Intake Differences?

[GoSlash27]

 Jarin,

 I'd love to see some empirical data on this myself, as I haven't done any testing myself.

Best,
-Slashy

[Sharpy]

On 27.12.2016 at 0:15 AM, String Witch said:

How did you obtain this graph?

I have a bunch of mod intakes I'd like to rate like that.

[Guest]

3 hours ago, Sharpy said:

How did you obtain this graph?

I have a bunch of mod intakes I'd like to rate like that.

Found on a YouTube comment but I've seen it before, possibly on the wiki.

Edited December 28, 2016 by Guest

[GoSlash27]

Sharpy,

 That graph is @Right's handiwork.

Best,
-Slashy

[tseitsei89]

There is some nice drag testing in that playlist. I recommend you watch all of these videos. It also answers to the question "Should I close my intakes?" among many other drag related things

[tseitsei89]

Did some of my own testing as well:

Some results: (not all tests are shown in the videos. At least not yet)

1. Fairing is the least draggy part in high speeds. Circular in take at low speeds.

2. Shock cone intake is not that good at low speeds and altitudes.

3. But good at high speeds. (That is why it was able to almost match ramp intake at airbreathing test but lost clearly on rocket test. We were traveling faster in the airbreathing test than in the rocket test.)

4. 1 circular intake can't feed 1 rapier enough at high altitudes. (I added engine nacelle to otherwise same craft with circ intakes and got a bettter performance.)

5. 1 shock cone intake can feed (at least) 3 rapiers. 1 ramp intake can't but it can feed 1 rapier at least.

 

So use ramp or shock cone intakes still not really sure wich  And use fairings as nosecones!

That's pretty much it at this point but I'll continue testing...

EDIT: These results are wrong! Sorry for that. Please see my other post little further down this page

Edited December 29, 2016 by tseitsei89

[mystik]

Thanks for the videos. So, shock cone is the thing that is needed for SSTO, and one is enough to feed 3 rapiers. This gives me an idea involving the tri coupler and fairings to hide the drag the coupler has.

 

I will experiment a bit with this design. I also saw that launching with the activated claw actually is better. That will also assist in my attempt since I intend to use that claw on my interplanetary miner cargo SSTO.

 

If this thing ever is completed I will make sure to upload it to the forum and credit everyone involved in the research and anybody that posted links. I will surely not be able to create it being able to land on Eve or Tylo but I will create some add-on disposable lander for it that will assist with that. Expect long late night testing times before anything usable appears.

[GoSlash27]

Tseitsei,

That isn't a good test for SSTO intakes. You'd want to fly it to 20km horizontally and note whether the engine cuts out before top speed is achieved (or the TV blows up). Any intake combo that can do this is suitable for SSTO work.

Best,
-Slashy

[tseitsei89]

43 minutes ago, GoSlash27 said:

Tseitsei,

That isn't a good test for SSTO intakes. You'd want to fly it to 20km horizontally and note whether the engine cuts out before top speed is achieved (or the TV blows up). Any intake combo that can do this is suitable for SSTO work.

Best,
-Slashy

Ok I'll try that sometime then

[Leafbaron]

I really appreciate all the work this community has done to diagnosing and dissecting drag. I never really thought too much about it until I started making SSTO's. I am sure glad smarter people than myself are hashing these problems out and sharing their data. 

[GoSlash27]

5 minutes ago, Leafbaron said:

I really appreciate all the work this community has done to diagnosing and dissecting drag. I never really thought too much about it until I started making SSTO's. I am sure glad smarter people than myself are hashing these problems out and sharing their data. 

Oh, definitely not "smarter". Just a bit obsessive

Best,
-Slashy

[sardia]

21 hours ago, GoSlash27 said:

Isn't that what all great scientists are about?

Edited December 28, 2016 by sardia
Mobile posting issues

[GoSlash27]

I farted around a little and confirmed some configurations that work..

2 radial structural intakes per engine
1 precooler per engine
1 nacelle per engine
1 shock cone and 1 precooler per 4 engines

Most intakes will get the job done on a 1-1 ratio, but these are lower- drag options.

 

Best,
-Slashy
 

 

[mystik]

2 hours ago, GoSlash27 said:

1 shock cone and 1 precooler per 4 engines

What type of engines used?

[GoSlash27]

Mystik,

I ran these tests with the Whiplash, which has the highest demand for air.

Best,
-Slashy

Edited December 28, 2016 by GoSlash27

[tseitsei89]

6 hours ago, GoSlash27 said:

I farted around a little and confirmed some configurations that work..

2 radial structural intakes per engine
1 precooler per engine
1 nacelle per engine
1 shock cone and 1 precooler per 4 engines

Most intakes will get the job done on a 1-1 ratio, but these are lower- drag options.

 

Best,
-Slashy
 

 

So a precooler and a fairing in the front looks like a good way to go for 1 engine craft then

Now I'm tempted to try 1 shock cone + 1 precooler/engine nacelle with 4 rapiers on a quadcoupler and then "hiding the quadcouplers drag behind a fairing, That might give good thrust to drag ratio  

EDIT: How about the ramp intake? all my tests indicate that it has lowest drag of all front/nose mounted intakes, but how does it perform as an air provider?

Edited December 28, 2016 by tseitsei89

[GoSlash27]

tseitsei89,

 The ramp intake worked fine, but I only tried it with 1 engine.

Best,
-Slashy

[GoSlash27]

Looking at the 1.2.2 drag cubes, here's what I see for the intakes:

Type           Area      Cd  

XM-G50       .728    .425
Circ Intake  1.213   .450
Shock Cone 1.213  .300
Radial Long  .121   .351
Mini Circ       .303  .400
RamAir        1.218 .375
Precooler     1.213 .886
Nacelle        1.494 .812
Fus Intake   1.718  .617

By way of comparison, some other things you might put on the front of a stack:

Nose Cone 1.213   .631
Adv cone A 1.213  .348
Adv cone B 1.213  .454
TailConnA    1.213  .132
TailConnB    1.213  .187
Mk1Cockpit 1.215  .232

 

Judging by this, I would expect the shock cone to still be the lowest drag intake for the front of a stack.

Best,
-Slashy
 

Edited December 28, 2016 by GoSlash27

[tseitsei89]

11 hours ago, GoSlash27 said:

Looking at the 1.2.2 drag cubes, here's what I see for the intakes:

Type           Area      Cd  

XM-G50       .728    .425
Circ Intake  1.213   .450
Shock Cone 1.213  .300
Radial Long  .121   .351
Mini Circ       .303  .400
RamAir        1.218 .375
Precooler     1.213 .886
Nacelle        1.494 .812
Fus Intake   1.718  .617

By way of comparison, some other things you might put on the front of a stack:

Nose Cone 1.213   .631
Adv cone A 1.213  .348
Adv cone B 1.213  .454
TailConnA    1.213  .132
TailConnB    1.213  .187
Mk1Cockpit 1.215  .232

 

Judging by this, I would expect the shock cone to still be the lowest drag intake for the front of a stack.

Best,
-Slashy
 

And once again I was wrong and you were right

Just noticed that my tests were done like an idiot... Shock cone intake for example weighs 60kg more than ramp intake. When I added some xenon as ballast so they weigh the same Shock cone is clearly better.

And other parts also behave as expected.

 

[GoSlash27]

4 hours ago, tseitsei89 said:

And once again I was wrong and you were right

Just noticed that my tests were done like an idiot... Shock cone intake for example weighs 60kg more than ramp intake. When I added some xenon as ballast so they weigh the same Shock cone is clearly better.

And other parts also behave as expected.

 

tseitsei89,

 Empirical testing and "idiot" should never occur in the same thought.

What would be *really* illuminating would be building a testbed aircraft and checking the drag of these parts at 340 m/sec. I'm home sick today, so I think I'll give it a shot...

Best,
-Slashy

[GoSlash27]

Okay, I build a simple Mk.1 testbed and recorded the actual drag for some parts at Mach 1.00 and 1km altitude. They pretty much line up like I'd expect from the drag cubes.

Front stack parts:                           Drag

Tail Connector A                             0.94
NCS adapter + small circular intake  1.17
NCS adapter + small nose cone       1.25
Shock cone intake                          1.26
Mk1 Cockpit                                  1.36
Tail Connector B                             1.44
Advanced Cone A                          1.95
Circular intake                               3.13
Advanced Cone B                          3.28
Adjustable ramp                             4.22
Aerodynamic Nose Cone                8.83
Nuthin' (structural fuselage)            101
 

Inline parts:

Precooler                                      0.68
Mk1 Inline Cockpit                         1.20
Engine Nacelle                              1.43
Mk1 Diverterless intake                  2.61

Radial parts:

Structural Intake Radial                  1.28 <-- but really you need 2, so 2.56
XM-G50                                        4.46 <-- I haven't tested to see how many are needed, but it doesn't matter. It sucks.

Best,
-Slashy

 

 

Edited December 29, 2016 by GoSlash27

[mystik]

I am playing around with my SSTO and replaced all intakes with shock cones. Would not go past 375m/s. Replaced some with ramp intakes and now the thing maxes out at 1350m/s. Well, I stop there because the claw starts to melt, at which point I point it towards the sky and it whooshes out of the atmosphere.

 

This is weird. Seems that the shock cone does not have enough power to get the thing up to the critical speed on it's own. It's a heavy beast at 400t, but still, having half ramps seems to change the flight significantly.

 

I don't understand. If all intakes are left as ramps the engines cut off on the runway. It has 6x2 rapiers, 5 whips, struggles to reach orbit since the ship had over 18 rapiers in it's original config but had a weight of over 500t.

[Aegolius13]

My overall thoughts on each intake:

Small Circular Intake: naturally paired to Juno due to size.  Not that practical for bigger/faster engines due to overall air capacity, bad high-speed performance and very low heat resistance.  

Circular Intake: good part for subsonic planes.  Also usable as a general purpose nosecone due to light weight and low drag.  Not practical for high-speed flight.

XM-G50 Radial Air Intake.  I don't really use this much; it doesn't seem like it's worth the drag.  I suppose it could be useful in early game, before you have the precooler and if you don't have a front node open.  

Engine Nacelle: Provides intake, fuel and housing for an engine, but not a great performer.  I've heard it's useful for VTOLs due to strong air capacity at very low speed.  

Divertless Supersonic Intake: For SSTOs, this is eclipsed by the Precooler.  But I find it very useful for disposable jet boosters, since it works at relatively high speeds, carries about the right amount of fuel, and is VERY cheap.  

Radial Adjustable Ramp Intake: It's extremely light, pretty cheap, and works at high speeds.  But it does not take in much air, especially at low speeds.  And it adds drag.  Sometimes useful for very small aircraft, but I don't use if I have more then one engine. 

Adjustable Ramp Intake: A very nice intake compared to the previous options, but it's overshadowed by the Shock Cone, which has better high-speed performance and (I think) slightly lower drag.  I don't use this much as a result.

Shock Cone: My go-to intake for most applications.  Ample air capacity at all speeds, great heat resistance, good aerodynamics, and not too heavy.  

Precoooler: The best inline intake, hands-down.  I use this when I want something else, like a docking port, on the front of my ship.  But since you need some kind of nosecone anyway, I generally use the Shock Cone instead when the front node is available.  

 

Edited December 29, 2016 by Aegolius13
I stand corrected on small circular intake air capacity at higher speeds.

[GoSlash27]

Surprisingly, the small circular intake is actually capable of driving the Whiplash at full thrust and 20 km altitude without flaming out. But yeah, the thermal limit makes it impractical for SSTO use.

Best,
-Slashy

[swjr-swis]

45 minutes ago, GoSlash27 said:

Tail Connector A                             0.94

I've noticed a tendency of using the tail connector as a nosecone since 1.2pre, because of the 'pointier is less draggy' change made back then.

While the drag nrs might seem tempting, I feel the need to point out that this wonderfully low drag is enjoyed only when the connector is flying exactly prograde in the airflow. Due to its length, even the slightest bit of deviation from exact prograde will very rapidly turn this 'perfect' nosecone into the draggiest part on a plane. The farther ahead of CoM it is placed, the more leverage it has, and thus tends to make for a very flip-happy craft.

Unfortunately, it also shows that same drag behaviour when placed as a tail cone. Despite it seeming like it should be a very aerodynamic choice both for nose and tail, if the craft using it cannot be kept exactly prograde, it will actually be a very draggy part.

The joys of calculating drag through cubes...

 

55 minutes ago, GoSlash27 said:

Tail Connector B                             1.44

Advanced Cone B                          3.28

I wish they had kept the 'B' and 'A' identical regarding drag. In the intended use case of slanting the cones to make the overall craft more aerodynamic, the slanted ones should actually provide a drag benefit, even if when looked at individually they would be more draggy than the non-slanted ones. Now they are un-intuitively always draggier than the straight cones. By letting them have identical drag figures as the straight ones, at least it would've averaged out over different use cases.

Thanks for tabulating your test results, @GoSlash27. Well wishes (even if that means less KSP play ).