Jump to content

eddiew

Members
  • Posts

    3,659
  • Joined

  • Last visited

Reputation

4,427 Excellent

5 Followers

Profile Information

  • About me
    the one with the ears

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. That is... bizarre, and I may actually try it out in mk3 format if I can make it look good My other thought was top-mounting a small cockpit on the fuselage where it would be out of the airstream. OTOH, I feel like I'm fixing a problem specific to this simulation of thermal spread, and maybe I'll just raise the temp threshold a couple of hundred degrees to get some leeway.
  2. Update Well it's been 2 years and a day since I posted anything in this thread. I think we can assume my previous career is dead But I have recently been noodling with KSP, if only in sandbox, and I'm particularly proud of this very-nearly-stock SSTO for JNSQ, so I feel like I want to log it here. But before I get into the flight log: I have previously landed this airframe at 100% stock. That's screenshotted somewhere on the WDYDIKSPT thread about a week prior to the date of this post. I have really had a hard time repeating that however. Every subsequent attempt, the quad adapters for the rapiers have overheated. Since there isn't really much alternative to these, I have instead added a tiny patch that ups their thermal tolerance by 400 degrees, and increases their mass by 25%. So this is not quite stock anymore; but I hope the added mass reasonably balances it out. They're only structural parts, there's not a good reason they can't be made a bit tougher. The ascent is just a little heavier, the descent is vastly safer. Stock parts are very close to being able to handle JNSQ's scale, but the margin of error is too tight in this case. I am considering also patching the cockpit itself, because it does get very, very close to its threshold on descent. I guess I'll see with other airframes. I do wonder if a radiator on the top might not also be a big help. Ascent A good, reproducible ascent under JNSQ made me throw away everything I knew about SSTOs in this game. Speed on air doesn't matter nearly as much as altitude. The rapiers are there to get us to 20km. Whether we're at 1200 or 1500m/s doesn't matter so much as our vertical velocity is as high as we can get it - ideally upwards of 150m/s by 20km. So, off the runway; stay low. Hotwings is heavy and rapiers have poor thrust when stationary. She'll unstick near the end of the runway, not before. Rear and mid flaps deployed, just coast barely gaining altitude, but picking up speed. Use deployment on the tailfin to correct any yaw problems once off the ground; we can't afford manual inputs because they're always too strong and the nose will drop. Around 400m/s, cut flaps, use forward canard deployment to begin a smooth ascent; again, no manual input. Aim the nose ~9 degrees above the horizon, which may only need 10 degrees on the canards. From here, the ascent is dead-stick. Honestly, the less those controls are touched, the happier it will be. Around 15-16km, the vertical velocity will stop increasing; add the nervas. Very soon after that, it'll stop increasing again; add the vector. Hands off that stick. If everything is going well, vertical velocity won't drop below 120m/s. If it goes below 100, we might not be going to space today. The pitch creeps up from 10, to 15, brushing 20; let it. We've got to have altitude in JNSQ. There is no point doing 2500m/s 'down' at 20km and on the level. With a payload below 6,000kg, stay hands off until about 50km, at which point gently bring the nose to prograde. We'll get about 200m/s more out of the vector. It's not critical, but it is efficient and it gives us more leeway on descent. With a payload above 6,000kg, just let the vector burn out before touching the controls. Rather than turn prograde with the nervas, we'll need to keep about a 10 degree pitch above the horizon. That's how close this ascent is to failing. Hotwings 4 can carry up to 14 tons, but it's really pushing the fuel budget and we might have to do the descent completely unpowered. It's possible, but I'm not sure I'd call it fun. The AP might start coming closer when the vector burns out, but that's ok; by 65km, Hotwings will keep the AP about a minute ahead, as long as the nervas are burning. (Oh, that's another thing about JNSQ ascent; we pretty much keep the engines on until we hit vacuum.) With this profile, a simple 5.3t relay can be lifted to a 90x90 orbit, with about 600m/s still left in the tank, with that cargo still aboard. There's enough leeway there to rendezvous with something around the 120-150km mark. Heavier cargos really don't have any orbital options however; although they will go further under their own steam. My standard 5.3t relay carries 3,800m/s and can go to Minmus, Mun, Duna, or Eve (probably an eccentric orbit at this last, but it should be an orbit). I'm not really sure how much range we could pack into the maximum 14 ton mass and fit it in that cargo bay... maybe I will try to work that out some time. Descent I wish I could say it was another hands-off experience, but it's not. The best profile I can come up with requires a high (over 30 degrees, ideally 45 degrees) AoA, which mandates manual input. Which does tend to make the nose bounce. Once we touch the controls (which we'll have to eventually), best advice is just pull back on the stick all the way down. It will wobble 30-40 degrees, but that will be fine. Anyway, descent burn; there's a handy equatorial lake to use as a landmark, west northwest of the desert airfield. Burn until we have an impact point that is about 1/3 of the way between KSC and the island airfield. If all goes well, we won't need any fuel at all to land with. Set rear and mid flaps to 30 degrees, canards to max, and deploy them all. The inner air brakes can be used; the outer ones will burn off, so keep them down (set deployment angle to zero). Go ahead and slap the atmosphere at a solid 45-50 degree AoA. By 60km, we're 20 into the atmosphere, but this aggressive AoA generates enough drag that no velocity has been gained in the descent. By 43km (probably wrestling the S key all the way), we've actually shed about 1km/s, and things are starting to get toasty. So, there isn't a big margin of error here. The cockpit will peak around 1430-1440 degrees, and is rated for 1500. Experiments were run with longer descent paths and lower AoAs, but the heat just creeps up for longer and still exceeds the threshold. Best as I can figure, this is actually the safest flight profile. It's a little tricky and needs to be flown by the numbers, but after three attempts, all three came down smoothly. Eventually by a little over 30km, we should be well under 2km/s, and starting to cool off. Cockpit temperature continues to rise even as other parts are cooling off - not really sure why that happens! But it should stay within its tolerance. Just. By 1400m/s we can pretty much fly as we like, based on our distance to KSC. Using the desert lake as a visual landmark for the descent burn, Hotwings should get to the runway without any input from the engines. Touchdown is better now that there are added wheels under the nacelles, rarely swerving or causing any explosions. On the whole, I feel like this is as good as this design can deliver - but I really need to go find out if anyone's lifting larger cargos in SSTOs, and especially if they didn't need to slightly patch some parts to do it Also, to the JNSQ team: I apologise. Please don't delete my craft file.
  3. Spent the last week stuck in JNSQ SSTO maximum load and reproducibility testing. (I.e. how much can it carry, and how to bring it home reliably.) Ok, before I get into the flight log: I have previously landed this airframe at 100% stock. That's screenshotted somewhere a few pages back. I have really had a hard time repeating that however. Every subsequent attempt, the quad adapters for the rapiers have overheated. Since there isn't really much alternative to these, I have instead added a tiny patch that ups their thermal tolerance by 400 degrees, and increases their mass by 25%. So this is not quite stock anymore; but I hope the added mass reasonably balances it out. They're only structural parts, there's not a good reason they can't be made a bit tougher. The ascent is just a little heavier, the descent is vastly safer. Stock parts are very close to being able to handle JNSQ's scale, but the margin of error is too tight in this case. I am considering also patching the cockpit itself, because it does get very, very close to its threshold on descent. I guess I'll see with other airframes. I do wonder if a radiator on the top might not also be a big help. Ascent A good, reproducible ascent under JNSQ made me throw away everything I knew about SSTOs in this game. Speed on air doesn't matter nearly as much as altitude. The rapiers are there to get us to 20km. Whether we're at 1200 or 1500m/s doesn't matter so much as our vertical velocity is as high as we can get it - ideally upwards of 150m/s by 20km. So, off the runway; stay low. Hotwings is heavy and rapiers have poor thrust when stationary. She'll unstick near the end of the runway, not before. Rear and mid flaps deployed, just coast barely gaining altitude, but picking up speed. Use deployment on the tailfin to correct any yaw problems once off the ground; we can't afford manual inputs because they're always too strong and the nose will drop. Around 400m/s, cut flaps, use forward canard deployment to begin a smooth ascent; again, no manual input. Aim the nose ~9 degrees above the horizon, which may only need 10 degrees on the canards. From here, the ascent is dead-stick. Honestly, the less those controls are touched, the happier it will be. Around 15-16km, the vertical velocity will stop increasing; add the nervas. Very soon after that, it'll stop increasing again; add the vector. Hands off that stick. If everything is going well, vertical velocity won't drop below 120m/s. If it goes below 100, we might not be going to space today. The pitch creeps up from 10, to 15, brushing 20; let it. We've got to have altitude in JNSQ. There is no point doing 2500m/s 'down' at 20km and on the level. With a payload below 6,000kg, stay hands off until about 50km, at which point gently bring the nose to prograde. We'll get about 200m/s more out of the vector. It's not critical, but it is efficient and it gives us more leeway on descent. With a payload above 6,000kg, just let the vector burn out before touching the controls. Rather than turn prograde with the nervas, we'll need to keep about a 10 degree pitch above the horizon. That's how close this ascent is to failing. Hotwings 4 can carry up to 14 tons, but it's really pushing the fuel budget and we might have to do the descent completely unpowered. It's possible, but I'm not sure I'd call it fun. The AP might start coming closer when the vector burns out, but that's ok; by 65km, Hotwings will keep the AP about a minute ahead, as long as the nervas are burning. (Oh, that's another thing about JNSQ ascent; we pretty much keep the engines on until we hit vacuum.) With this profile, a simple 5.3t relay can be lifted to a 90x90 orbit, with about 600m/s still left in the tank, with that cargo still aboard. There's enough leeway there to rendezvous with something around the 120-150km mark. Heavier cargos really don't have any orbital options however; although they will go further under their own steam. My standard 5.3t relay carries 3,800m/s and can go to Minmus, Mun, Duna, or Eve (probably an eccentric orbit at this last, but it should be an orbit). I'm not really sure how much range we could pack into the maximum 14 ton mass and fit it in that cargo bay... maybe I will try to work that out some time. Descent I wish I could say it was another hands-off experience, but it's not. The best profile I can come up with requires a high (over 30 degrees, ideally 45 degrees) AoA, which mandates manual input. Which does tend to make the nose bounce. Once we touch the controls (which we'll have to eventually), best advice is just pull back on the stick all the way down. It will wobble 30-40 degrees, but that will be fine. Anyway, descent burn; there's a handy equatorial lake to use as a landmark, west northwest of the desert airfield. Burn until we have an impact point that is about 1/3 of the way between KSC and the island airfield. If all goes well, we won't need any fuel at all to land with. Set rear and mid flaps to 30 degrees, canards to max, and deploy them all. The inner air brakes can be used; the outer ones will burn off, so keep them down (set deployment angle to zero). Go ahead and slap the atmosphere at a solid 45-50 degree AoA. By 60km, we're 20 into the atmosphere, but this aggressive AoA generates enough drag that no velocity has been gained in the descent. By 43km (probably wrestling the S key all the way), we've actually shed about 1km/s, and things are starting to get toasty. So, there isn't a big margin of error here. The cockpit will peak around 1430-1440 degrees, and is rated for 1500. Experiments were run with longer descent paths and lower AoAs, but the heat just creeps up for longer and still exceeds the threshold. Best as I can figure, this is actually the safest flight profile. It's a little tricky and needs to be flown by the numbers, but after three attempts, all three came down smoothly. Eventually by a little over 30km, we should be well under 2km/s, and starting to cool off. Cockpit temperature continues to rise even as other parts are cooling off - not really sure why that happens! But it should stay within its tolerance. Just. By 1400m/s we can pretty much fly as we like, based on our distance to KSC. Using the desert lake as a visual landmark for the descent burn, Hotwings should get to the runway without any input from the engines. Touchdown is better now that there are added wheels under the nacelles, rarely swerving or causing any explosions. On the whole, I feel like this is as good as this design can deliver - but I really need to go find out if anyone's lifting larger cargos in SSTOs, and especially if they didn't need to slightly patch some parts to do it Also, to the JNSQ team: I apologise. Please don't delete my craft file.
  4. It's an interesting idea, but I don't really want to have to calibrate for every descent possibly bouncing and needing to loop back. That's more fuel than I want to try to carry with JNSQ. Margins are already kind of tight... (Although I'm pretty sure my mk3 design could climb to orbit, land, and go to space again on stock Kerbin.) For my own purposes, I have made a little patch that raises the mk1 cockpit's maxTemp to match the mk3, and the skinMaxTemp to a bit less than the mk3, and add 10% mass by way of paying for it. It is 'enough', but doesn't particularly let you take any liberties. Realising that the inline cockpit is lighter than the pointy one, I swapped it in on my mk1 design and gave it a shot. It is... a bit of an unnerving descent, some thermals get within a hundred degrees of failure, but it does survive the drop. Again, a very high 30-40 degree AoA seems to do the trick with spreading out the heat and preventing the shock intakes from melting. Unfortunately this one does need some jockeying at higher altitudes, since it doesn't have any reaction wheels and doesn't like holding that inclination without manual input. Somewhere around 2400m/s it settles out and can be allowed to drop to a more sedate 25-30 degrees, which it will hold on its own. The mk1 seems a lot draggier for it's weight than the mk3, and this descent fell short of target and needed a bit of flight time to get back to KSC. Fortunately it can be nudged with the rapier at 17-20km up and get a lot of mileage out of the 10% (160 units) of fuel that remained. This is of course, not a useful spaceplane. It doesn't carry any cargo, doesn't have a docking port, and doesn't have enough spare fuel to pipe into anything else. But it is a proof of concept, working spaceplane (not actually an SSTO) for JNSQ with only a very slight thermal tweak, and I'm basically ok with it if the little ones don't work so well at this planetary scale. That my mk3 design flies to space and back (shown a few posts ago) is a massive surprise to me, and I'm happy with that
  5. Figured I'd give my Hotwings 4-ST SSTO for JNSQ a try at 100% re-entry heating to see what explodes and needs patching. All parts in use are currently set to stock thermal tolerance. The rapiers are accidentally 10% heavier because I forgot to comment that out when resetting their thermals, so that added about 1.6 tons to the overall mass, but that's only about 1% of launch weight so no biggie. Same old flight profile on the ascent. With flaps deployed, Hotwings 4-ST will unstick before the end of the runway. Fly flattish up to ~250m/s, at which point drop the flaps and begin a gentle 5-10 degree climb to 6km. At this point the rapiers are hitting their power band and push transonic. Don't worry too much about maxing out on rapiers, 200m/s here or there disappears into the overall delta-v requirements under JNSQ and this vessel usually hits LKO with 700m/s spare. Kick the nervas and vector engines around 18-20km. By this point the nose has probably risen to about 10 degrees above the horizon; let it continue to rise during the rocket burn and by the time the oxidiser runs out the AP should be past 60km and even the tiny LV-N thrust is adequate to go the rest of the way. Altitude is important for both thermals and drag reasons, the nervas can't fight the atmosphere at 40km. (And yes I do tend to fine control vessel pitch by adjusting the deployment on the canards; it leads to a lot less jumpy behaviour and doesn't get into the control wobble state where you have to wrestle the stick all the way up. Most of the ascent is hands off.) Deploys a relay satellite in LKO as usual before preparing for a descent approximately 90 degrees around the planet from KSC, setting a point of impact somewhat west of the runway. (I find these much easier to judge than long shallow glides.) Roughly the same 30 degree AoA as the actual space shuttle feels good. All flaps deployed and a pair of 2.5m SAS units seem capable of holding it as long as you don't manually fiddle the controls. (Seriously don't fiddle the controls. I'm never sure whether this is an oddity of KSP or something I don't understand about aerodynamics, but if you touch that pitch, it's going to drop the nose real hard and you'll be jockeying it the rest of the way down. Leave it alone and it's fine.) Anyway, let's see where the hot spots are. The quad couplers are currently the scariest red, with some deep orange thermals across other components. Ok, so maybe the previous rapier patch would be good, and a similar touch for the cockpit might be nice. Wait a moment... What do you mean it's cooling down and nothing blew up yet? Ok, so it does get very, very close to the quad couplers' thermal limits; I guess they're a bit blunt and they stick out into the airstream under the vessel, so they're taking a beating - but they just barely held together. I might patch them just a bit more mass+tolerance, because that wouldn't feel deeply cheaty to me. You can't tell me those are solid blocks of steel and couldn't be made a bit sturdier. I did whiff the landing a little, but it's on the runway with all crew present and that's good enough. I'm sure we can get back to using the KSC just as soon as the radiation from the exploded nerva clears up. (Side note; I might put some landing gear under the nacelles to prevent this happening. Not sure if it's wing flex or that I'm misjudging how close the rapiers are getting to the tarmac, but it's happened a lot with this airframe.) But anyway, er... stock mk3 SSTO for JNSQ. An accidental success story.
  6. Noodling around, wondering exactly how much work is involved in making patch files for MK1, 2 and 3 parts such that a stock spaceplane can survive ascent and re-entry in JNSQ without heating changes. Started with mk1; the cockpit is the biggie. 2000 skin temp and 1500 (core?) temp limits are just too low for handling mach 12 in any kind of atmosphere. After a little experimentation, bumped those up to 2400 (less than stock mk3) and 1700 (200 more than stock mk3). I thought I was also increasing the skin temps of the mk1 LF tank and rapier... but they don't actually seem to have skin temps by default. Since their max (core?) temp is 2000, this seems completely fine. I'll remove my configured skin temps (2400) and test again. Max temps on landing gear raised from 2700 to 2900 - though I'm not sure this was necessary. The dry masses of these parts have been increased by 10%, which hasn't helped an SSTO who's delta-v was already struggling, but it is possible to get my previously-posted Hotwings airframe to orbit with about 380m/s remaining. The flight profile changes a bit; stays low off the runway, builds speed to about 200m/s, aims at 5 degrees and basically holds it all the way. The under-wing shrimp SRBs kick in about 20km, shortly followed by the nervas. Kicking off the shrimps, TWR drops to about 0.8, which results in a slight altitude dip around 40km, but eventually the forward velocity starts to carry the ground away faster than Hotwings is falling, and eventually it ends up orbital within the atmosphere. It does get hot, but with the new settings, nothing explodes. For re-entry, it goes with a solid 30-40 degree AoA, which seems to de-focus the heat from the cockpit and allows it to dump speed really quickly, such that by 30km it's well within its thermal comfort zone. I got the landing coords horribly wrong, but that's another issue and I don't need to address it right now. But now I'm pondering just how close the stock mk3 I built is to handling the heating of JNSQ... it might only be a couple of hundred degrees short. One to experiment with
  7. SAS is on Though to be fair, turn it off and you might settle into a stable all-feet configuration that feels better to the eye.
  8. Early experiment with adding ablator (and thermal tolerance) to a mk1 cockpit and flying to orbit with JNSQ at 100% re-entry heating. Even Jeb seems a little concerned at the kinks that haven't been ironed out of this one. Maybe he has reservations about sending a pair of nuclear engines into an uncontrolled descent back to Kerbin. (Honestly, I suspect I'll end up adding thermal limits and ablator to so many parts it's probably not worth the effort, and I should just keep the heating setting dialled down. But it's been a fun experiment, and I learned a whole new flight profile where this spaceplane doesn't even hit the SRBs until supersonic at 10km. Now it uses them to get as much vertical velocity as possible once the rapier maxes out, rather than to get off the runway. I didn't know it could get off the runway without the SRBs!)
  9. There is now a setting for same vessel interaction... I have yet to try it.
  10. Oh dear, now I feel compelled to create a station with a very long robotic arm that can Klaw any vessel within 50 metres... but that might have to wait until I actually kick off a proper game instead of just noodle around in sandbox. Mostly I was thinking to have a big ore station in LKO, and a small blacksmith 'pod' that would float around in orbit and glomp on to any vessel that needs the attention Although especially using modded engines, I'm sure it's possible to build a useful SSTO that also carries a docking port and has balanced RCS. Oh hey, I remember spirographs...
  11. Interesting... I might give this a go. I was pondering whether to make some thermal tolerance patches using OPT's parts as my guideline, but the idea of an expendable heat coating feels quite good. Although that does mean the SSTO needs to be rendezvous/docking capable... or that the ablator factory needs to have a Klaw and be able to come to them. Which might be the best answer. Probably need about 1000 units of ablator on those frikking air brakes though
  12. Literally the day I needed it, thank you! I was wondering how to survive my new game without a porkchop plot
  13. I could try burning out all my remaining fuel, would take orbital speed down to about 3050m/s. Might be survivable - although the drop is going to be much harder, of course.
  14. If you want to try doing that, feel free My gut feeling is that you'd have to be almost skimming the atmosphere when you drop the probe, and have it decelerate really hard so's to be in the clouds very quickly. You'd also likely end up doing this either at the daylight terminator, or into the night side, because you're coming from close to the sun and would reach Joolian periapsis around the back of it. This means you have no control, because Kerbin has got to be on the sunward side. Although I guess the Galileo probe had no concept of remote control, so it might well have just dumped all it's data to the orbiter for later transmission.
×
×
  • Create New...