Tech level 7 SSTO to Minmus surface and back

[pacovf]

Exactly what it says on the tin.

The idea comes from the common (?) wisdom that Panthers can't really get anything worthwhile to orbit without extreme amounts of effort. So I took it upon myself to see how far I can push spaceplanes with only a level 2 R&D centre. My current best is to get onto a Mun orbit and come back, with about 200 m/s delta-v of spare fuel on landing, but it was with a fairly unoptimized design and trajectory, so a Minmus landing shouldn't be impossible. Let me see what you can do!

Rules:

• Vehicle must be a SSTO spaceplane. No splitting the craft at any point during the mission.
• No refueling.
• No parts that require a level 3 R&D centre.
• Vehicle must be crewed.
• Only stock parts.
• No (or at least very minimal) clipping.
• No gameplay mods (aside from informational ones).

Optional constraints:

• Make the vehicle as light as possible.
• Bring as much science from Minmus' surface as you can; if you take science from more than one biome, then each science spot has to be at the very least 30 km away from the previous one (do a suborbital hop, basically).
• No parachutes to land it back on Kerbin.

Hard (impossible?) mode:

• Do the same, on the Mun.

 

If you want a way to score your submission:

Score = Science/Wet_Mass

A screenshot of the vehicle landed on Minmus, with the amount of fuel (etc) leftover in your tanks, and another of it landed (or splashed) back on Kerbin, will be considered a valid submission. If you've landed more than once, then you need to show a picture for each landing. I encourage you to provide a description of your profile ascent too!

Good luck!

 

SUBMISSIONS:

Minmus:

1. ManEatingApe 71.2
2. Wanderfound 44.12
3. Aerospacer 30.5

Mun:

1. ManEatingApe 23.5
2. Cunjo Carl 7.7

Edited May 16, 2017 by pacovf

[*MajorTom*]

it's simple challenge IF SSTO can drop it cargo (spaceship) to  LKO /or split apart . Then go to Mun/Minmus and come back to Kerbin , dock and land.

Or SSTO-ship must be in ONE piece? 

[pacovf]

Must be one piece, no splitting apart. Will update the OP accordingly.

Besides, who likes to spend time docking?

Edited April 8, 2017 by pacovf

[Cunjo Carl]

I'm actually working on this right now for a few questions posed in the gameplay help forum. I'll go ahead and post them here for cross-pollination sake.

@Abastro's low cost, low tech panther spaceplane question : Link

@Tychonoir's low mass, high science SSTO question: Link

If I get anywhere I'll be sure to post! I like the optional constraints section; I'll probably miss most of the marks, though. My work so far has focused on using SRBs for a big kick right after the panther drops off and before the low TWR liquid fuel engines can really kick in. The low drymass of the hammer feels fairly perfectly suited for the task, but the main impediment to its use is its horrific drag. I'm still in proof-of-concept phase so we'll see how it goes!

[ManEatingApe]

@pacovf Here's a simple Panther based spaceplane that I threw together to get you going...

It'll get to LKO with about 2,000 dV total. Not enough for a Mun landing and return (about 2,700 m/s needed) but plenty for a Minimus jaunt (about 1,600 m/s total necessary).

 

Ascent profile is straightforward, climb at 15° initially, the let the node rise naturally to about 20°. Stage the nuke and sparks at about 10km. The nuke on it's own doesn't have enough TWR, but the Sparks add enough punch to push AP far enough ahead in order to circularise.

 

Bob went to Minimus and visited 2 biomes 30 km apart 

 

So I know you insisted for this challenge the biomes be at least 30km apart but normally what I do for a Minimus mission is land on a high plateau then simply roll downhill to each new biome, ending on a flat. Much cheaper than a sub-orbital hop.

My 2 favourite routes, covering 4 and 5 biomes respectively are in the spoiler

Spoiler

 

Full Album

Craft File Download Link

1217.5 science was returned for an initial weight of 17.1 tons, for a total score of 71.2

Edited April 8, 2017 by ManEatingApe

[pacovf]

@ManEatingApe Nice submission! You ended up with a similar design than what I had, though mine had a lot more wing area (couldn't lift off from Kerbin without it) and air intakes.

The 30 km constraint was to avoid people just rolling from one biome to another and artificially inflate their score. I never thought about rolling through 5 different biomes that way though, I will definitely keep it in mind the next time I start a Career game.

[Cunjo Carl]

It's not much for score, but I just managed a Mun shot!

...

The landing wasn't exactly as intended though...

Spoiler

I'd been finding that I was losing a lot of fuel to drag while trying to accelerate using just the LV-N after the Pather dropped off. I originally was planning on using a low-drymass SRB to help fill in the gap. Sadly, the insane mass of the LV-N made the low drymass advantage mute, meanwhile the high drag of the SRBs kept the panther from really hitting its stride.... Dead end!

Swapping out the hammer SRBs with an equivalent mass of terriers sure did the trick, though!

 

It had a lot of deltaV, and with a bit of tweaking, I managed to get it up to orbit with ~5tons (1050units) of fuel.

 

 

Would it be enough? I could only hope, because off I went to the Mun!

 

 

The low TWR landing was much more touch-and-go than I was expecting. Thanks to liberal use of the F9 key, I eventually managed to land near a 5000m mountaintop at the edge of a great crater. Success!

 

 

I planted a flag, took a surface sample and saddled up for the trip back home! Like on a hike, heading home would feel faster than going out, right?

 

 

During a gentle multipass aerobrake, I noticed some odd popping sounds and smoke clouds  from my tail.... uh-oh. I had used basic fins for control and aesthetic edgings because they're just so cute! Unfortunately they didn't like melting, and between losing them and returning 4.5 tons lighter, the plane was completely unstable! Stangely, I didn't notice at first. It flew find while SAS was in control, but the moment I touched the keyboard it would flop like a fish and suddenly deathspiral. Things did not look good. I was just about to have Jebidiah bail out over some convenient ocean, when inspiration struck.

 

 

That little panther in wet mode has the best low speed plane TWR in the game! I pointed vertical, gunned the throttle, and hoped for the best. I won't lie, it took a half dozen tries, but when I finally got the hang of steering that floppy fish, it worked!

 

 

Success! Not the most glamorous return, but it sure beat freefall. The next version of this craft will feature a much longer and more developed tail section and probably a fairing for drag abatement on that LV-N!

The wet mass was 23.46 tons and it only got 180-ish science, so I doubt it will be winning the contest in terms of points, but as a heavyweight Mun goer it was a fun proof of concept!

 

 

[ManEatingApe]

@Cunjo Carl Interesting asymmetrical approach - out of curiosity what speed did the single panther top out at?

PS Loved the landing!

 

 

[Cunjo Carl]

16 minutes ago, ManEatingApe said:

@Cunjo Carl Interesting asymmetrical approach - out of curiosity what speed did the single panther top out at?

PS Loved the landing!
 

Thanks! The panther doesn't want to pass mach1 on its own (without a dive to break through), but it can easily haul the craft up to 7km where the other engines start to become efficient. At this point by lowering the AoA a bit and activating the NERV, the craft can break through and hit that 800-950m/s sweetspot panthers in wet mode do so well at. It's funny, for every 1kN the NERV actually provides, the tiny boost in speed makes the panther 2kN more efficient and the drag 2kN less! (numbers just for effect, I don't know them for real)

After that, with a tilt to the AoA the panther drops off due to altitude, and the terriers kick in at ~15km when drag starts to drop off. Their speed really kicks in from 20-25km and from there its a toodle for the last 600m/s up to orbit. I suspect it's a very similar thought process to your plane but a lot more cantankerous! I especially like your plane's backswept wings with the NERV up front and the light engines out back.... Seems pretty stable!

The asymmetry takes a lot of fiddling to make work right, and even then it makes the craft fidgety in flight. Still, it has its perks!

[pacovf]

@Cunjo Carl: That's really impressive, congratulations on your exploit! I have to say, the asymmetrical layout is very interesting. I assume the objective was to be able to put only one Panther on it to reduce the dry weight?

Also, I would have never guessed that a single Panther could get 23t off the ground! I assume when you say that Panthers do well at 800m/s+, you really mean that's the maximum speed they can achieve, not the cruising speed? I find Panthers tend to start losing thrust when you hit 700m/s or so (depends on altitude).

Any opinions about how to score Mun shots? I think I will just need to keep Mun and Minmus attempts separate, I don't know how to compare them otherwise. Multiple landings are a lot easier on Minmus.

[Cunjo Carl]

1 hour ago, pacovf said:

@Cunjo Carl: That's really impressive, congratulations on your exploit! I have to say, the asymmetrical layout is very interesting. I assume the objective was to be able to put only one Panther on it to reduce the dry weight?

Thanks! That was the intent, but given the shear mass of the single NERV, I'm not sure how much it gains us in the end.

1 hour ago, pacovf said:

Also, I would have never guessed that a single Panther could get 23t off the ground! I assume when you say that Panthers do well at 800m/s+, you really mean that's the maximum speed they can achieve, not the cruising speed? I find Panthers tend to start losing thrust when you hit 700m/s or so (depends on altitude).

It's true! I always pretilt the wings to 5 degrees on my SSTOs, which really helps the anemic thrust keep you aloft. Even so, I think I lifted off only barely before the runway disappeared . I've also been finding ~700 m/s has been the optimum on my flight path, which was weird. I had remembered it being 800-900 so vividly I didn't even think to check! Maybe it was on a lower altitude plane? Since I noticed that, I've been stressing less about increasing speed early, and it's been making life easier.

1 hour ago, pacovf said:

Any opinions about how to score Mun shots? I think I will just need to keep Mun and Minmus attempts separate, I don't know how to compare them otherwise. Multiple landings are a lot easier on Minmus.

I think your scoring system is a good one (especially with the must suborbital hop clause). Just making a separate category for Mun seems like a simple and nice way to do it.

Oh, welcome to the forums by the way. Nice challenge!

[AeroGav]

If you're talking Panther/NERV 've had one of these in my inventory for a while on KerbalX, in the shape of the Voodoo Ray, later modded to become the  Auto Ray (with trim flap).  I made a launch video here when first built, but have since improved my technique with a flatter trajectory , especially after going supersonic, that saves time and fuel.https://kerbalx.com/AeroGav/Auto-Ray

Spoiler

 

I left the throttle wide open as our AP passed over 70km, and didn't back off till we were just under the Mun's SOI to max the Oberth.  The type b nose cones that hold the nukes to the wing tips got hottest, reaching about 73% max thermal.  Not being great with orbital mechanics, I wasted delta V by not matching inclination before launch.

Spoiler

Correction burn was 40m/s. Another 5 m/s to raise our PE above 70km, then we wait an orbit or two till the transfer window is open. 

 

Given our high starting PE, only 108 m/s for the transfer burn and midcourse correction   Spent another 303 on the capture, and deorbit burns.    I decided to go for a horizontal landing, using our Vernier and Oscar B tank..

This experience was fed back into improved undercarriage for the next version ! We had a good 500dV left and could have visited a lot of Minmus biomes by taxying over the surface.  Pretty sure we could have driven to the polar biome, but this is time consuming and you can't quicksave when moving over surface.

Spoiler

 

With a 47km PE, re-entry direct from Minmus was ok.  Made two high drag aerobrake passes, then realised we were going to be landing at night, so shut the service bays, raised gear, and flew the next ones in prograde to keep us aloft longer.   After about 3 passes it became apparent our next one would take us more or less directly above KSC.  If we could loose enough velocity on this one, we could land next pass.  Feeling impatient, i pulled out all the stops, going radial in and holding a sharp negative lift attidue, forcing us below 30km.  This turned out to be overkill and i spent 100-150 dv on the nukes extending our sub orbital flight.   I also ran the panther for a while in dry mode on the final flight up the coast to KSC.   Taking more time, you could have done this much better and not used any fuel here at all.  But , as you can see from the landing,  we certainly got close to KSC

 

[pacovf]

Nice! Did you fly it in the current aerodynamic model? Can you say how much it weighted? Also, at what speed/altitude did you turn on the NERVs?

...that landing video was... interesting.

 

I have to find the time to try a Mun landing...

[AeroGav]

5 minutes ago, pacovf said:

Nice! Did you fly it in the current aerodynamic model? Can you say how much it weighted? Also, at what speed/altitude did you turn on the NERVs?

...that landing video was... interesting.

 

I have to find the time to try a Mun landing...

I linked it on KerbalX, the craft is 22.78 tons.    I think we were about 770 or 780 at 16km when i started the nukes.

Panther looses power very quickly once you exceed mach 2.5 or 14km.   Your close cycle propulsion system has to add a lot of extra velocity just to get orbital.  Fortunately with nukes,  you have quite a high delta V system.

This is my transition to nuke power on a previous flight, though if you look at the screenshots i hid in the spoiler tag,  you can see i also used them to get through sound barrier.

 

[Gordon Fecyk]

On ‎4‎/‎8‎/‎2017 at 2:34 AM, pacovf said:

The idea comes from the common (?) wisdom that Panthers can't really get anything worthwhile to orbit without extreme amounts of effort.

You were just daring AeroGav to respond with that line.

How about the Kerbin 2 challenge done on Tier 7 parts? And returning all parts safely to KSC? That was one of my prouder accomplishments, even if it took fixing the aerodynamic model (Ferram Aerospace) to do it:

Spoiler

 

Don't consider this a valid entry for the challenge, as it does ferry a lander to and from LKO. But conventional wisdom be damned.

Edited April 14, 2017 by Gordon Fecyk
This is not a valid entry for challenge

[pacovf]

@AeroGav: Thanks for the extra details! Why do you fly nearly horizontal for a while with your NERVs on, is it just so that you don't lose the lift from the wings? Also curious about why your time to Apoapsis goes from 14 min to 0s around 11:30, any idea why that is?

Also, everybody seems to be able to reach cruise speeds at altitudes above 10k with Panthers, but every time I try I see my thrust plummet, so I am just flying at 20 degrees all the time until the Panthers give up (around 14 or 15k). 

@Gordon Fecyk: That's a more sensible design than what I am asking for here, but even if it was single stage, FAR would invalidate the submission... I like it though! How did you dock the lander back inside the spaceplane without RCS?

(unrelated: what mod is changing the way your exhausts behave? Your Swivels/Reliant behave funky).

Edited April 15, 2017 by pacovf

[Gordon Fecyk]

Hey @pacovf. I have no problem with this example being invalid; it was more of an example of how to use Panthers in SSTOs effectively. Mark Thrimm's original craft were meant for stock aero, and there's no way I could haul 4.5 t to orbit in the original LTS Kestrel with stock aero. I live by Ferram Aerospace now, having figured out how to land safely after several tries.

The Panthers will fail around 15 km as you've noticed. The Kestrel-FC only works because of sheer momentum after being pushed with four of them. The Sparrow-FC can only reach 13 km by comparison, closer to what you're seeing, even with Ferram Aerospace.

The engine plumes are from RealPlume Stock. It's a visual mod only; the engine performance is unchanged. Be warned though: This will stress your graphics card and processor.

As for docking, I flew that mission twice. The first time Valentina managed to push the lander into the bay and close one of the doors to align the ports. The second time, I backed the lander into the cargo bay's aft end and then had Jeb pitch up until the lander settled into the bay on its own. The docking port magnets took over from there. That second approach was much easier.

Edited April 15, 2017 by Gordon Fecyk
I found the RealPlume Stock thread

[AeroGav]

4 hours ago, pacovf said:

@AeroGav: Thanks for the extra details! Why do you fly nearly horizontal for a while with your NERVs on, is it just so that you don't lose the lift from the wings?

Two reasons - first is cosine losses, second is drag.

Cosine losses means engines thrusting downward rather than horizontally, fighting gravity rather than adding to forward velocity.  It is far more efficient to get lift from the wings instead, because wing parts generate more than 10 pounds of lift for every pound of drag they create, even in supersonic flight.  This craft has quite a lot of wing area, and the wings are angled upward as they attach to the fuselage, so it can get enough lift to fly with the fuselage nearly horizontal.

Second reason is drag.  In KSP, the drag generated by a part is affected by three things.  Mach number, air density, and AoA.   Over 90% of the drag comes from the fuselage in KSP aerodynamics, so i want to keep fuselage AoA as close to zero as possible.  For wing parts themselves , optimum Lift to Drag ratio happens at about 7 degrees AoA,  so the wing parts are angled up when attached to fuselage so i can have them make lift at a near-optimum AoA while the fuselage is flying prograde.

4 hours ago, pacovf said:

Also curious about why your time to Apoapsis goes from 14 min to 0s around 11:30, any idea why that is?

I came to this game as someone who played flight sims rather than rocket sims, so i tend to look at the vertical speed indicator rather than the time to AP.      

From 10:14 to 11:30 we are descending so the "time to AP" is going to be pretty meaningless.  I think the game assumes you are currently at PE so your AP must be on the other side of the planet somewhere.    After 11.30 we are starting to climb again, so "time to AP" goes to zero.  This is a prediction based on a ballistic trajectory, not taking account of lift or thrust.  Even though  we were climbing at this point, if the wings were decoupled and the engines stopped, we'd start descending again very quickly, which i suppose is correct !

What's happening is we'd got ourselves into a zoom climb situation.  I'd started the nukes at about 12km, suddenly adding extra thrust, we were going sufficiently fast in the lower atmosphere that our lift was significantly exceeding our weight.   I'd exacerbated the situation by pitching the nose up slightly, creating more lift.  So, our vertical speed starts increasing and we get into a steep climb.  At 9.14 we pass 19500M after which the air is too thin to get enough lift at 800 m/s,  for the next two minutes our lift is less than our weight and our vertical speed starts decreasing, though our momentum carries us up past 26km.

Eventually, that descent is arrested because

• as we fall back to 22km the air gets thicker again
• we are getting faster so more lift is created
• we accelerate from about 30% of orbital velocity to around 60% of orbital velocity.  At 60% of orbital velocity, orbital freefall is supporting 60% of the airplane, we only need to find enough lift to support the other 40%.
• minor effect, we are getting lighter as propellant is expended.

The thing you need to understand about this kind of TWR < 1 spaceplane ascent to orbit, is that it needs two things to work.

First, you need to generate sufficient lift to avoid falling back into the lower atmosphere.     At the beginning of the closed-cycle climb, lift needs to be almost as high as your weight, but as you get closer to orbital velocity , freefall effect supports more and more of the vessel.  Eg. at half of orbital velocity, it supports half the vessel.   This enables you to fly higher where drag is less...

Second, thrust needs to be > drag.   Preferably, significantly higher than drag, or you end up with "gravity losses".    

[AeroGav]

5 hours ago, pacovf said:

Also, everybody seems to be able to reach cruise speeds at altitudes above 10k with Panthers, but every time I try I see my thrust plummet, so I am just flying at 20 degrees all the time until the Panthers give up (around 14 or 15k). 

Thrust declines with altitude  but it also increases with speed.  I climb subsonic to about 8km then shallow dive though mach 1 with a burst from the nukes.   If i tried to keep climbing subsonic, i wouldn't get enough thrust , and also to get enough lift at such a low speed and high altitude i'd be pulling the nose up to a steep angle which creates lots of drag.  On the other hand if i tried to go supersonic at sea level drag would build up quicker than thrust and i'd probably get stuck at 280 m/s.

remember

1. thrust declines with altitude, but more slowly than drag up to 14km , provided you got enough wing area to maintain nose angle near prograde.  13-14km is best altitude for max speed in level flight.

2. thrust increases with speed up to mach 2.5, then falls off fast.

3. between 85% and 125% of the speed of sound is the transonic region, which has very high drag.   Above 125% speed of sound, drag is lower than when transonic, though not as low as it is when subsonic.

4. keep the body of the airplane as close to prograde as possible for low drag.  Certainly not more than 5 degrees difference between where the nose is pointing and where you are going.

My supersonic flight transition in detail - 

1.  Subsonic climb for the first few minutes.  I want to keep AoA low, but also not exceed 240 m/s and get into the transonic region.

By this point, it's becoming impossible to avoid one or the other, the air is too thin.   I either need to let my speed increase to transonic region, or  pitch the nose up more to get enough lift to keep climbing.  Both of these would result in high drag

So, I use an action group to deploy some trim flaps that lower my nose from +2.26AoA to -1.1AoA.  This reduces lift and drag, causing us to arc over to a shallow dive, and i trigger the nukes briefly.

With the extra power, and flying "downhill", we accelerate very quickly through the high drag transonic region.    At Mach 1.3 , we are through the worst of the transonic drag.  The Panther is also starting to benefit from the "ram air" effect and is producing much more power.  I cancel the nukes as they are no longer needed, and start easing the nose back up so we can resume a climb

[pacovf]

Thanks for the indepth response! Very interesting considerations, I will definitely try to get more horizontal in my next attempt. I know about the sound barrier (usually my way to punch through it is just to add moar Panthers, probably not the best way), my problem is that once I reach ~750 m/s at 10km my thrust starts decreasing (even if I keep the plane level), and then so does my speed. So instead I just keep climbing rather than waste speed for no reason. Will see if adding wing area and angling it with respect to the fuselage helps.

Note that, unless my math is wrong, the "freefall effect" increases with the square of the velocity, so at half the escape speed, it's only supporting one quarter of your weight. Don't think it changes anything to your post, but it emphasizes the importance of lift area to support you while you get to escape velocity (or of TWR if flying a rocket instead).

[AeroGav]

Yes 750 m/s really is the limit for Panther

once you reach that, you may as well use any excess power to climb rather than stay level and have it tail off anyway.  On my first flight in the Ray (the one in the video), i start the nukes up as soon as i hit 750.   On my Minmus mission, i ease into a climb but i toggle the nukes on  and off with an action group to maintain about 750-800, and try to keep thrust/weight more constant so as not to get into a zoom climb/dive cycle.   

With a single nuke / 2 panthers,  you want to try get into a zoom climb to give you more time to circularize, but with 2 nukes the Ray is most efficient just cruising up as steadily as possible.

 Re:  Mun mission

According to the community delta V map that's another 960dV required vs Minmus landing mission.   I don't know if it's possible. 

I've Ray actually has better TWR than a lot of my spaceplanes, but i don't know if you could simply add enough extra fuel to it.

With one panther for it's 22 tons it gets through mach 1 pretty easily.   (Accelerates from 0.84 to 1.3 mach in 30 seconds, which i think is faster than an F-15).  The most underpowered ssto i made weighed 50 tons and had one rapier, 2 nukes.  Had to run the nukes for a couple of minutes to get supersonic, which cost a fair bit of fuel.   So there's limits to how much fuel you can add here before you're into adding extra panthers.

In terms of getting to orbit, like i say, thrust needs to > drag.    Low drag mk1 with even more wing area, can get 40 tons to orbit with 2 nukes.   But , that was a sandbox ship, was using Big S parts,  all those wings were helping with fuel stowage.  At tech level 7 they are just dead weight, and a mission to the Mun has much of the delta v requirement out of the atmosphere.

On the other hand if you add extra fuel tanks but no extra wings, the Ray would be lower in the atmosphere for any given mach number, so it would have more drag from the liquid fuel tanks (remember, 90% of drag is from fuselage).  It would also get heating effects worse.

A third option is to add a terrier booster engine.    It only needs enough fuel to boost you to mach 4 or so, and orbital freefall supports more of your weight,  meaning you fly higher, meaning less drag, and you can get away with a lower nuke TWR.  So potentially, adding 1 ton of oxidizer and half a ton of terrier might mean you can leave 3 tons of nuke behind.    However, you still need enough nuke twr for powered landing Mun (though should be easy as most of your fuel gone by then).

Incidentally, i made a 2 stage to orbit ship with 2 nukes and 1 terrier in the upper stage.  It's unfinished but the upper stage works really well, despite being launched from a Wheesley carrier that only gets mach to 450 m/s or so.  The oxidizer for the Terrier runs out at about 1000 m/s. Then again, it  is not dragging dead jet engines to orbit, and it abuses Big S wings.

finally, if i was doing a moon mission with an airplane, i might take the easy way and abandon 100% re-use for such a historic flight.

Terriers are really cheap - 450 funds.  LF/O tanks are worth half as much empty as when full, so given their low empty weight, they are not good candidates for disposal.   But LF tanks are worth very little when dry.  So I might put some engine pods under each wing with a small hardpoint, made with LF tanks and terriers.     The oxidizer tanks can stay in the main body and be carried to orbit.

 

[pacovf]

I am aware that making it SSTO is putting extra constraints on the relatively low performance of the Panther at high speeds and altitude, that's why it's a challenge  I came across the Kondor while looking for inspiration for this challenge, it's a beautiful craft.

But hey, Cunjo Carl managed to land his spaceplane on the Mun, so it's possible! I have a feeling using Sparks (like ManEatingApe did) instead of Terriers to boost yourself to orbit might be more efficient. My version used 2 Terriers, and I have a feeling I wasted a lot of thrust due to having to go into a zoom climb (hence very little lift) because of the relatively short but powerful boost at a low altitude. Maybe only one Terrier works too, but then you run into symmetry problems.

Edited April 15, 2017 by pacovf

[AeroGav]

It's a tough one.   Really hard to say no to the Terriers knockout ISP.   But the Sparks are smaller, more granularity in matching to size of ship.  Unless building really frickin big.

[pacovf]

The way I see it, you want to maximize your delta-v once you are already in orbit, which means that, for the same amount of fuel, you want the parts that get you into orbit to be as light as possible. Sparks have nearly twice the TWR of Terriers (1/5 the weight but 1/3 the thrust), while having an ISP that is close to the Terrier (320 to 345 at vacuum, though it's 270 to 85 at atmospheric pressure, so maybe they are a bit closer at the altitudes you start them). Since the amount of fuel you need increases exponentially with the delta-v, it will depend on the exact weight of the craft which engine ends up with a lower dry weight (sparks will need more LF/O).

It's clear that, if you want to land on the Mun this way, either you have a very good intuition about what works and what doesn't, or you need to run some numbers. The delta-v left once on orbit seems to be the key parameter to optimize.

[Wanderfound]

4 hours ago, pacovf said:

It's clear that, if you want to land on the Mun this way, either you have a very good intuition about what works and what doesn't, or you need to run some numbers. The delta-v left once on orbit seems to be the key parameter to optimize.

As a datapoint: I was messing around with this yesterday, and the Panther/Nuke/Spark combo is working well. I'm getting about 2,500m/s ΔV in LKO from an easy-to-fly design with four Sparks, one Panther and one Nuke. Nuke on the centreline, Panther mounted above it. The Panther's gimballing ability lets you get away with quite a lot of thrust offset.

If you allow a few 0.625m drop tanks, you could probably push that to 3,500m/s or more.

Edited April 16, 2017 by Wanderfound