Jul 15 2019

Progeny Mk6 Block II Flight 3 Analysis

Although we did not expect to fly the Mk6 Block II again, Bluedog Design Bureau and DMagic Orbital Sciences joined together to fund another mission to explore the radiation belts – this time on the night side of the planet. We know from past Mk6 Block I flights that the belts are blown into an extended elliptical shape by the kerbolar wind however we’ve never been able to explore the outer belt with the Block I. The Block II had the potential to not only reach it but travel all the way through and reveal the full extent of both belts. It was also equipped with the new hemispherical ion trap instrument for additional readings of the belt environment.

The Flight

After first being bumped a few days due to a delay of the Progeny Mk7-A debut launch scheduled prior to this one and then delayed further thanks to a bad decoupler between the first and second stages, the rocket flew its mission on July 9th, lifting off on schedule at precisely 12:30:00.08 local time. The ascent was nominal when compared to past Block II launches, with a clean radial booster drop and core booster separations leading up to the final boost of the liquid-fueled engine. As the rocket exited the atmosphere, it did something new for a Block II and dumped the upper payload fairings to expose the ion trap instrument to space, which managed to affect the rocket’s ascent angle and push its final apokee out to a whopping 4.3Mm!

Everyone was stunned by the huge increase over the previous mission that saw an apokee of just 3.1Mm and while it was exciting to know we were exploring farther out into space than ever before a serious concern was now directed at the battery supply. Basing our planning on the previous mission, we had decided to change the hibernation routine. Previously the rocket logged instrument readings every minute for the entire flight and it returned with 64% power still remaining. This meant we could shorten the hibernation intervals to 30s and leave the rocket awake to log data every second while it passed through the inner and outer belts. Since we did not know their exact positions, the rocket would log its altitude when exiting the belts so it could be sure to re-activate before then on the way back down so that with the two-way trip we could get exact boundary measurements.

However all this was planned for a mission length that took into account an apokee of 3.5Mm at the greatest, allowing for slight changes to the unguided ascent. 4.3Mm added almost another hour to the mission duration. The only Block I launched at night failed to fully penetrate the inner belt so without knowing the size of either belt we had no way to calculate how long the rocket would remain fully awake and thus how much power would be drained. As such, there was a lot of tension in mission control as the rocket flew out of comms range just as it was coming towards the edge of the inner belt.

After that, there was nothing we could do but wait. Thankfully, even with our planned maximum height of 3.5Mm we were still afraid of a potentially runaway scenario where the probe core failed to hibernate properly and implemented a power fail safe. If the battery levels dropped below 3% while out of contact with KSC the rocket would power off and remain that way until it neared the atmosphere and would be within range of ground station Arekibo. It would then have enough power left to wake up, lock signal and dump whatever data it had collected. Still, we were of course hoping the rocket would have power remaining when it got back in contact.

Using the data from its last contact we plotted its trajectory and estimated when it would be back in range for contact. Nearly 3.5 hours after it had launched mission control was once again full of kerbs waiting and hoping to hear from the spacecraft as it fell back towards the planet. A resounding cheer went up when, several minutes early, the probe contacted Arekibo and dumped its data while also continuing to stream current telemetry – it still had power! Albeit only about 4% but it was nearing the lower boundary of the inner radiation belt and everyone remained on the edge of their seats to see if it would make it all the way through before losing power.

It did! The rocket squeaked out from the inner belt with only 0.85% power remaining and managed to hibernate three more times before finally falling below the horizon prior to re-entering the atmosphere. Its final actions before completely losing power would have been to arm the parachute, which is triggered by a backup pressure sensor in the event the batteries die. Whether or not the rocket survived splashdown in the waters west of Arekibo we may never know, as a recovery effort would be cost-prohibitive. But who knows? It may be found someday if it remained intact. Regardless, we got the data we wanted and the mission was deemed a resounding success.

Flight Telemetry Data

Flight Analysis

Radiation Belt Visualization

With the telemetry collected from this mission and the previous one over the day side, we can at last draw a picture of the radiation belts based on actual data – all past illustrations have simply been free-drawn based on our conceptual understanding. This is very exciting to see but does come with a few caveats:

  • We may be slightly off in the rotation of the trajectory for the day side launch relative to the night side. In the time between the launches Kerbin has moved some distance around the sun. Since the belts always extend away from the sun, the orientation has shifted during this time. The rotation of Kerbin is also not the same as the length of its solar day, which needs to be accounted for in the time between launches since the position of the sun, and thus the belts, is not the same at 12:00 in April as it is in July.
  • The day time data is taken once every minute throughout the entire flight which means the boundaries of the belts is not as exact as the night time data. This mostly affects the size of the inner belt since that is when the rocket was traveling fastest, although at this scale these differences may be negligible.
  • The shape of the magnetosheath is still mostly free-drawn. We know where the boundary is in front of Kerbin and we know where the boundary isn’t on the other side since the Block II on this flight did not leave it, but beyond that its shape is still based on conjecture.
  • This is a top-down representation of a toroid-shaped inner belt, meaning the edges change based on what angle the rocket passes through it. Due to the unguided nature of the Mk6 the two flights were not flown at precisely 0° inclination and thus the start/end boundaries of the belt would not be expected to be uniform, hence the slightly misshapen appearance when we would expect a more definitive thinner belt sunward and thicker belt on the other side. Also the outer belt is expected to connect to the radiation at the poles and thus has a non-uniform shape as well

Despite all this it’s still exciting to see all our past predictions are on target and the shape of the belts conforms with expectations.

Extreme Apokee

How we ended up flying so high was determined to be due to both detaching the upper fairings and the slight change in rocket center of mass. The fairing detachment proved to be an issue on the previous Mk6 Block I mission, and despite additional efforts to ensure a clean separation including waiting until the rocket was out of the atmosphere one or both still managed to get hung up, spoiling the rocket’s balance and introducing a large pitch wobble that significantly raised the nose upwards for an extended period of time.

Marked events are radial booster drop, 2nd stage ignition, 3rd stage ignition and payload fairing detachment. Click for full resolution

This was coupled with the CoM shift that came from moving the payload instruments from the lower truss to the upper truss. This was done so that after the fairings were removed they would be as far away as possible from direct exposure re-entry heat (we know, the effort went to ensure the payload survived when we had no plans to recover it may seem questionable). The change in CoM had the affect of causing the rocket to pitch over more than the previous flight (seen also in the plot above), which in turn allowed it to pick up more speed.

Click for full resolution plot

It may not look like much of an increase, but at these velocities and at this distance from Kerbin the effects are very large. Thankfully it did not ultimately negatively affect the mission.

Radiation Damage

Despite extra shielding around our probe core that was implemented towards the end of 2017 after our first flights through the inner belt, it seems the extended flight time and passage through the larger areas of the belts managed to add up. There is some unexplained out of order operations log instructions late in the mission and during the second passage through the inner radiation belt a 25s data gap appears in the telemetry. While not having an affect on the mission, we can now at least begin to quantify radiation damage over extended periods of time in space.

Future Plans

Once again we have no plans to launch any more Progeny Mk6 Block II rockets – not only have we exhausted payload options but also mission objectives. There is nothing more substantial we can learn about the radiation belts from a sub-orbital mission. We were wrong before and could be wrong again but as of right now no future missions are being worked on.

One thing that will come out of this mission thanks to the evident radiation damage will be a stronger push for the development of radiation tolerant hardware as opposed to radiation hardened hardware. The difference is that rather than just shielding electronics from the effects of radiation they will have the ability to deal with it, mainly at first by using two computers that can cross-check each other and correct any errors that may be affecting one or the other. This will require larger probes than Progeny is currently capable of carrying up into space but is something the Ascension program can pioneer (in fact, when originally announced, dual computer redundancy was on the list of improvements however the switch from unkerbed orbital to kerbed suborbital occurred before large enough probes could be built).