Jun 02 2017

Operations Summary – Week of 5/29/17

View post on imgur.com

Progeny Mk4 Launch Analysis Still Ongoing

During its second launch yesterday the Mk4 once again suffered a break-up in flight, this time later in the ascent during the 3rd stage boost. You can watch the video of the entire ascent, including the break-up, unfortunately some hardware issues created a few errors in the data that we couldn’t completely scrub out. The tracking cameras will get some hardware replaced and be inspected prior to the next launch. Progenitor engineers spent all day today going over the telemetry data as well as the radar contact data sent to us from MSV Tongjess and of course the video recordings of the ascent. They worked late today but were still deliberating on analysis results and overall conclusions when Lead Engineer Simon finally told them all to get out for the weekend and come back fresh on Monday.

What we can say right now is that while the failure of the rocket during ascent is of course unfortunate, remember that we once again intentionally pushed it as hard as possible to exceed any unknown or theoretical limitations. As much as we would have loved for it to have survived everything that was thrown at it, we now at least have a much better idea of the forces at work on the craft during a rapid ascent through the upper atmosphere. We can also state that although telemetry data was lost before the rocket’s apokee reached past 70km, a good portion of today’s analysis was projecting the probable apokee at the time of break-up, which does in fact come out in excess of 70km. We can confirm that space is within our reach with the Mk4!

We’re currently expecting to release a full analysis sometime late in the day on Monday.

Weird Asteroid-Related Weather?

A persistent and very large band of moisture appeared to work its way around the planet in the days following the passage of Pilirani through our atmosphere, impacting the launch of our latest Progeny Mk4 rocket. It first reached the Arekibo Radio Observatory construction site and poured rain down on them for two full days, halting construction for a day when the dish bowl became flooded and had to be pumped out. It then struck Kravass, which wouldn’t have been a big deal for an underground cavern except for the fact that Kravass General Airport flight trainees under instruction of our Captain Jebediah were carrying out some landing pattern work in a Civvie as the valley was fogged in within minutes. Luckily the student pilot was able to land in <50m visibility with only minor damage to the aircraft and no injuries to himself. The storm reached KSC later that day and also stuck around for over an entire day. The intensity of the storm was low, with only moderate winds and rainfall, lightning and thunder were reported but no ground strikes were seen. Meteorologists have done what research they could into the storms after the fact, and have put forth theories that include the asteroid’s passage dumping large amounts of material in the air that were carried aloft by winds to form huge cloud banks filled with moisture – perhaps most of it from the asteroid itself? They are planning to perform a more detailed study after this next pass of Pilirani over the weekend.

Kravass Flight Training Takes to the Skies

For the past several weeks Commander Valentina and Captain Jebediah have been taking turns training a cadre of new pilots for the Civvie aircraft, who will in turn eventually begin to train the dozens of want-to-be private kerb pilots who have expressed interest in buying the aircraft and experiencing the surface of Kerbin from above. C7 Aerospace built a factory in Kravass to manufacture new Civvies and so far 4 have rolled off the line. Two of those have gone to us to expand our fleet to three aircraft while the other two and three of the four coming in June will be for Kravass General Airport to train kerbs on. Last week Jeb started allowing trainees to take the Civvies up for actual flying, talking to them from the ground by radio to guide their flights and landings. So far, other than the incident mentioned in our previous story, everything has been going very well – although there have been moments. The four trainees will continue to pilot actual aircraft for the remainder of their training and each will conclude with a cross-country flight between KGA and KSC. Best of luck to everyone!

Successful Solo KerBalloon Inland Mission

Today the KerBalloon crews along with Bill and Bob carried out a mission to launch & recover a balloon from the highlands northwest of KSC. Normally they would simply recover the balloon that was dropped & deployed by a Civvie aircraft in cooperation with the Genesis program but as they are still trying to determine why deployment failed on the last attempt, the launch was done from the ground this time. While KerBalloon would be willing to partner up with Genesis again in the future, the distances needed to make the joint venture worthwhile will have to be much larger than current contracts are asking our UTVs to travel.

ATN Database Update

The weekly update for the Asteroid Tracking Network database was posted here. It contains now a total count of 615 asteroids. A surprising revision was made to BDN-506(C), which was predicted to pass 938km from Kerbin in 2030 – turns out it is actually passing that close to Minmus. The mistake was noted on the second observation and recording of the asteroid’s orbit, so this could hold up. We’ll see if the next orbital check up in September still shows this encounter.

Celestial Snapshot of the Week

This is a really cool image of Minmus taken by Bob, which looks like a good chunk of the normally roundish moon is completely missing. What’s really happening is the lowland areas like the ones you see lit at the top are completely shadowed by the higher elevations.

From the Desk of Drew Kerman

Out of Character Behind the Scenes stuff

Written on 5/23/17

Uuggghhhh I was doing pretty well in getting ahead and could have written this on 5/21 if something hadn’t come and waylaid me – been spending way too much time not being able to get out of bed. Feeling better now though, probably just nasty seasonal allergies which I hear have been pretty bad all around this year so far. Bleh. OMG so many notes. Let’s get to it.

Progeny Mk4 Launch

I went into this launch knowing it was going to be a pain in the ass and made things worse by being rather incompetent at trying to pull it off. The reason I knew it was going to suck was that getting the rocket to behave very similarly on the actual launch and the various replays I do for video recording was going to be difficult due to the dynamic nature of the wobble (actual occurrence, not induced for plot) that the rocket would pick up when separating from the second stage. I went so far as to write a kOS script that logged the precise time I triggered staging actions so I could play things back as accurately as possible. Still, this led to some pretty wacky ascent trajectories (green is actual, yellow is the recreation attempt) and I actually had to tweak the staging events just a little bit to get this plot, the closest after several tries.

Beyond the difficulty of recreating the ascent conditions when doing things right, I did many things oh so wrong… Let’s see for one attempt since the Camera Tools camera tends to move while targeting a vessel it eventually clipped into the Tracking Station since that’s where I place it above, blocking the view. Another launch when I decided to start using Persistent Trails so I could compare the ascent trajectory and see how close I got afterwards I forgot to disable the trail rendering so a green line was visible following the rocket. I forgot on another launch attempt to reset the SmokeScreen particle count that was lowered to save FPS during the actual launch so flame/smoke trails were almost non-existent. Another launch I forgot to set the throttle so the 3rd stage never ignited….

I finally made a checklist using the checklist program I recently downloaded so that when I reset to try again for a better trajectory match I wouldn’t keep forgetting things and having to redo it again. Checklists are good. I have actual checklists I use when launching rockets and flying planes in KSP. Highly recommend the app I linked for Windows.

Anyways the best launch trajectory I got also ended up clipping into the damn Tracking Station again which is why the video has “data issues” and you see that fuzz/static show up towards the end of the ascent to explain the cut in the video when I had to remove the 3 seconds of footage showing a texture from the Tracking Station blocking the view. It came out looking better than I thought and used an annoying issue in my video editor Sony Vegas where if I had two video clips layered atop each other they’d interfere and cause rendering to spazz out.

I also wasted an hour or two mucking around with Flight Manager for Reusable Stages so I could go back to the booster releases and track their trajectories (sneak peek!) and see how they land after decoupling. However turns out FMRS doesn’t track objects that don’t have a parachute, control unit or command pod attached to them, so I had to figure out what I needed to patch into the boosters using Module Manager to meet those requirements without having to add any extra parts/mass to the rocket. Finally figured it out but turns out that when reloading the boosters they don’t behave anything like I would expect them to. However the addition of the Command module to the boosters has let me select “Control from here” at launch so that when the booster is decoupled, the rest of the rocket flies off and I can see & track the booster. So I just have to recreate the launch a few extra times, which is something I’m already good at anyways. So yea, no longer using FMRS for now.

MM configs for visual enhancements

In the process of hunting down a Scatterer issue I rewrote and added several MM configs for myself to not have to constantly be editing files in the Stock Visual Enhancements install whenever I update it. I also moved some textures I was including myself in the SVE install into my own KSA folder within GameData. So yea now I can just delete the SVE folder and drop in the updated one and not have to worry about any of my own changes being overwritten. Still, I wish there was a way to pre-empt MM patches from even being performed in the first place rather than having to just delete them or overwrite them with your own patch. The MM configs I have also let me adjust properties of the EVE cloud layers since if I do that in the game it messes up the Scatterer integration with EVE (known issue). Thankfully if I setup cloud conditions using the offset property of the cloud layers (like forming overcast skies) in the game and then restart with the settings added to my MM config the clouds show up exactly where they should.

If anyone is interested more in what I did for their own installs, just contact me. One example is using MM to make sure you never have city lights.

Better-quality clouds from the ground

It was pointed out in this post how much better clouds from the ground can look when you tweak up their detail levels (comparison – 10 on left, 30 on right). The reason the SVE dev gives for keeping them at a low level is very valid, but it’s a shame he didn’t bother to mention anywhere how to improve things from the ground given that plenty of people spend a lot of their time in the atmosphere of Kerbin.

Civvie EC usage still screwed up

So it turns out that the issue I spoke about in a previous Desk Notes about the Civvie’s engine alternator no longer producing EC while running at idle is due to a bug fix in SolverEngines, a mod dependency of Advanced Jet Engine, where the alternator was putting out 15% of the total power capability at idle, which didn’t work from a balance perspective for all engines. They still haven’t come up with a good solution so now no engines produce EC at idle. Well okay then so I just added my own to the Bonny cockpit I can toggle on with engine start:

  name = ModuleGenerator
  isAlwaysActive = false
  activateGUIName = Activate Alternator
  shutdownGUIName = Deactivate Alternator
    name = ElectricCharge
    rate = 0.05

So yea battery usage of the Civvies on their recent flights will be different from subsequent ones but I don’t really expose that data anyways so not a huge deal. I’ve taken what I feel is the proper amount of batteries out of my parts inventory and we’ll just leave it at that and continue on…

Non-rechargeable batteries

Speaking of batteries, I mentioned in a previous Desk Notes how I do non-rechargeable batteries by simply disabling them when the engine alternator can keep up with any demand and enabling them when it can’t. So they would always drain and never recharge. Well, I got tired of always having to have multiple Part Access Windows pinned to my display so I could enable/disable batteries without having to hunt for them inside the plane while flying, so after some experimentation on my own that led nowhere I took to the forums and asked if anyone had a better idea. Turns out someone did, and it’s pretty awesome. Now I don’t have to manage the batteries anymore and they drain without recharging. w00t

No more orbit jumping

Last week’s Desk Notes I talked about a problem where loading asteroids would cause their orbits to change drastically. Turns out it was an issue with PersistentRotation, which I had determined at the time was one of the mods causing problems but after a recent update was released it seems to have fixed all the problems I was seeing even with some other mods.

Even though this is fixed, the effects of this bug will be ever-lasting in the orbits of most the asteroids I visited while it was an issue. For example, this is probably why one Trojan asteroid was found to be on a course that actually brushed through Jool’s SOI. Upon re-visiting asteroids their orbits would jump and change their periods by several minutes or hours – some NKOs on course to re-intercept Kerbin’s SOI probably wouldn’t be doing so otherwise. In the end though, it all doesn’t have to be justified in any way since things like this aren’t really exposed to the normal follower. I am in fact thinking it would be cool if the asteroid orbit would continue to have a small percentage of a jump due to “errors in previous observations” to add a bit more randomness to things…

One last thing…

Did anyone notice the ceiling crane?