On the 9th of September this year BIROS released in-orbit the BEESAT-4 picosatellite by means of its Single Picosatellite Launcher (SPL). This device is a spring-based mechanism accommodated on BIROS with the ejection track parallel to the z-axis of the spacecraft body frame.
BIROS bus and payload assembly
The orientation in space of the ejection track and the time of the release have been computed to meet several objectives:
This last point posed requirements on the characteristics of the relative trajectory created by the picosatellite ejection. In particular, a passively-safe configuration of (anti-) parallel relative eccentricity and inclination vectors had to be achieved as soon as possible, to cope with the fact that BIROS and BEESAT-4 are not communicating with each other, and to mitigate any effect of possible short-term system failures on BIROS. Moreover, BEESAT-4 should lead the formation in flight-direction, so that the natural effect of the differential aerodynamic drag (strongly perturbing the relative orbit at this height) would naturally increase the inter-satellite separation. Finally, the x-component of the relative inclination vector should be the closest possible to zero, to obtain a J2 almost-stable relative trajectory. This factor was sought to avoid wasting fuel in frequent, and expensive, out-of-plane corrections. And it constrained the ejection to occur at a pole of the BIROS orbit. We chose the orbit North-pole, almost overlapping with a Svalbard ground station (SPZBG) supported pass and in light during the current season.
Given the 1 kg mass of the picosatellite, the equivalent delta-v generated by the SPL amounted to circa 1.5 m/s, which is a very big value when dealing with formation flying activities. If we would have directed the ejection track solely in flight-direction, for example, the so obtained semi-major axis difference would have produced an along-track drift of 27 km∕orbit (or 405 km∕day). Thus the picosatellite would have quickly escaped BIROS, definitively not enabling the AVANTI experiment! This should not surprise: SPLs have been primarily conceived to release spacecraft in space, to quickly and safely establish a certain relative distance, disregarding how different the final dynamics of the two vehicles are. The fact that we used a SPL as a prelude to a formation flying experiment was part of the challenge!
The spacecraft relative dynamics tells us that a single maneuver, which for us was the action of the SPL device, cannot establish alone the relative trajectory required to start AVANTI. Thus, after the ejection, BIROS needs to perform maneuvers using its propulsion system, to completely achieve the goals listed above. However such maneuvers cost fuel and 1.5 m/s is definitively something we did not want to spend!
Basically our ejection strategy design could be summarized is this way:
Is there a smart combination of time, direction of the SPL activation, and sequence of maneuvers performed by BIROS so that the majority of the initial conditions required by AVANTI are achieved at the expense of the delta-v provided by the SPL?
The answer is yes and on the 11:00:15 UTC of the 9th of September BIROS looked like this:
BIROS attitude during BEESAT-4 ejection (spacecraft not in scale for visualization purpose)
Prior to the ejection the operations team had to verify that the chosen attitude was correctly assumed by BIROS, and only then the command to release the ejection mechanism was given. Shortly after the command was executed the telemetry reported back that the mechanism was successfully released. Immediately three maneuvers were uploaded to the BIROS spacecraft to be executed during the next two orbits in order to reduce the drift between the two spacecraft. The maneuvers were designed in such a way that the drift would only be partially reduced to cope with a possible underperformance of the release mechanism, with a total amount of circa 14 cm/s delta-v.
In the following ground station contact, GPS tracking data could be retrieved from the BIROS satellite and used to estimate the effect of the recoil of the ejection on the BIROS obit. With a mass ratio of about 125:1 between BIROS and BEESAT-4, the BIROS satellite was expected to experience a delta-v of more than 1cm/s. The recoil could be estimated very well and it was detected that the release mechanism had indeed a slight underperformance, but the picosatellite was ejected in the expected direction. This result made it possible to perform a first coarse relative orbit determination and plan two more drift reduction maneuver for the following 12 hours. This in fact helped to save fuel as the drift was stopped before the along-track separation could become too large.
In the following days, with BEESAT-4 flying free some tens of Km forward BIROS, the ordinary activities of the spacecraft were resumed and the commissioning phase of the AVANTI experiment could start. Would BEESAT-4 show-up now and then in the pictures taken with the BIROS star-tracker?