratings Gate-Source / Gate-Drain voltage: -50 V


Specifications (TA = 25°C unless otherwise noted)

Limits Parameter Symbol Test Condition

Min Max Unit

Static

Gate-Source Breakdown Voltage -V(BR)GSS IG = -1 µA, VDS = 0 V 40 -

Gate-Source Cutoff Voltage -VGS(off) ID = 0.1 µA, VDS = 15 V 0.4 1.5 V

Saturation Drain Current IDSS VDS = 15 V, VGS = 0 V 30 90 µA

VDS = 0 V, VGS = -30 V, TA = 25 °C - 1 pAGate Reverse Current -IGSS

VDS = 0 V, VGS = -30 V, TA = 150 °C - 2.5 nA

Gate Operating Current -IGSS ID = 0.1 mA, VDG = 15 V - 0.5

Drain Cutoff Current ID(off) VDS = 15 V, VGS = -5 V - 10 pA

Gate-Source Forward Voltage VGS(F) IG = 1 mA, VDS = 0 V - 0.7 V

Dynamic

Common-Source Forward Transconductance

gfs 70 210 µS

Common-Source Output Transconductance

gos

VDS = 15 V, VGS = 0 V, f = 1 kHz

- 3 µS

Drain-Source On-Resistance rDS(on) VDS = 0 V, VGS = 0 V, f = 1 kHz - 18 kCommon-Source Input Capacitance

Ciss - 3

Common-Source Reverse Transfer Capacitance

Crss

VDS = 15 V, VGS = 0 V, f = 1 MHz

- 1.5

pF

Equivalent Input Noise Voltage en VDS = 10 V, VGS = 0 V, f = 1 kHz - 20 nV/Hz



3 OpAmp1 - CMOS Micropower OpAmp for single and multi color detectors

LIE-235 / -241 / -245 / -251 Features Single [(4.5 ... 15) V] and split supply [(+3.5 / -1.5 ... ±8) V] operation

Ultra low supply current [20 µA typical]; ultra low input bias current [10 fA typical]

Rail-to-Rail output swing; high voltage gain [140 dB]

Absolute maximum ratings Differential input voltage: ± supply voltage

Voltage at input / output pin: (V-) -0.3 V ... (V+) +0.3 V

Supply voltage (V+ - V-): 16 V

Current at input pin: ±10 mA

Current at output pin: ±30 mA

Current at power supply pin: 40 mA


Specifications (TA = 25 °C; V+ = 5 V; V- = 0 V; VCM = 1.5 V; VO = 2.5 V; RL > 1 MΩ unless otherwise noted)

Limits Parameter Symbol Test Condition Min @

85 °Ctyp

Max @ 85 °C

Unit

Static Input Offset Voltage VOS - 100 1300 µVInput Bias Current IB - 0.01 4 pAInput Offset Current IOS - 0.005 2 pA

Common Mode Rejection Ratio CMRR 0.0V VCM 12.0 VV+ = 15 V

63 85 - dB

V+- 2.5 V+- 1.9 - V [MAX]Input Common-Mode Voltage Range

VCM V+ = 5 V ... 15 V

for CMRR 60 dB - -0.4 0.0 V [MIN]200 4000 - V/mV [Sourcing]

RL = 100 k60 3000 - V/mV [Sinking]80 3000 - V/mV [Sourcing]

Large Signal Voltage Gain AV RL = 25 k

35 2000 - V/mV [Sinking]4.850 4.990 - V [MAX]V+ = 5 V

RL = 25 k to 2.5 V - 0.010 0.150 V [MIN]14.925 14.990 - V [MAX]

Output Swing VO V+ = 15 V

RL = 100 k to 7.5 V - 0.010 0.075 V [MIN]V+ = +5 V, VO = 2.5 V - 20 40 µA

Supply Current IS V+ = +15 V, VO = 7.5 V - 24 48 µA

Dynamic

Slew Rate SR V+ = 15 V; 10 V step

voltage follower7 35 - V/ms

Gain-Bandwidth Product SR - 100 - kHzPhase Margin GBW - 50 - DegEquivalent Input Noise Voltage m f = 1 kHz - 83 - nV/HzEquivalent Input Noise Current en f = 1 kHz - 0.2 - fA/Hz



OpAmp1 Specifications (TA = 25 °C; V+ = 5 V; V- = 0 V; VCM = 1,5 V; VO = 2,5 V; RL > 1 MΩ unless otherwise noted)

© 2001 National Semiconductor Corporation



4 OpAmp2 - CMOS very low power OpAmp for single and multi color detectors

LME-335 / -337/ -341 / -345 / -351 / -353 / -392 / -553 / -541 / -551; LIM-052 / -054/ -162 / -262 LMM-242 / -244 ; LFP-3041L-337 ; LFP-3950-337 Features Single [(4.5 ... 16) V] and split supply [(±2.2 ... 8) V] operation

Low supply current [70 µA typical]; very low input bias current [100 fA typical]

Rail-to-Rail output swing; high voltage gain [120 dB]

Absolute maximum ratings Differential input voltage: ± supply voltage

Voltage at input / output pin: (V-) -0.3 V ... (V+) +0.3 V

Supply voltage (V+ - V-) 16V

Current at input pin: ±5 mA

Current at output pin: ±30 mA

Current at power supply pin: 40 mA


Specifications (TA = 25 °C; V+ = 5 V; V- = 0 V; VCM = 2.5 V; VO = 2.5 V; RL > 1 MΩ unless otherwise noted)

Limits Parameter Symbol Test Condition Min @

85°Ctyp Max @ 85°C

Unit

Static Input Offset Voltage VOS - 200 1500 µVInput Bias Current IB - 0.1 10 pAInput Offset Current IOS - 0.05 5 pACommon Mode Rejection Ratio CMRR 0.0 V VCM 2.7 V 70 83 - dB

V+ -1.0 V+ -0.8 - V [MAX]Input Common-Mode Voltage Range

VCM -0.3 0.0 V [MIN]

RL = 1 M 350 1700 - V/mVLarge Signal Voltage Gain AV

RL = 100 k 100 350 - V/mV4.800 4.880 - V [MAX]V+ = 5 V

RL = 100 k to 2.5 V - 0.090 0.150 V [MIN]9.700 9.980 V [MAX]

Output Swing VO V+ = 10 V

RL = 1 M to 5.0 V - 0.010 0.150 V [MIN]V+ = +5 V, VO = 2.5 V - 70 150 µA

Supply Current IS V+ = +10 V, VO = 5.0 V - 80 150 µA

Dynamic

Slew Rate SR 1.5 ... 3.5 V stepvoltage follower

50 120 - V/ms

Gain-Bandwidth Product GBW - 200 - kHzPhase Margin m - 63 - Deg

Equivalent Input Noise Voltage en f = 1 kHz - 19 - nV/Hz

Equivalent Input Noise Current in f = 1 kHz - 0.6 - fA/Hz



OpAmp2 Specifications (TA = 25 °C; V+ = 5 V; V- = 0 V; VCM = 2,5 V; VO = 2,5 V; RL > 1 MΩ unless otherwise noted)

© 2001 Texas Instruments Incorporated


OpAmp3 - CMOS very low power OpAmp for use in single supply applications

LME-336 / -346 Features Single supply [(2.7 ... 10) V] operation Low supply current [15 µA typical]; low input bias current [1 pA typical] Rail-to-Rail output swing; high voltage gain [95 dB]

Absolute maximum ratings Supply voltage (V+ - V-) 12V Current at output pin: 50 mA Power dissipation: 10 mW (RL≥10kOhm) Specifications (TA = 25 °C; V+ = 5 V; V- = 0 V; RL > 10kΩ unless otherwise noted)

Parameter Symbol Test ConditionLimits

UnitMin typ Max @ 70°CStatic

Common Mode Rejection Ratio CMRR 0.0 V ≤ VCM ≤ 2.7 V 65 83 - dB

Large Signal Voltage Gain AV RL = 1 MΩ 12 800 - V/mVRL = 10 kΩ 6 12 -V/mV

Output Swing VO

V+ = 5 VIOL = 50 µA to 2.5 V

4.95 - V [MAX] - 0.015 - V [MIN]

V+ = 5 VIOL = 500 µA to 2.5 V

- 4,5 - V [MAX]- 0.15 - V [MIN]

Supply Current, no load IS V+ = +3 V, VO = 2.5 V - 11 30 µAV+ = +5 V, VO = 2.5 V - 13 30 µA

Dynamic

Slew Rate at unity gain SR 1.5 ... 3.5 V step,

RL=10kΩ, CL=100pFvoltage follower

10 25 - V/ms

Gain-Bandwidth Product GBW - 65 - kHzPhase Margin θm - 2 2 - Deg



OpAmp3 Specifications (TA = 25 °C; V+ = 5 V; V- = 0 V; RL > 10 kΩ unless otherwise noted)

© 2001 Texas Instruments Incorporated


www.lasercomponents.com 02/11 / V10 / HW / infratec/ application_library.pdf

detector basics - mark allen

Documents