Mar 17 2017

Operations Summary – Week of 3/13/17

View post on imgur.com

Progenitor Program Returns to Flight with Progeny Mk3

The big news from this past week was, of course, the launch of our first rocket in 2017 and the first launch at KSC in exactly 3 months since the last launch of the Progeny Mk2.1 in December. This was also our 10th rocket launch, bringing our launch count up into the double digits. While ultimately successful in that the rocket did not explode and the payload was recovered intact, the flight profile did not perform as expected, with the rocket failing to travel downrange over the water. Instead, a low TWR of 8 allowed lift at the nose to push the rocket from its 3° launch pitch to completely vertical before enough acceleration was achieved and the rocket spun up to allow it to continue on a straight trajectory. While previous Progeny launches had similar TWR at takeoff, the longer length of this rocket greatly increased the amount of torque the lift at the nose could apply. This lead to a potentially dangerous situation as the spent stages fell close to the KSC facilities and the crowds of onlookers. The first stage managed to land nose-first while spinning and drilled itself halfway into the ground.

Other than the failure to travel downrange, the flight went off as planned with each booster hot staging at the tail-end of the previous booster’s run. In the video of the launch (with sound!) you can see that the booster still appears to be running when the next stage is triggered. This is not the case and is just the booster expending its remaining exhaust after burning through all its solid fuel. The puff of smoke seen during each staging are the stack decouplers being obliterated by the heat of each engine’s exhaust. A quick chute deployment after apokee was to allow the boat in the water, deployed from a Maritime Service Vessel stationed offshore, time to get underneath and secure the payload before it splashed down, as it does not float and would immediately sink, shortly dragging the parachute under as well. Instead it landed just 356m north of the launch pad. The rocket’s official apokee was recorded by the Telemetry Data Unit as 7,546.63m ASL, you can see the full plot here. Altitude however was not the purpose of this flight, instead we were subjecting the rocket to the highest-possible amounts of dynamic pressure to ensure it would remain intact, which it did under a load of 190.237kPa.

In order to help ensure future Mk3 flights travel downrange, the launch base will be modified to tilt the rocket 6° at launch, slightly more than the 5° pitch of the Progeny Mk2.1 rockets. This will lower our apokee but ultimately we have to think about the safety of our facilities and staff. In addition, the fins for the next Mk3 launch will be re-aligned to a smaller degree of pitch for each stage, as Lead Engineer Simon called the rotational speed of the first launch “ludicrous”. It was spinning so fast our tracking camera couldn’t even resolve the marker pattern, and the vertical pitch gave us no roll data in the TDU, so “ludicrous” is the most accurate spin rate we have. The wider base was thought to need more fin angle to spin up, but apparently we overcompensated. A bit. The launch base will also receive some modifications to the lower rail to allow for a more secure mounting of the first stage booster, which nearly set back the schedule of this first flight.

All these changes will take time, which means we’ve pushed back the launch of the second Mk3 to the original launch date of the third, which is now TBD pending the results of the second flight next week. Even though the water landing did not occur, we were pleased to work flawlessly with the MSV Lymun to monitor the eastern range and get it set up for recovery. We look forward to continued joint operations with the Maritime Service for future launches.

One more piece of Progenitor news this week comes from USI’s HQ in Sheltered Rock, where they had a first successful 5s test fire of their Liquid Fuel/Oxidizer (LF/O) engine that will power the Progeny Mk4 closer, if not all the way, to space later this year. Kerzzah!

Genesis Program is Back on its Wheels

This week saw the completion of the Civvie Production model, which was a design overhaul of the original Civvie Prototype using the lessons learned from its various flights around the region of KSC under the command of pilots Val and Jeb. Even better news for the Genesis Program, which has been without an air-worthy aircraft since the accident in January, is that the ground trials for the new Civvie were completed without serious issue earlier today, meaning the aircraft is ready for flight trials next week, maybe as soon as Monday. Commander Val put the Civvie through its paces, first helping the engineers to tune the suspension on the landing gear, then performing various engine runups and takeoff runups. The lighter weight of the new Civvie means much better performance, and it managed to achieve near-takeoff speed in only 624m at 50% thrust, whereas the older Civvie required nearly a kilometer of runway to takeoff at full thrust. Val says she can’t wait to see how it handles in the air, although Capt Jeb will be the first to try it out, as the C7 crash team completed its investigation this week and did not recommend suspension for Jebediah. They did, however, recommend that Rwy09 approaches be avoided until their full report is released, which we are expecting next week.

The plan for the older Civvie, which should be receiving its repair parts at the start of next week, is still to patch it up and fly it up to the Air History Museum in Kravass City, where it will be retired. Assuming no major flaws are found in the new Civvie design, several more will be ordered.

Asteroid Tracking Network Update

The latest release of the ATN catalog can be found here, containing 390 asteroids. They have also determined that Meeny will not be returning to the Kerbin system anytime withing the next 27 years, however like all known NKOs they will re-observe it after every orbit to see if anything changes.

Celestial Snapshot of the Week

Val caught a triple transit with Jool hovering overhead near zenith and was able to track it much easier for a nice close shot. Notice the path of Laythe below Tylo – it will indeed have its north pole darkened slightly when it passes behind Tylo.

From the Desk of Drew Kerman

Out of Character Behind the Scenes stuff

Written on 2/16/17

Being that it’s nearly 1am technically this is being written on the 17th, but it’s really just an extension of the work I was doing most of the day today so, still the 16th. I realized earlier in the week that with my entire upcoming weekend packed with Real Work, I needed to jump ahead to at least next Monday, since I would want a day to recover afterwards, so I haven’t made any more significant progress towards my v1.2.2 upgrade as I’ve just been working to advance my lead during a very busy week, what with a new rocket launching and aircraft getting ready for flight. But it’s all going well so far – I hope to bang out the weekend and Monday tomorrow so I can relax these next three days while taking care of Real World stuff. I have a lot of notes to expand upon this week, so let’s get to it.

Progeny Mk3 Launch

This was exciting, as it really has been over three months in real time since I last got to launch a rocket in KSP. Well, okay I will admit I threw the Mk3 up into the sky just messing around when I first built it, but this was the first launch where I had to follow an actual flight profile – and I flubbed it twice 😛 First time I totally forgot to remember where to look for a readout of the booster fuel and my eyes darted from the staging column to the resource panel to the PAW I had pinned open for the engine. I couldn’t focus on one and missed the hot stage. Then the second time I staged once too many times because I forgot the final booster auto-decouples from the payload once it is spent. I finally got a good launch on the third try. Here’s what my screen looks like during a launch:

Too much info? Well that’s why it’s all video recorded so I can go back later and look at each data point individually if I want to. The Hyperedit window is how I snap the EC back to full charge 2min before takeoff to simulate coming off ground power.

Wait I kind of got off track here, I was saying how exciting it was to launch the rocket for real, which is kind of one of the reasons I like playing the way that I do, it really makes each launch feel like an accomplishment. It’s like – yea I worked real hard over all this time getting everything setup and BAM! Off it goes! Whooo! It’s a bit hard to describe but I really think it’s a bit of a minuscule amount of what real rocket engineers feel when they spend months/years on a mission prior to its launch.

So another thing I sort of tried to do for this launch was to setup as many ways as possible for it to appear like it was going to fail. I had ominous weather, an unplanned recovery ship, a hacked-together mounting job, a thrown-off launch schedule… I really hope people following along were getting a bit of dread in the proceedings, given that KSA launches have failed in the past.

Photos of the event were tricky a little bit, since I had specified publicly where all the kerbs would be spectating from so I had to make sure I didn’t show any of those areas because I couldn’t really fill them with spectators. The astronaut bunker photo was also a lot trickier than I expected it would be, because the kerbals in v1.1.3 still suffer from that ground friction bug that lets them shift slightly over time along the ground, which meant that successive photos, even using a KerbCam position key to reset the camera, were not exactly aligned. So although you hopefully can’t tell, Bill on the left had his lower body replaced, as did Jeb, and Bob had his entire body replaced from another image that had to be resized to match. Bleh. Came out good in the end though.

I also discovered in my setup of v1.2.2 that I never had a RealChute patch installed for Sounding Rockets in v1.1.3. There was a patch created, but although it was sent as a pull request to RealChute it never got accepted, and Sounding Rockets never distributed it. So I tried to install the patch and got NRE spam under v1.1.3 so nevermind about that.

This launch video had sound, because it was a new rocket engine, a bigger rocket engine and I really really wanted to have sound. Still, it took me a few hours trying to source some sounds from YouTube, which I still think is the best resource although to be honest I didn’t spend time looking for a better place to get sounds. I should do that next time. YouTube videos are great, but the audio inevitably either has cheering over the launch soundtrack, no sound at all, or music instead of rocket sounds. Still, I managed to find this clip to use for the first stage booster, and then some other clips for the 2nd & 3rd stages. There was a bit of mixing involved to get the sound the way I wanted, with a gradual fade of the rumble before cheering entered the soundtrack. I think it turned out really good.

One annoying fact though is that the video doesn’t exactly match the events if you calculate the times using the telemetry data image posted. The payload chute deployment, for example, is way off. This is because despite my best efforts the rocket does not always fly the exact same way twice. I can usually get closer to the proper times, but I forgot to really try and the first capture turned out so well, zooming in just as you can see the payload fairings deploy, that I didn’t want to take the time to do it again.

Inventory Rework

I realized belatedly that I really didn’t have the spare nose cone chute to have performed that recovery test of the low-altitude KerBalloon, as all three were already integrated for use in Progeny Mk3 rockets. So, in this case I simply forgot to even look at whether I had a spare part available to use, but it still made me realize that my inventory spreadsheets were not adequate for items that can be reused across various vessels, because while it was easy to see how many I had available to use, it was difficult to check where the ones that were in use were being used. So now I have a new format (original format in these Desk Notes):

So on the left are parts that are reusable and I can see how many I have to assign and exactly what they are being used for at the moment, and how many are available (blank cells). On the right is the original check-in/check-out system that automatically calculates how many parts remain of a non-reusable part. Once these parts are taken out, I don’t really care where they went (even though I do make note of it with a comment) because once they’re gone they won’t be coming back. We’ll see how this goes.

Asteroid Tracking Rework

Another deficiency made itself known this past week, which was in how I was handling asteroids. Every single day new asteroids are generated by the game and I track them all. Each day the ATN pulls a certain number (sometimes none) from this collection of tracked asteroids and officially “discovers” them. So there are significantly more undiscovered asteroids than there are ones that have been listed in the ATN database so far. I won’t say how many but my old pre-v1.0 saves only handled about 500 at a time and I have way more than that. The game actually handles it great as long as I don’t try to show them all at once in the tracking station (I zoom in on Kerbin so only the ones in its SOI are shown before I make them visible). But switching from the tracking station to the asteroids to get their orbital data and makeup has started to take over a minute and process so much that my music even skips while the scene is loaded, which makes me want to put my fist through my monitor. So to circumvent having to load into the flight scene in a game with 500+ asteroids after I “discover” the ones for the day I copy them from the ATN save file to one of my sandbox test files with no other craft. There I can load in and out of the flight scene in a jiffy and it’s overall much faster and more efficient now.

I don’t know if any of that even made sense to someone reading this, but delving into how the ATN functions would take an entire Desk Notes entry on its own. I’ll keep that in my back pocket for when I have nothing else to write about.