Drag, or: how do I find out which design have more drag?

[Jestersage]

The general rule of thumb is "Mk2 fuselage = draggy"

However, I was looking at the drag coefficent in the wiki, and notice that Mk2 parts is not that high -- in particular, I was trying to use the 1.8m cockpit and a 1.8-2.5m adapter (and a nosecone at front), and when I compared the figures of equivalent Mk2 parts (mk2 cockpit+mk2-2.5m adapter). I noticed the drag coefficent is about the same, if not lower, with Mk2 parts.

So which is right? Or, if you do not have an answer, how should I test out the two different design to find out which has the lower drag?

Edited March 24, 2019 by Jestersage

[KevinW42]

You can view the drag produced by individual parts in flight by opening the alt+f12 menu and going to the physics -> aero section. There should be an option called something like "show aero data in action menus". If you turn that on then right clicking a part while you're flying will show you the exact amount of drag it's producing.

And someone else may have a better answer for this part, but if I understand correctly the problem with Mk2 parts is that even though they seem to have good drag values on paper, the actual drag shoots through the roof as soon as they become angled even slightly relative to your direction of travel.

Edited March 24, 2019 by KevinW42
spelling

[bewing]

Are you intending to fly it for hundreds of km through the atmosphere? If not, then it doesn't matter all that much.

If you are building an airplane, then try the two designs with the same engine -- and see which one breaks mach 1 the easiest. An alternate technique is to build your craft with two stacks attached by a wing to each other. Then see which way it tries to yaw.

 

[Jestersage]

49 minutes ago, bewing said:

An alternate technique is to build your craft with two stacks attached by a wing to each other. Then see which way it tries to yaw.

I think I will try that.

The two plane are litterally identical except the parts I mentioned. The problem is that the one using 1.8m parts have better TWR (for some reason).

EDIT: Decide to use the build in AeroGUI (not the action menu). Keep the same flight profile (at 13k). Did 2 tests. Results are below:

Design1: using Mk2: 330kN, +/- 20kN
Design2: usign 1.8: 300kN, +/- 20kN

What is weird when they are combined as one, the plane yaw to design2's side.... So why is it the case, if I test it indidvidually the 1.8m design have less drag?

Edited March 24, 2019 by Jestersage

[Aeroboi]

6 hours ago, Jestersage said:

The general rule of thumb is "Mk2 fuselage = draggy"

However, I was looking at the drag coefficent in the wiki, and notice that Mk2 parts is not that high -- in particular, I was trying to use the 1.8m cockpit and a 1.8-2.5m adapter (and a nosecone at front), and when I compared the figures of equivalent Mk2 parts (mk2 cockpit+mk2-2.5m adapter). I noticed the drag coefficent is about the same, if not lower, with Mk2 parts.

So which is right? Or, if you do not have an answer, how should I test out the two different design to find out which has the lower drag?

The point is that Mk2 is draggier then Mk1 so if you want to reduce drag as much as possible Mk2 is a discarded option. The difference is indeed negligible, yet it isn't.

When you talk about frontal cross section the difference isn't that great. The mk2 is only slightly more draggy then Mk1. When your Mk2 fuselage is at a angle due to the pitch of the aircraft it gets a lot worse and the drag of Mk2 becomes noticeable especially if the whole plane consists of it.
Solution: Use wing incidence so the nose is pointed at prograde for the duration of your flight.

Another issue with Mk2 is that you often only use a front Mk2 cockpit with mk1 adapter nodes at the back making the nose draggy which causes instability in flight or during aerobraking.

Another reason why you would want to use Mk2 anyway is due to space plane shape, ability to use mk2 cargo bays which might look weird if it were a mk1 plane only to have mk2 adapters and cargo bay but that's ultimately any design wish.
The actual reason is because mk2 are space plane fuselage modules their drag is very good at aerobraking. That means you would require very little wings, like a very efficient SSTO only required to do a takeoff and a supersonic speedrun (you wont need many wings for this) and many high dv ssto's are minimalistic on wing design.
That means the mk2 fuselage + a few wing strakes and some smallest elevons for control has enough stopping power to do a interplanetary aerocapture, otherwise using mk1 you'd need more wings which is a higher part count and ultimately not that lighter because for the Mk1 design you'd need more wings for takeoff and aerobraking.

Edited March 24, 2019 by Aeroboi

[Jestersage]

5 hours ago, Aeroboi said:

<snip>

So you are saying that the "lift" surface, and not just the drag coefficient, is what caused the drag?

[Aeroboi]

2 hours ago, Jestersage said:

So you are saying that the "lift" surface, and not just the drag coefficient, is what caused the drag?

No, and I could have been clearer tbh. What I mean is that the cross section increases at any degree above or below prograde. That effect is what makes mk2 planes noticeably draggier and is part of the problem people face. This problem usually occurs with mk2 planes when using very little wings and more so if not using wing incidence with it. Sometimes people build very cool pointy looking mk2 planes with sparse amount of wings and then wonder why they don't accelerate past 400m/s while the nose is at 10°. While the example in itself is excessive the drag associated rises exponentially the further the nose is from prograde when trying to accelerate past 400m/s  

The less wings like on a wing minimalistic design would want to have more degrees of incidence and follow a according flight path to stay near prograde. If you want maximum Dv in orbit you would want to design in that particular way.

Edited March 24, 2019 by Aeroboi

[Laie]

16 hours ago, Jestersage said:

So which is right? Or, if you do not have an answer, how should I test out the two different design to find out which has the lower drag?

Lately, I've started to do drop tests on Eve: roll out vessel, teleport it to 20km altitude at Eve, drop. Much easier with hyperedit, of course.

Simple pointy rockets will impact the ground at 700m/s+, but more complicated designs will reach their top speed at some higher altitude, then the thick atmo will start to slow them down. That's a great way of telling whether this or that modification will have a major or minor impact on aerodynamics.

It's advisable to unload most fuel: the same amount of drag will have more effect if the vessel weighs less, so try to remove as much ballast as possible while keeping the CoM where it needs to be. It's not strictly necessary that vessel mass is 100% identical between tests.

Dropping stuff on Jool may work better, I don't know. I only ever did it on Eve and only thought of Jool just now. ETA: my version of Hperedit won't let me use the "lander" function on Jool, so it seems that Eve is the easier destination.

 

Edited March 24, 2019 by Laie

[Jestersage]

2 hours ago, Laie said:

Lately, I've started to do drop tests on Eve: roll out vessel, teleport it to 20km altitude at Eve, drop. Much easier with hyperedit, of course. 

Okay, I do have HyperEdit, but most of the time anything that is not in Orbit (but especially when "landing") will end up clipping in the ground spectacularly... so what are the parameter?

Edited March 24, 2019 by Jestersage

[Foxster]

10 hours ago, Jestersage said:

Okay, I do have HyperEdit, but most of the time anything that is not in Orbit (but especially when "landing") will end up clipping in the ground spectacularly... so what are the parameter?

Use the alt-f12 set-orbit cheat function first. Only use HE for the landing bit. 

HE gets confused when switching SOI. 

[Laie]

18 hours ago, Jestersage said:

so what are the parameter?

I don't have that problem and can't provide a workaround. I litearlly roll out the vessel, and, on the launch pad, enter any old coordinates on Eve (0/0 is fine) and click "land". Once there, I click "drop". That's all.