So even flat surface is better than Rocket Nose Mk7?

[SaturnV]

HarvestR said the new drag system will make streamlined designs fly easier than non streamlined ones, then I tested it.

How the tests were performed:

1. Load and launch the rocket

2. Turn on SAS, full throttle, press space

3. Switch to map view, click on the AP label, observe the maximum altitude it reached before ran out of fuel. (The number will decrease after that, due to the air drag)

Ships used in the tests:

Very Flat / Flat / Pointy / Very Pointy

* By adjusting mono fuel and adding/removing parts in the service bay, all ships are now exact 20,000 kg, all of them have 1,553 m/s delta V.

* They're all the same except the amount of mono fuel, the content of service bay and their nose.

Result:

Very flat: 10,000 m

Flat: 55,000 m

Pointy: 45,000 m

Very pointy: 58,000 m

You can test it yourself (craft file):

http://1drv.ms/1KxxOd3

Edited May 10, 2015 by SaturnV

[Kuu Lightwing]

Does it go supersonic? If not - add more boosters and repeat the test.

[m4v]

If you go close and past the sonic barrier nosecones do mater. Still their usefulness are kind of dubious, fairings specially.

[Murph]

Flying better isn't only about maximum range. Your test is invalid due to staying vertical, and focussing only on the maximum altitude. A valid test must include a realistic ascent profile, and consider whether it improves aerodynamic stability and control. I.e. does it improve flight characteristics, helping to avoid loss of control, on a realistic profile? A valid test should also keep to appropriate speeds on the way up, i.e. no prolonged balls of fire due to excessive TWR, no mach 3+ at low altitude.

If you want to conduct a more meaningful test, I suggest using MechJeb to get a repeatable "hands off" realistic ascent profile (assuming that MJ ascent guidance is working for 1.x now, I've yet to try it as I have chosen to do the first bit of career hand flying). You'd still need to do some separate hand flying, and subjective evaluation, to decide whether it helps for control and stability.

N.B. I'm not saying that the nose cone is ok, only that your test is not ok.

[DoToH]

Your test is invalid due to staying vertical, and focussing only on the maximum altitude. A valid test must include a realistic ascent profile, and consider whether it improves aerodynamic stability and control. .

[Jappards]

The results must be reproduceable, otherwise it isn't valid either, so I will be carrying out tests too. At first, I will try to reproduce SaturnV`s results using his/her craft file. Some other things people want me to test?

Also: very flat gives a higher altitude than flat?

Edited May 10, 2015 by Jappards

[ferram4]

So long as the masses are identical so that the effect of gravity is the same in all tests, this method of testing is valid. Further, the high TWR of the rockets ensures that the aerodynamic forces are very high, which should reduce the error in the calculations.

I see nothing invalid with this method of straight-up testing, since it provides a very reproducible method of launching, as well as a simple method to determine the energy lost to aerodynamic forces, which will be proportional to the reduction in altitude.

In all cases, this is concerning, since the flatter rockets should not only have higher drag, but with all that drag ahead of the CoM, should be harder to control. If that isn't the case, there is a serious issue that needs to be fixed.

[Jouni]

It may be an issue with the service bays. I built a test rocket without it, and the results were different.

Both rockets weighted 23.3 tonnes. The one with a nosecone climbed to 20080 m, while the one with an extra battery only reached 14017 m.

[Jappards]

I did my first test(#1), here are the results:

Very Flat: 6,380m

Flat: 35,610m

Pointy: 25,300m

Very Pointy: 39,440m

(results where rounded to the nearest ten)

I wanted to eliminate even more variables: This test was done with the same pilot every time: Jebediah Kerman. I did use Display aero forces in flight in the debug menu, but that did not interfere with the flight. The SAS stability assist was used to stay straight in the air. I had my hands off the flight keys and only looked at the map for a little bit. Crash results where used.

Looking at the map for max altitude isn't reliable, because the number changes very fast.

Edited May 10, 2015 by Jappards

[Brotoro]

Nice set of tests there, SaturnV. Have a kerbal snack.

[sal_vager]

The tests could be valueless however, if Jouni's findings with the service bay are accurate.

[m4v]

In all cases, this is concerning, since the flatter rockets should not only have higher drag, but with all that drag ahead of the CoM, should be harder to control. If that isn't the case, there is a serious issue that needs to be fixed.

Those rockets have a lot of torque installed though.

[Jouni]

Three more tests with my 23.3-tonne rocket.

With an empty service bay on top of the rocket (and 300 kg less monopropellant), the rocket reached only 13260 m. Apparently the service bay has a slightly larger cross-sectional area than most 2.5 m parts.

Then I removed the extra 2.5 m battery and replaced it with a pile of 1.25 m batteries on the floor of the service bay. This reduced the drag, and the rocket managed to climb to 15908 m. On the other hand, when the batteries were hanging from the roof, the maximum altitude was only 13379 m.

Finally I moved the remaining 2.5 m battery on top of the service bay with 1.25 m batteries on its floor. This fixed the issue, and the rocket climbed only to 13759 m.

- - - Updated - - -

Still more experiments.

I moved half of the 1.25 m batteries from the service bay on top of it. When the remaining batteries were on the floor, the rocket reached 16015 m – a bit more than when all batteries were on the floor. With the remaining batteries hanging from the roof, the altitude was only 13447 m – also a bit more than with all batteries inside.

With all batteries on top of the service bay, the maximum altitude was 13289 m.

My conclusions: Parts on the floor of a service bay reduce drag, unless the service bay is fully occluded.

Edited May 10, 2015 by Jouni

[SaturnV]

Your test really helps!

My test was done assuming that things in the service bay won‘t affect anything related to the aero performance, it's not true however.

The result looks really strange, I‘ll check the service bay later.

[pincushionman]

Since it appears the service bays are impacting the test negatively, I would repeat the nosecones test, but with a locked-off fuel tank to vary the mass rather than a variable number of parts to worry about.

[Cdr_Zeta]

Hello,

And the reason for this? Just confirms atmospheres work? I dont get it...

There are worse problems with atmosphere than this test.

Zeta

[Lord Aurelius]

Hello,

And the reason for this? Just confirms atmospheres work? I dont get it...

There are worse problems with atmosphere than this test.

Zeta

The purpose of this test is to verify that nosecones do what Squad said they would do: reduce the drag of a rocket stack. This was a much-hyped feature of the new aero (since nosecones were worse than deadweight before) and someone just wanted to verify that the new aero was working as advertised. It didn't, which is why we have this thread which has some excellent experimental results that the devs can hopefully use to narrow down the root cause (too high of a drag value on the nosecone?) and fix the problem.

I agree that the atmosphere has other issues as well, but this is actually a pretty significant issue if the new drag system isn't working correctly.

[klgraham1013]

The purpose of this test is to verify that nosecones do what Squad said they would do: reduce the drag of a rocket stack. This was a much-hyped feature of the new aero (since nosecones were worse than deadweight before) and someone just wanted to verify that the new aero was working as advertised. It didn't, which is why we have this thread which has some excellent experimental results that the devs can hopefully use to narrow down the root cause (too high of a drag value on the nosecone?) and fix the problem.

I agree that the atmosphere has other issues as well, but this is actually a pretty significant issue if the new drag system isn't working correctly.

I was going to use more snark, but this is probably the better answer.