The Truman Kerman Craftyard

[Hotel26]

Magellan Light Cruiser

Craftyard

Description
1:gen 0:gimbals

A stock ion cruiser called Magellan Light Cruiser or LMC for short. Built with 214 of the finest parts, its root part is mk2Cockpit.Standard.

This is an extension of Light Saber and is capable of landing on a lunar surface.

3.5 km/s dV, 2.3 m/s/s and TWR 1.6 (Mun). Transports 10 krew.

AG1 controls the fuel cell array providing sufficient current to run all engines continuously.

No clipping was used in the manufacture of this craft.

Modified Zephyr lifter included.

Built in the VAB in KSP version 1.7.3.

 

[Hotel26]

Truman finds the following, small, Python program indispensable for computing stopping distances while de-orbiting.

>deorb 552 2.31 2100
65.99 km 18.9 deg
>

It takes 2 to 4 arguments, being: orbital speed, max deceleration (KER), AGL and radius of the body being orbited (default 200km).

It produces line-of-sight distance to the target for commencement of the decel burn and (great circle) degrees before a location when a target is not already present there.

#  deorbit.py

import math
import string
import sys

if __name__ == "__main__":
  if len(sys.argv) < 3:
    speed = 545
    acc = 9
    alt = 6000
    radius = 200
    print('deorbit speed acc [alt]')
  else:
    speed = float(sys.argv[1])
    acc = float(sys.argv[2])
    alt = float(sys.argv[3]) if len(sys.argv) >= 4 else 0
    radius = float(sys.argv[4]) if len(sys.argv) >= 5 else 200

  dist = speed * speed / 2 / acc
  dist = math.sqrt(dist * dist + alt * alt)
  print("{0} km {1} deg".format(round(dist / 1000, 2), round(dist / (2 * math.pi * radius * 1000) * 360, 2)))

I hope and trust posting a very small amount of code is not in violation of forum rules (which I have just checked) -- and that this snippet may provide fun & utility to others!

The formula, d = v² / 2 / a, is derived from d = v₀ + 1/2  * a  * t², and can be easily understood arranged as d = (v / 2) * (v / a), which is the average speed while decelerating multiplied by the total time for deceleration.

 

Edited December 29, 2019 by Hotel26

[Hotel26]

PARDON OUR DUST -- AS THIS SITE UNDERGOES A PERIOD OF REFURBISHMENT

Truman has made a commitment to bring the site index up-to-date and keep it that way.  The lead page will list the Hangars only and link to a page for each Hangar, containing its index.  The intention is to make it easier to browse.  It seems also that the layout could be made friendlier for small-screen devices.

The rest of this page will, during construction, be the prototype for the Personnel Lifter Hangar index.  Craft will be listed, newest to oldest.  Thanks for your patience.

                                                                                                           

What a rotten procrastinator I am/Truman is!  How hard could this be?  Just do it!

OK, let's pick my Lifter hangar and start with the photos...

 

Star Knife

https://kerbalx.com/Hotel26/Star-Knife

 

 

 

  https://kerbalx.com/Hotel26/Blister

   https://kerbalx.com/Hotel26/Spearhead

Venom V

A stock resupply rocket called Venom V.  Built with 84 of the finest parts, its root part is probeStackLarge.

I will use this ship to keep my LKO space station(s) supplied with monopropellant (3,750 kals).

Includes a Sparrow and a recoverable Spunk lifter.

Built in the VAB in KSP version 1.7.3.

 

   https://kerbalx.com/Hotel26/Xenon-III

   https://kerbalx.com/Hotel26/Terrapin

   https://kerbalx.com/Hotel26/Aquila-VIII

   https://kerbalx.com/Hotel26/Mule

   https://kerbalx.com/Hotel26/Methuselah

 https://kerbalx.com/Hotel26/Spunk

   https://kerbalx.com/Hotel26/Archer

   https://kerbalx.com/Hotel26/Zephyr

   https://kerbalx.com/Hotel26/Titan

   https://kerbalx.com/Hotel26/Escort

   https://kerbalx.com/Hotel26/Forward-Assisting-Rear-mounted-Thruster

Edited May 27, 2020 by Hotel26

[Hotel26]

Jetstar

Craftyard

Description
1:RAPIER 2:NERV 3:boost 4:cut

A Mach 7 hypersonic aircraft called Jetstar. Built with 38 of the finest parts, its root part is mk2SpacePlaneAdapter. Perfect for fast trips to destinations, 1,000 to 1,884 km, distant.

Cruise 2.187 km/s @31km.

Atmospheric Autopilot [AA] is a must in order to achieve cruise flight.

Inspired by the pioneering work of Laie in the 80 Minute Challenge, this is the commercial application for up to 8 pax.

Preflight check-list
- View: Free
- Open AA Cruise Flight Controller
- set initial desired heading and select Heading control
- dial in Speed control: 2220 m/s but leave disengaged
- set altitude to 31km and climb rate 10m/s and select Vertical speed
- ensure Advanced options: max_climb_angle: is 15 degrees
- select Cruise Flight controller for take-off but leave Master Switch: Off
- pin a RAPIER tweak menu open to the side

Take-Off
- activate RAPIERs, SAS Hold, disengage Brakes and go Throttle: Full
- rotate at 90 m/s; just raise the nose to lower the tail-wheel to the runway
- when clear and gear: up, engage Master Switch to establish a shallow climb and turn on course
- when on course, engage Altitude control to start the climb toward an ultimate apoapsis of 36.6km altitude (even though the Altitude control is set for only 31km!)
- at Mach 1, activate the NERV
- at speed: 1km/s (attitude centered in the navball prograde), engage SAS Prograde and disengage AA
- engage RAPIER boost when their thrust drops to 100 kN
- kill the throttle when the apoapsis arrives at 36.6km
- wait for 25 km altitude and engage AA to level off at 31km and go Throttle: Full!
- engage Speed control: 2220 just before the speed arrives at this number
- when speed is stabilized at this target, shutdown the RAPIERS
- periapsis will stabilize at 31km, altitude at 31km, apoapsis at 36.6km
- immediately reduce speed setting to 2,187 m/s once Pe shows 31km
- Congratulations! You have now achieved cruise flight profile. Enjoy your flight!

Descent
- begin 500 km from the destination
- set throttle to idle and kill Speed control
- engage Standard Fly-By-Wire
- begin mild porpoising to wash off some speed gradually to below e.g. 2 km/s
- begin a slow descent
- beware of heat indicators as you approach 20-25km altitude; keep the AoA at/near zero as much as you can
- Good Luck. Patiently done: there is sufficient safety reserve

Thanks, Laie!!
 

[Hotel26]

@Mars-Bound Hokie just opened an interesting can of worms on KerbalX @ https://kerbalx.com/Hotel26/Cobra

 

With 1.2 km/s dV arriving at LKO, is this spaceplane suited for further-afield operations?

Well, now, what a wonderful question because I have my second transfer window for Jool opening up in just 24d.  (The first fleet haven't even arrived yet!)

And, I'm ashamed to admit, I have never even been in orbit around our sister-planet Laythe, let alone set flipper there...

So I've started an exercise -- more of a "High Space Kommand initiative" really, now that the Big Wigs are involved -- to boost a few of my spaceplane candidates into orbit, refuel them and then compare them by a) dV and b) krew load.  I am also somewhat interested in viability in terms of local operations [Laythe] for exploration.

My preliminary appraisals are showing that Cobra may have some greyhound space legs on it.

I'll be reporting back here from time to time...

Edited April 5, 2020 by Hotel26

[Hotel26]

@Mars-Bound Hokie you may find the below results of interest.

How far can various space planes go after refueling in space?  Of particular interest is Cobra below (originally known as Duster).  It arrives at LKO with 1.2 km/sec dV, which is a respectable reserve, but won't go far on a voyage.  Yet, with refueling, it's enough to go to the Jolian system.

(I'm puzzled by some of the performances, particularly that Freebird didn't do better, but will be analyzing these to glean more insight.  UPDATE: it has not NERVa, but Reliants.  Yech.)

So here's the leaderboard, so far, showing LKO departure dV after full refueling:

 

4.688 km/s dV

Epiphany

 

 

 

 

 

 

4.142 km/s dV

Jolly Roger

 

 

 

 

 

3.638 km/s dV

Cobra

 

 

 

 

 

3.224 km/s dV

Atkara

 

 

 

 

1.882 km/s dV

Freebird

 

 

 

 

 

 

1.373 km/s dV

Jackrabbit

 

 

 

 

 

 

 

Note that LKO to LVO [Vall] requires 2.910 km/s dV according to the subway map.

[Evidently, Jackrabbit is headed for the scrap heap...  I think the take-away here is that any SSTO that does not include NERV power is not going interplanetary without further assistance and some mighty special good reason for doing so.]

Edited April 6, 2020 by Hotel26

[Hotel26]

This is Epiphany...

It's an all-LF machine...  twin Whiplash and twin NERVa...

Total capacity is 3,380 kallons.

But it departs KSC at sea-level with a load of just 1,620 kallons in the tanks...

By the time it arrives in 72² km LKO, there are only 326 kals remaining for a pitiable 596 m/s dV...

But when it's fully refueled in LKO: its dV is 4.688 km/s.

[Hotel26]

[Ahem.  Truman has been very lazy, lately; at least, about the Craftyard.  Look at that: nearly 5 months since the last post!  Procrastinating about doing some clean-up here in this site.  "Rome wasn't built in a day", and other excuses.]

OK, I have been on a kick lately with VTOL, learning a lot from some of the masters of that particular craft.  In an earlier post, I outlined my 1-2-3-Hover^[tm] control scheme for VTOL control.

I have used another scheme as well that I call GoHover and my VW Beetle of VTOL machines, Hexapen, is a very good example of its use.

In its very simplest form it is as follows:

• k1 toggle horizontal thrust
• k2 toggle vertical thrust
• k3 toggle boost
• k4 VERT reference
• SAS Hold 'F': HORZ reference

This is 4 keys to do what is normally done with 6.  (Since the HORZ ref selection is overloaded on SAS Hold, which still operates for that purpose, it doesn't get counted.)

First observation is that the VERT engines are set up in Boost (turbocharged; wet) mode in the SPH, while the HORZ are defaulted in Normal (dry) mode.  For VTOL, there are circumstances in which you wish to operate both sets of engines (during the transitions to/from hover) but you are never going to require Boost on both sets simultaneously.  It'll be one or the other.  Therefore, the Boost key toggles boost between the HORZ and VERT engines; one set or the other.

Second observation is that, when you are entering the hover and wish to select the VERT ref, you will then select SAS Radial Out, which will hold a neutral reference attitude for the hover.  Leaving SAS Hold selected, you can 'blip' control movements for small inputs and rely on SAS Hold to return your attitude to neutral.  This is a big assist with the pilot workload while targeting the soft touchdown.  You can tap 'F' to exit SAS Hold and still retain the VERT ref to continue using a WASD helo cyclic control scheme (explained below).  But, as a result of the likelihood that SAS Hold may be engaged while in Hover mode, it is absolutely essential that SAS Hold of Radial Out be CANCELED before changing to HORZ ref.  Therefore: binding HORZ ref selection to the 'F' key not only saves a key, but ensures the pilot faultlessly transitions into Go mode.

So a note about the VERT ref.  If a vertical-facing dock is not available (generally not on atmospheric VTOL), then I plant a cubic octastrut on the fuselage somewhere; attach an Okto2 to it; and then rotate it and shift-gizmo it into the fuselage such that one end of the octa strut is barely visible/clickable (to pull it out for maintenance such as re-assigning the VERT ref key binding or changing parts, etc).  The important task is to orientate the Okto2 such that a standard QWE/ASD control function scheme is presented to the pilot.  My choice is to place the Okto2 such that the "OKTO2" caption faces the stern of the aircraft and is right-side up.  That gives WASD as a helicopter cyclic control:

• W nose down, same as conventional flight
• S nose up, ditto
• A roll left (just like Q in convention flight)
• D roll right (ditto E)
• that leaves Q & E controlling the less-important yaw and operating in reverse: Q yaws starboard and R yaws to port

                                                                   

Hexapen uses a slightly more complicated variation of GoHover in which k1 & k2 "activate" the HORZ and VERT jets, respectively -- and toggle the other set.

So tapping '1' gets you ready for conventional flight and tapping it again repeatedly yields control of the VERT jets.  Tapping '2' makes your VERT jets the main set but allows you to control the HORZ on and off by repeated toggling.

This can be quite intuitive actually.  Fly Hexapen, if you want to try it in the cockpit.  The only consequence of this is that you need another key to shutdown all engines.  On Hexapen, I bound it to the ladder key.  Make sure you are already on the ground when you put the ladder down!!  Make sure you pull the ladder up before you start engines for flight!!  For craft with no ladder, or by preference, I also bind it to Abort, as a reasonable choice.

Edited August 27, 2020 by Hotel26

[Hotel26]

I am intending to standardize my VTOL craft on this scheme and so will document it here.  I would be quite happy to hear comments from others and also encourage those aviators who use & recommend other schemes to outline them in this thread as well.

3-2-1-Fly, a VTOL control scheme
(adapted from a @swjr-swis concept)

Definitions: Vt and Ht engines form two sets, providing vertical and horizontal thrust respectively.  Vr and Hr define control references for the vertical and longitudinal axes, respectively.

• [1] Flight Mode: conventional flight utilizing lift from wings and powered by Ht engines.  Vt is inoperative.  Hr is selected.
• [2] Transition Mode:
• [3] Hover mode: craft is not moving quickly in any direction across the ground but may be ascending, descending or simply keeping station.  Only Vt is operative and Vr is selected, most usually with SAS Radial Out assisting the pilot by keeping the hover attitude.
• [4] Shutdown: all engines on the VTOL vehicle are shutdown.

Flight <-> Transition <-> Hover

Vertical Departure and Transition to Flight

• Use [R] to open the hover bays (if any) to prepare for a vertical departure.  (Otherwise, your lunch will quickly being to cook...)
• Select [3] Hover mode to start the Vt engines and select Vr control reference.  Activate SAS and engage Radial Out.  Spool up the engines to ascend and clear obstacles.
• Be absolutely sure to tap [F] to clear Radial Out/select SAS Hold
• slightly depress the nose to commence forward motion
• quickly tap [2] to engage Transition mode, which will start the Ht engines (keeping Vt operating) and select the Hr control reference.  Keep the nose down while airspeed builds.
• as soon as flying speed is attained, tap [1] to shutdown the Vt engines and ensure the Ht engines are running.
• you may now close the hover bays, if any, with [R]

Transition to Hover:

• as you approach the landing target, still retaining sufficient forward speed to coast to it, begin the transition with [R} to open the hover bays, if any.
• tap [2] to engage Transition mode, which will shutdown Ht and engage Vt.  You can begin ramping up Vt power to compensate for anticipated decline in wing lift and to maintain the coast (speed) and glide path as desired.
• tapping [2] repeatedly now toggles Ht on and off as you need it, should you have misjudged your initial approach speed upon entry to the Transition.
• Hr is still engaged during the transition because KSP control surfaces can react erroneously while Vr is selected and this often leads to fatal consequences when there is significant airflow over those surfaces.
• only engage Hover mode when the horizontal speed component is much closer to 0 than it is to 30 m/s, which is the speed that true hover control becomes nigh impossible.
• complete the transition by tapping [3], shutting down Ht, leaving only Vt operating and engaging the Vr.  Quickly then select SAS Radial Out.  You are now in the hover.
• Use the [WASD] joystick (with [QR] opposing yaw) to momentarily  'jiggle' the attitude away from the neutral hover attitude to control descent to a spot landing.
• with roughly TWR=1 Vt operating, use the Falling Leaf technique to hasten and control quicker descent to the ground for a safe landing

                                                                                                                     

For anyone who wishes to try this immediately in practice, I think my Hexapen Deluxe machine is currently equipped with 3-2-1-Fly and is an ideal trainer for VTOL.

Good luck!

Edited August 3, 2021 by Hotel26

[FleshJeb]

I'll make the argument that using a second, vertical reference necessarily reverses at least one control axis from the "expected" direction.

From a cognitive standpoint, I find manually holding my heading indicator on the horizon line leads to more instinctual and timely error correction in the hover.

I would be interested in a comparison of the two techniques when VTOLing in uneven terrain. I can't even boot KSP right now, otherwise I'd test it myself. However, I've taken off and landed next to every monolith on Kerbin multiple times.

[swjr-swis]

1 hour ago, FleshJeb said:

I'll make the argument that using a second, vertical reference necessarily reverses at least one control axis from the "expected" direction.

It does, and it is a very valid argument - one does need to be keenly aware of the axis effects and use controls accordingly (assuming a control reference pointing vertical up, pitch stays the same, but roll and yaw switch places control-wise, and yaw control is additionally reversed).

Even once accustomed to this and in full awareness of it, there can be side-effects that go beyond the pilot's control, eg. KSP's SAS/control code can often, but alas not consistently, get confused about what to do with any control surfaces whose longitudinal vector pass near the CoM, as is often the case with those that normally affect roll. Any significant airflow over those surfaces will then have unpredictable effect, feeling much like the controls fighting you.

This flight/control scheme is, admittedly, still only a best compromise of sorts, from the options KSP gives us. There's a number of alternatives, but they all have their own disadvantages due to a few quirks and/or missing things in the controls, capabilities, SAS, and keybinding that KSP offers. @Hotel26 and I have been having many musings on this over time, me from a purely gameplay point of view, and his from someone with actual RL pilot experience. I think we're both very keen on hearing how others a) experience this particular control scheme, and b) what other alternatives are being used to good effect in KSP.

 

1 hour ago, FleshJeb said:

From a cognitive standpoint, I find manually holding my heading indicator on the horizon line leads to more instinctual and timely error correction in the hover.

I will admit, even accustomed to the scheme as exposed here, I still at times do as you say.  I haven't yet figured out what the specific circumstances are in which I pick one over the other. Also, I play KSP exclusively with mouse and keyboard, which means that control inputs are only as subtle as I can 'short-tap' the keys... which is never going to be as good or accurate as joystick controls.

Note though that what you propose is also still readily available in this control scheme. The extra binding of the horizontal control reference to the [F] key is a quick way to revert to your preferred (and 'stock') hover setting with the navball focused on the horizon. Part of the compromise.

[Hotel26]

I can't throw much light on this, but perhaps a little.

Let's say that positive control with a horizontal reference is the most accurate way to control the aircraft including in the hover.  I think it is.

While hiking up a mountain in Japan, clothed in fog, I witnessed a medivac hovering very precisely, absolutely motionless about 10 meters perhaps over the surface of a track, presumably,  just above the trees.  I presumed someone had had a heart attack and were being treated and possibly about to be evacuated.  The helo hovered for what might have been ten minutes.  This is an awfully long time for a pilot to control the hover -- and yet: it was quite obvious that the helo was Parked in autopilot.  It didn't move a centimeter from where I could tell, including not in the vertical axis.  Due to the fog, I could not see any activity below the helicopter, not even on a ladder that might or might not have been there.

So my point is as follows: if you are in a hover, controlling a motion to a target, go ahead and do it and probably a horizontal reference (and standard controls) is a good or the best way to do it.  No contest.

But if you want as near as you can get to hands-off, divided attention hovering, SAS Radial Out gives you auto-normalization and will hold the hover.  You can adjust the descent rate with the throttle.  You can 'jink' easily with short taps.  And you can get a faster descent with the Falling Leaf technique, all the while dividing the workload with SAS, which will be working tirelessly to bring your craft back into the hover.  This works because you don't have precision goals other than to be making some progress toward your target touch-down.

The odd man out is only Yaw, with which you will have opposing control.  I'm equally at home with both control schemes with the only confusion being "which one I am in" during a transition.  3-2-1-Fly makes the state obvious in the single-push of a button and that has worked really well in my mental workflow.  (Lotsa credit to swjr-swis here!)

Bottom line is that you are Not Compelled to use Hover [3] if you don't require it or do not choose it.  It's always nice to have the option on standby, I do feel.  And I will add that I very much save Hover for the real Hover and not the approach to a Hover; I regard it as almost a labor-saving device only.

An alternative scheme following the suggestion (don't use a vertical reference) rigidly would just be a subset and I wonder whether eliminating Hover (and usage of one key) would actually simplify the scheme sufficiently to justify the deletion?  Especially when you are building VTOL craft for others to use, offering Vref (in a superset) seems wise.

As a coda (should have mentioned it above), the ascent during a vertical departure is the perfect example in which Radial Out is absolutely all you require since it's purely a simple climb followed by a flat pivot onto the departure heading.

I doubt there is any material disagreement here.

Edited August 4, 2021 by Hotel26

[FleshJeb]

@swjr-swis @Hotel26 This is exactly the type of discussion I was hoping to foster.  No, we don't have any material disagreements.

I wonder if one could retain the horizontal flight control scheme while using a vertical reference by using Action Group Sets. (I'm not sure this is possible.)

• Bind all reaction wheels, control surfaces, and gimbals to KAL-1000s. One for each axis.
• In AGS 0 (horizontal flight), assign each KAL to its respective axis as normal.
• In AGS 1 (vertical flight), assign the KALs and swap the Roll and Yaw axes. Then use the reverse input button for whichever the "odd" axis turns out to be.

Things I'm unsure of:

• Can reaction wheels and gimbals even be assigned per-axis? (I haven't played KSP since April.)
• Can Action Group Sets be switched by a button push?

IF this works, you could transform ANY 90 deg rotation of control axes you wanted.

[Hotel26]

Fascinating direction.  There's a simple mod in here somewhere bursting to get out (as a last or simple resort).  Pity i don't recall any of the three of us being keen on gratuitous mods, but it might make a lovely experiment. 

A mod like that could also add the "missing tweakables", principally the things that can only be toggled but not set to a particular state. 

[swjr-swis]

5 minutes ago, Hotel26 said:

A mod like that could also add the "missing tweakables", principally the things that can only be toggled but not set to a particular state.

KAL might be the stock workaround for (some of) that too: eg. you can deterministically open or close -instead of only ambiguously 'toggle'- service bay doors by assigning the door deployment to a KAL controller.