Drag Revolution

[Superluminaut]

So we all know the drag model in KSP is not so great, but I wanted to know what KSP would be like with more earth like drag.

I did some quick research and real rockets typically lose 2% launch ÃŽâ€V to drag. With gravity, losses are about 16% launch ÃŽâ€V.

Kerbin atmosphere is about 4 times larger than that of earth with respect to the planet. So drag losses could be approximated at 8% ÃŽâ€V, with a total of 22% ÃŽâ€V losses.

What does this mean?

In earth like drag KSP, if you want to launch to a 70km orbit, 2290m/s, it would take you about 2936m/s to do so. A far cry from the 4500m/s it takes now.

The question comes to mind, will the final drag model revolutionize launch requirements, or will it attempt to mimic current requirements?

Edited September 17, 2013 by Superluminaut

[leax256]

well the way the ship is built will also affect the actual peformance, ie. the noce cone will decrease drag overall.

play with FAR(farom's aerospace research) for a look a kinda what it will be like.

[stupid_chris]

Correctly rigged up rockets in far can get to orbit for as little as 3300m/s afaik. A better drag model is indeed gonna help a bunch.

[loknar]

+1

http://forum.kerbalspaceprogram.com/showthread.php/20451-0-21-Ferram-Aerospace-Research-v0-9-6-2-Aerodynamics-Fixes-For-Planes-Rockets

[Kosmo-not]

Your percentages are off by two orders of magnitude.

[AmpsterMan]

Your percentages are off by two orders of magnitude.

Some how his calculation if correct though.

From my limited experience with FAR, it does seem like Delta-v reqs would be smaller. I personally wouldn't mind that as it will allow us to send larger payloads into orbit.

[Kerbart]

Kerbin atmosphere is about 4 times larger than that of earth with respect to the planet. So drag losses could be approximated at 0.08% ÃŽâ€V, with a total of 0.22% ÃŽâ€V losses.

How does this exactly work? There's still about 50km of atmosphere the rocket must battle, about the same as on Earth (yes there's atmostphere above 50km but it doesn't really induce that much drag)

The question comes to mind, will the final drag model revolutionize launch requirements, or will it attempt to mimic current requirements?

Well, if the atmospheric drag model becomes vaguely realistic, the following will happen:

• Four levels of asparagus staging will become deadly as the short-but-stubby Kerbal designs are no longer effective (too much drag)
  
• Enjoy how your rocket explodes in a spectacular fashion as it hits the sound barrier
  
• If you survive that, enjoy how the atmosphere rips your rocket to pieces as a rapid pitch to 45° at 10,000m suddenly is not that good of an idea when the velocity vector is still pointing straight up.

I'm sure a realistic drag model will revolutionize launch requirements. But likely in a different way than projected by the OP

[Superluminaut]

My bad, fixed it.

[Van Disaster]

FAR removes most of the dV loss from drag, yes ( given how Kerbin's atmosphere is much thinner than earth, if you're comparing the two, I suspect pretty realistically ). I bank on losing 50m/s at most to drag, and start turning at 300m off the pad going full throttle all the way. FAR is actually a bit overly kind atm, some of the "fat bodies cause tons of drag" code isn't functional atm but it will give people a good idea how realistic launches would go. Rocket design is quite important, especially if you don't want it tumbling.

[Superluminaut]

How does this exactly work? There's still about 50km of atmosphere the rocket must battle, about the same as on Earth (yes there's atmostphere above 50km but it doesn't really induce that much drag)

Well, if the atmospheric drag model becomes vaguely realistic, the following will happen:

• Four levels of asparagus staging will become deadly as the short-but-stubby Kerbal designs are no longer effective (too much drag)
  
• Enjoy how your rocket explodes in a spectacular fashion as it hits the sound barrier
  
• If you survive that, enjoy how the atmosphere rips your rocket to pieces as a rapid pitch to 45° at 10,000m suddenly is not that good of an idea when the velocity vector is still pointing straight up.

I'm sure a realistic drag model will revolutionize launch requirements. But likely in a different way than projected by the OP

It's just a rough estimate. Maybe a better estimate would result by comparing the scale height of Kerbin atmosphere vs Earth atmosphere. However I think more precision would only show even lower ÃŽâ€V costs, this estimate is very generous yet still clearly makes the point.

Interesting thought about sudden changes in direction at speed, I wonder how the new gyro SAS would do during launch.

[arq]

Just since it hasn't been mentioned yet, a stock launch requires right about 4500m/s for most profiles. I haven't done much with FAR, except for a few basic planes, but using the above 3300dV figure we can make some bounding statements about efficiency.

Let's say LKO is 80km. A circular orbit at this altitude is right about 2279m/s. For simplicity, let us assume a uniform gravitational field at 8.5m/s^2 (really it goes from 9.81m/s^2 at 0km to 7.6m/s^2 at 80km). So let's say that 2279m/s of our dV goes to our orbital velocity and the rest is getting there. Suppose we did two impulsive forces in a vacuum. We would need 1166m/s to blast up to 80km and then 2279m/s to circularize, minus the 175m/s we get from Kerbin's rotation. Thus the orbital insertion can be done for 3270m/s in this two-step process, which is the least efficient 'nontrivial' (where you actively work against yourself) profile.

Of course, we know that the most efficient strategy in this vacuum scenario is a Hohmann transfer. A 0x80km orbit at Kerbin requires 2500m/s at the periapsis and 2206m/s at apoapsis. To circularize would then require an additional 73m/s at apoapsis. So we need 2500+73-175=2398m/s in a vacuum with infinite TWR and no atmosphere.

Unfortunately, the atmosphere greatly complicates both of these matters by inducing drag losses (and thereby incurring additional gravity losses too), meaning that even the Hohmann is not an ideal profile. We could not get away with these impulsive transfers in the presence of an atmosphere because the drag losses would be incredible. But we do know that the gravity, drag, and path inefficiency losses together do not add up to more than 2100m/s in stock and 900m/s in FAR, so we can certainly bound any of those three sources to be below 47% in stock and 20% in FAR (for an 'efficient' rocket).

Now, the curious mind might just ask: how fast one would need to be going horizontally near sea level to coast out to 80km through the atmosphere? It must be hundreds of km/s, at least...

[magnemoe]

Well, if the atmospheric drag model becomes vaguely realistic, the following will happen:

• Four levels of asparagus staging will become deadly as the short-but-stubby Kerbal designs are no longer effective (too much drag)
  
• Enjoy how your rocket explodes in a spectacular fashion as it hits the sound barrier
  
• If you survive that, enjoy how the atmosphere rips your rocket to pieces as a rapid pitch to 45° at 10,000m suddenly is not that good of an idea when the velocity vector is still pointing straight up.

I'm sure a realistic drag model will revolutionize launch requirements. But likely in a different way than projected by the OP

The pancake launchers and cup launchers would be out yes, but the standard 1+6 design would work well as long as you keep the boosters close to the core. yes this will make strutting and separation harder. You could also add an skirt of solid stage boosters or high twr short burn time liquids to get up to 80-100 m/s fast.

Sound barrier does not appears to be an major problems for rockets. The space shuttle is probably of the most kerbal launch system used regarding aerodynamic but aerodynamic was not any issue here. Yes I'm sure its lots of tweaks on it to reduce drag bit it shows than other designs than arrows works.

The gravity turn would be an major change, we have to do an much more rounded turn, 60-70 instead of 44 in mechjeb accent autopilot adjust trajectory.

In short, you must always have the pointy end at the direction of travel or drag would slow you down fast or even rip you apart.

[Temstar]

The pancake launchers and cup launchers would be out yes.

Oh?

http://en.wikipedia.org/wiki/SERV

So clearly either Chrysler's Space Division engineers are all duds, or the drag model in Real Life is not realistic enough?

[AceofPengwins]

I want to build a SERV now.

[magnemoe]

Oh?

http://en.wikipedia.org/wiki/SERV

So clearly either Chrysler's Space Division engineers are all duds, or the drag model in Real Life is not realistic enough?

Still look more aerodynamic design than the shuttle at launch position.

Think this is more that we call non aerodynamic designs.

An Eve lander.

Note that struts itself is an major drag source.

[WafflesToo]

Oh?

http://en.wikipedia.org/wiki/SERV

So clearly either Chrysler's Space Division engineers are all duds, or the drag model in Real Life is not realistic enough?

I think there's a reason they didn't win the contact XD

[Temstar]

SERV losing the STS contract was more to do with the ambitious aerospike engine design and the fact that NASA pretty much already made up its mind before even starting the competition that they wanted a large spaceplane. That said it definitely wasn't what you would call an aerodynamic design. The idea was that yes, you lose delta-V due to drag at launch from the tubby shape. But then that tubby shape has already been extensively studied and tested during Apollo for the command module and is proven to make re-entry easier per Blunt Body Theory. Hence the thermo protection system on SERV will be simple and lightweight and that saving in weight of the TPS more than offsets the drag loss at launch.

TLDR: under a realistic drag model drag will affect stubby rockets more than streamlined ones. But on the other hand under a realistic drag model drag isn't going to be a big deal like it is now so both streamlined rocket or flying pancakes will reach orbit easier. And since asparagus staging has a lot of advantages elsewhere it won't be going away.