A.

In Figure 1 the basic design of the HSB system is shown. The Intelligent Power Module (IPM)contains all necessary components for the generation of the HSB secondary power, the control processor and the communication interface. The current system is designed that two different types of sensors can be read out, namely integrated electrical temperature sensors with an I²C interface and fiber-optic FBG temperature sensors. Therefore two independent modules are designed, the I²C interrogator module (IIM)for electrical temperature sensors and the fiber-optic interrogator module (FIM)for fiber-optic temperature sensors.

Figure 1:

Concept of the HSB system with Intelligent Power Module (IPM), I²C Interrogator Module (IIM)and Fiber-Optic Interrogator Module (FIM).

Specially for the FIM fiber-optic components are necessary. Thus the main focus of the paper lies on the radiation test results of the fiber-optical and electro-optical components.

B.

The HSB system shall be implemented onboard the German technology demonstrator “Heinrich-Hertz Satellite (H2-Sat)”. This satellite is based on the Small-GEO platform from OHB-System AG. All mission specific requirements such as radiation loads, power and mass budgets and all other requirements for the contribution on the H2-Sat mission are derived from the “General Equipment Requirement Document (GERD)”. For component te sting in terms of radiation tests two important parameters are to be considered, namely the total ionizing dose (TID)value and the non-ionizing energy loss (NIEL). NIEL results in degeneration due to displacement damage in the crystal lattice structure. Electronic components show a different sensitivity to TID or NIEL effects. A detailed list summarizing which type of component is sensitive to which radiation effect can be found in the ECSS-E-10-12A standard [1].

In Figure 2 the TID for silicium (Si, red curve)and the NIEL for silisium (NIEL-Si, black curve)in dependence of the alumina (Al)shielding is plotted. For the system design we assum an aluminum shielding thickness of 5 mm (including satellite housing and HSB chassis)resulting in a TID of 100 krad and a NIEL of 2·10⁶ MeV·g(Si)^-1.

Figure 2:

Total Ionizing Dose (TID) for silicium and non-ionizing energy loss (NIEL) in dependence of aluminum shielding thickness for the H2-Sat mission with a duration of 15 years in geostationary orbit [2].

A.

An FBG is an intrinsic optical sensor implemented directly in the core of an optical fiber. The grating is inscribed in the fiber by UV laser light [1]. The UV light compacts the glass molecules which results in a local change of the refractive index in a grating manner along the fiber. The photosensitivity of the fiber can be enhanced by increasing the germanium (Ge)concentration of the fiber core [2] or by hydrogen loading of the fiber [3] at high temperatures and high pressures.

The maximum reflectivity of an FBG is present at its Bragg wavelength λ_Β, which is a function of the effective refractive index n_eff and the grating pitch Λ, and can be described in approximation by the equation [5]

If an FBG is illuminated by a broadband light source or a tunable laser light, only light corresponding to the Bragg wavelength given by equation (1) is reflected. Other parts of the wavelength spectrum can pass the FBG without be altered. Thus the FBG acts as a wavelength dependent band-stop filter. This is illustrated graphically in Figure 3.

Figure 3:

FBG sensor array is illuminated with a broad light source (here tunable laser) and the transmission (right side) and reflection spectra (bottom left) is shown.

By the use of this characteristic a sensor array can be implemented in a fiber. Each implemented FBG must have a different Bragg wavelength λ_Β creating a definite assignment between the FBG sensor and the reflected spectral peak. The number of embedded FBGs is limited by the optical broadness of the illuminating light source, which is in the case of a tunable laser approximately 40 nm, and by the necessary wavelength bandwidth of each sensor.

The Bragg wavelength of an FBG is a function of the grating temperature and the mechanical strain applied to the grating. Assuming an operating wavelength of 1550 nm in the case of the HSB system, the Bragg wavelength shift (BWS) Δλ_Β in dependence to the temperature change ΔT and to the applied strain Δε can be calculated to [3]

When evaluating the (BWS)conclusions concerning temperatures and mechanical strains can be made. However, for the HSB system the interest lies only in the evaluation of temperature changes. Changes in strain and/or vibrations are no issue here.

B.

Considering the space environment wherein the system will be used, a detailed knowledge of radiation induced changes regarding the measurement accuracy is necessary. Because of the possible use of FBG sensors also in nuclear facilities many studies and tests have been carried out to observe the BWS of an FBG during and after irradiation with γ-rays [4].

By Grobnic et. al. it has b een shown that the FBG’s written in F-doped fibers from Fujikura by the fs-IR (femto-second infrared) technology have the lowest BWS of only 3 - 7 pm after 100 kGy (10 Mrad). No change of the BWS was observed for these fibers either if the fiber was H₂-loaded (type I)or not (type II). The results were identical for type I and type II FBG’s. The H₂ loading of FBGs in standard fibers shows a high impact on the BWS. For an unloaded SMF28 fiber a BWS of 10 pm after 100 kGy was observedn contrast to 60 pm at the same fiber loaded with hydrogen [5]. A change of 3 - 7 pm results in a temperature deviation of 0.3 - 0.7 °C according to equation (2).

Hoeffgen et. al. pointed out that also UV written FBGs can be resistant to radiation. In the case discussed above the fs-IR written FBGs have only a 10% - 20% smaller BWS. The highest impact on the BWS has again the hydrogen loading of the fiber prior to the writing process [6]. The pressure of the hydrogen loading has only small influence which has been shown by Henschel et. al. [7]. It has also been shown that the BWS is two times higher at temperatures around -50 °C in contrast to the situation at 20 °C. This thermal annealing effect is caused by the annealing of radiation induced color centers, which change slightly the refractive index according to the temperature. In most cases the BWS shows a saturation behavior which causes after a certain dose that the overall shift remains at a constant value. If the FBGs are written in a N-doped fiber the BWS shows no saturation at doses higher than 1 MGY [8]. Therefore such fibers are absolutely not suitable for temperature monitoring in space applications.

Analysis of the BWS with different fiber coatings have been carried out by Gusarov et. al. [9]. They have shown that the stripped fiber has the lowest radiation sensitivity, whereas the ormocer coated fiber has the highest shift. This might be caused by the radiation induced aging effect in the coating which generates a mechanical stress between the fiber core and the coating.

C.

Two possible interrogator designs are considered in the HSB project. The first concept is based on a modulated grating Y-branch (MGY)tunable laser. This type of laser diode can be tuned from a wavelength of 1528 nm to 1568 nm with the help of three input currents. Inside a control FPGA a look-up table is stored wherein the relation between laser wavelengths and reflector currents is implemented. During a measurement cycle the tunable laser scans through the spectrum and the reflected light from the FBG array is measured by a photodiode. The reflected intensity signal is assigned to the output wavelength again in terms of a look-up table (LUT). Afterwards, the setup wavelength and the reflected signal are evaluated in the FPGA. The position of the reflected peak of the FBG can be estimated either by a finite impulse response (FIR)filter or a centroid algorithm. The BWS will then be evaluated by the logic implemented in the FPGA and assigned to a temperature.

Figure 4:

Fiber-optic interrogator based on a tunable laser with control FPGA, optical circulator, FBG sensors and photodiode.

The second interrogator design is based on a spectrometer and a broad-band light source, for example a superluminescent laser diode (SLD). The SLD illuminates all FBGs in the optical fiber simultaneously and the spectrometer can query all sensors nearly at the same time. Some complications arise when more than one sensor fiber shall be queried. Here the use of optical switches might be necessary. The most sensitive element in this setup is the spectrometer itself, because it is sensitive to vibrations and temperature changes and also to cosmic radiation. The design of a very stable free-space spectrometer would be the biggest challenge for this setup.

Figure 5:

Fiber-optic interrogator based on a spectrometer and a broad-band light source.

D.

The tunable laser, which is the core element of the scanning laser interrogator, is a monolithic laser diode whose wavelength can be controlled electronically by three input currents. Basically two resonators are combined within the structure of this laser diode. Both resonators share the same partially light-transmissive end. At the other end a grating structure acts as second reflector as shown in Figure 6. The MGY laser diode emits light at a wavelength where the modes of each resonator overlap. The grating structures can be modulated by carrier injection allowing a refractive index modulated by carrier injection allowing a refractive index modulation of the grating material. Electric currents supplied to the grating reflectors adjust the free spectral range (FSR)of the resonators which originally is 630 GHz (5.05 nm)and 700 GHz (5.61 nm)respectively [10]. Thereby the spectra of both resonators can be shifted against each other which allows tobring different modes of the reasonators to overlap. By this so-called Vernier effect [11] wavelength tuning is achieved.

Figure 6:

Principle of the MGY laser diode with left/right reflectors and common phase reflector. By carrier injection the FSR of both reflectors is changed and a tuning of the output wavelength is possible.

Additionally a semiconductor optical amplifier is placed behind the MGY structure to increase the output power.

A.

MGY Tunable Laser Diode

This section describes the radiation testing of the MGY laser diode. The radiation test procedure is derived from the ESCC No. 22900 [12] and adapted for the laser diode test. The TID test was carried out by the help of a Co-60 source. In a first step the LUT of the laser diode is generated and saved. This is necessary in order identify the correct stable wavelengths of the laser and to perform a detailed analysis of the radiation induced shift of the wavelength after the test.

The following parameters are measured in-situ during the test:

The fiber-optic interrogator module is more sensitive to changes in the output wavelength than in the output power. So a special measurement procedure has been established for the test. wavelength step starts at the same definite mid-wavelength λ_mid. During the whole test this wavelength can be investigated often and a long term drift over radiation exposure time can be identified. This is illustrated in more detail in Figure 7. The single wavelength points are taken from the most stable regions of the laser diode. So, no loss of wavelengths can occur during test test. In summary 37 different wavelengths plus 37 times the mid-wavelength where measured in one cycle.

Figure 7:

Measurement cycles for a scan through the lasers spectrum. Each wavelength (λ₁ to λ_n) is se tup from the same mid-wavelength to identify long term drifts.

The wavelength is changed at each step according to Figure 7 until the TID of 100 krad is reached. Then the radiation exposure is stopped and the measurement is continued for at least one additional hour. Afterwards a room temperature annealing under bias is performed. All parameters are measured 12 h, 24 168 h after exposure respectively. The test requires the use of a biased diode and an unbiased diode. The unbiased laser diode is measured every 10 krad TID. Thereby the on-time for the measurement is approximately 2 minutes. The laser threshold current and the slope efficiency are also measured every 10 krad by changing the laser current and monitoring the optical output power. During the test three different dose rates are applied. The duration of each step and the resulting TID is summarized in Table 1.

Table 1:

Different dose rates during the Co-60 radiation test of the MGY laser diode.