Superheterodyning

The original design driver for selecting superheterodyning [1] has been that for white-light stellar interferometers on separated spacecraft (DARWIN, TPF) the absolute distance as well the relative velocity have to be both controlled.

Both metrologies can simultaneously be measured by well-known superheterodyning interferometry [1] where two relative heterodyne laser metrology subsystems with laser frequencies at fixed offset are simultaneously read out in two separate phasemeters. This principle has been selected and breadboarded within the Astrium HPOM 1&2 studies for ESA, and has been a candidate demonstrator instrument for PROBA-3.

The principal setup is depicted in figure 1 and consists of:

Fig. 1.:

Design schematic of laser locking subsystem, super-heterodyning subsystem, and initial dummy test setup

While the first laser receives a heterodyning of net 4 MHz, the second laser sees a heterodyning of net 1.9 Mhz, allowing them to be filtered into separate phasemeter channels that measure phase ϕ₁ and ϕ₂, respectively.

where D is the inter-spacecraft separation.

The difference phase ϕ₁ − ϕ₂ can be written as

and corresponds to a synthetic wavelength Λ which only depends on the frequency offset between the two lasers and the performance of the frequency-locking PLL, but not on individual absolute laser stability. As the derivative of D to individual phase measurements scales with the synthetic wavelength, but not with the individual wavelengths,

very low noise of the individual phasemeter channels is essential as well as precise timing in case of non-zero-average relative velocity. 32 micron absolute accuracy spec therefore correspond to less than 32 μm/10⁵ = 0.32 nm accuracy for the phasemeter noise.

Breadboard level

Consequently three subsystems have been breadboarded:

The design schematic is provided in figure 1, whereas figure 2 shows the actual hardware.

Fig. 2.:

Hardware setup from top left in clock direction:

Laser subsystem and Iodine stabilization [2], Super-heterodyning subsystem, phasemeter, dummy optical head subsystem. Beam combination and polarization control have been breadboarded in three different approaches as discrete optics, fiber-pigtailed discrete components, and purely fiber-based optics

This setup will eventually be tested with an

in a thermal vacuum tank.

EM Engineering Model level

The optical head will be connected to the super-heterodyning subsystem and phase-meter detectors via polarisation-maintaining optical fibres with LTP flight qualification. The head holds the beam combiner bench and the telescope, which collimates the laser beam to be retro-reflected by the target spacecraft. Part of the return beam is steered towards a position sensor device for sensing the target transverse displacement. The separation of emitted and return beams minimises the backscattering towards the laser sources. The optical head has a space envelope of 245 × 180 × 140 mm and a mass less than 2 kg including baffles and cover

To withstand radiations, the optical design includes mainly silica optical parts. The structural parts are all in silicon carbide ceramic, which, in addition to its properties of stiffness, lightness and thermal conductivity, has a thermal expansion which matches the silica one. The afocal telescope has a con-focal mirror design with intermediate image plane, such that a field stop could be implemented to reject the sunlight. The magnified emitted beam has a diameter at 1/e² intensity of 20 mm. The telescope aperture is oversized to allow for lateral shifts of the return beam.

Fig. 3.:

Optical head consisting of telescope on top and beam-combiner bench below central base plate. An intermediate field stop at the focus of the Dall-Kirkham is implemented allowing operation close to sun direction. The receive aperture is tolerant to lateral beam shifts.

The telescope is of Dall-Kirkham type with primary mirror off-axis closely hyperbolical (f/D~0.7) and a spherical secondary mirror.

Fig. 4.:

Aspherical off-axis primary mirror f/D = 0.7 (left) and telescope in alignment tool frame (right)

The beam-combiner bench is based on 50-50% silica beam-splitter plates with the same thickness to balance the interfering paths. The beams have a diameter at 1/e² irradiance of 2.5 mm. The space envelope of the miniinterferometer including collimating and focusing lenses is 100 × 100 × 25 mm.

Fig. 5.:

Balanced beam-combiner scheme (left) and as built ultra-stable all-silica all-glued (right) from Kylia

Stand-alone tests of breadboard

Building up the system with discrete optics as well as with pigtailed free-beam combiners, the setup has turned out to be performing just like “plug and play”. A noise level of 12 μm 1-σ noise was measured “indefinitely”, corresponding to below 0.1 nm noise per phasemeter channel, insensitive to stimulated temperature fluctuations of several K. This is recorded in figure 6, showing very good long-time stability.

Fig. 6:

Long term static distance measurement, compared to the required 1-σ (32 micron) and 3-σ lines. The distance measurement has an RMS/1-σ of 12 micron and is well in spec, and shows no correlation with room temperature fluctuations.

Surprisingly using fused fiber-splitters a significant temperature correlation of about 70 μm /K was observed in the heterodyning subsystem, independent from specific setup of optical head and polarisation cleaning. This could not be explained by first-order theory but must be a higher order effect.

TV tests on EM level

Dynamic velocity and TV tests together with the EM-level optical head are soon to be started at TNO Delft, but will have to be reported in a subsequent paper. These tests will demonstrate