Terminal Velocity Tests using Graphotron 2000 in KSP 1.0.2

[LitaAlto]

Click here for the telemetry spreadsheet.

I've been experimenting with Graphotron 2000 since it was released for KSP 1.0, and as I'm playing version 1.0.2, I figured I'd see if I could pull some interesting data from the game and see what it has to say about terminal velocity.

(Now, let's be up front, each craft will have its own unique terminal velocity based on how much drag it creates. I'm not about to claim these numbers have any direct applicability to anyone's gameplay. However, it may prove insightful.)

I built a 1-ton payload (really 1.04 tons but I can't quite shave off the last 40 kilograms....) and equipped it with an accelerometer; a barometer; two thermometers, one inside and one outside the cargo bay; and the Graphotron 2000.

After sending it up on a BACC booster, I let it drop, recording data starting at the edge of the atmosphere and continuing until about 500 m ASL. At that point I was about 150 m from lithobraking, so I paused to save the data recorded.

Already there have been some attempts in the Google+ KSP Community to puzzle out the implications in the data. As they say--the more the merrier. Feel free to grab the data and use it as you will. I only ask that you credit me for generating the data.

[bakanando]

Very nice data, thanks for the share. I'll see if I can help with it.

[Pax Kerbana]

Nice vid! Nice data! Also this Graphotron thing looks sweet!

[Zephram Kerman]

I attempted a similar test in 1.0, before the patch came out. Dropped a little rocket from 150km straight down. The poor thing never did reach terminal velocity, except maybe for the last few seconds. So I did a similar experiment with a little rocket, the Sniffer. I wasn't aware of Graphotron, but a friend suggested VOID, so I used that instead. Sniffer1 took measurements in 0.90, Sniffer2 repeated them in 1.0, and this morning Sniffer3 flew in 1.02.

Here's a graph of the results:

After reading lots of complaints about "Squad reverted back to the soupasphere in 1.02", I did not expect the actual density to look like this. It's a bit hard to see, but 1.0 and 1.02 have practically the same density profile all the way up. (There are subtle variations, probably due to time of day.) The bigger differences is between 0.90 and 1.0, where the density actually increased except near sea level. So I think this means the perception of changed density is due to other related changes of the drag model, etc.

(Your data would have been used for this, but VOID measures density (g/m³) while Graphotron measures pressure (Pa). Instead of fumbling with conversions, I just made the third test consistent with the prior two. I can supply raw data and/or selected data points if anyone wants to see.)

Very nice vid, by the way!

Edited May 9, 2015 by Zephram Kerman

[NathanKell]

Correct, nothing about the atmsophere changed 1.0 to 1.0.1, just the drag multipliers. Part of the increased density in .90 was that density was linear with pressure rather than computed from pressure and temperature (and temperature during the day on the equator in 1.0+ is...rather warmer than baseline, that's why your curves don't meet at sea level). If you go north until sea level temperature is 288K, then sea level density should be 1.225 again.

As to the main topic, as cool as the research is, simply measuring current velocity isn't really going to get you terminal velocity. You need to calculate current drag and current weight and then calculate what your instantaneous TV is for those.

[Zephram Kerman]

If you go north until sea level temperature is 288K, then sea level density should be 1.225 again.

Make sense. Very interesting. Maybe I'll deliver Sniffer4 to a northerly launch site and run the experiment again from there. But unfortunately RL will delay that for a while.

As to the main topic, as cool as the research is, simply measuring current velocity isn't really going to get you terminal velocity. You need to calculate current drag and current weight and then calculate what your instantaneous TV is for those.

I'd be very interested to see someone tackle this, too. ATM, I have no idea how fast I should be flying. I like to adjust the turn so that I can keep full throttle until MECO.

[NathanKell]

AeroGUI will calculate your instantaneous terminal velocity.

[LitaAlto]

Correct, nothing about the atmsophere changed 1.0 to 1.0.1, just the drag multipliers. Part of the increased density in .90 was that density was linear with pressure rather than computed from pressure and temperature (and temperature during the day on the equator in 1.0+ is...rather warmer than baseline, that's why your curves don't meet at sea level). If you go north until sea level temperature is 288K, then sea level density should be 1.225 again.

As to the main topic, as cool as the research is, simply measuring current velocity isn't really going to get you terminal velocity. You need to calculate current drag and current weight and then calculate what your instantaneous TV is for those.

No, that's understood. Thank you for being more explicit about it. The data includes mass and dynamic pressure so the drag can be calculated.

And I stress again, I am not claiming any of this data applies directly to anyone else's craft. Again, this is at best a baseline.

- - - Updated - - -

...I should see if any of the G+ folk care to share their number-crunching here, actually. We've got some nice graphs coming out of this data.

Edited May 9, 2015 by LitaAlto

[NathanKell]

Ah, gotcha. Keep on doing awesome data work then. :]

[Gargamel]

I built a 1-ton payload (really 1.04 tons but I can't quite shave off the last 40 kilograms....)

forum.kerbalspaceprogram.com/threads/76231-1-0-x-Kerbal-NRAP-Procedural-test-weights!-v1-5-0-4-30-04-15

Try this, NRAP.

[Zephram Kerman]

Correct, nothing about the atmsophere changed 1.0 to 1.0.1, just the drag multipliers. Part of the increased density in .90 was that density was linear with pressure rather than computed from pressure and temperature (and temperature during the day on the equator in 1.0+ is...rather warmer than baseline, that's why your curves don't meet at sea level). If you go north until sea level temperature is 288K, then sea level density should be 1.225 again.

Ok, did it! Spent a peaceful afternoon cruising around Kerbin in the spotter plane, and found a coastal location that reaches 288K twice each day. At that moment, the density reading was 1220 g/m³. Almost exactly like you said; I think the difference was due to the 5.21m height of the equipment. Built a mobile launcher for Sniffer4, got it there, and redid the experiment. The yellow line is the new data, launched from 288K beach.

AeroGUI will calculate your instantaneous terminal velocity.

Thanks for the tip! I'm enjoying AeroGUI right out of the box.

[LitaAlto]

forum.kerbalspaceprogram.com/threads/76231-1-0-x-Kerbal-NRAP-Procedural-test-weights!-v1-5-0-4-30-04-15

Try this, NRAP.

Thanks for the suggestion. I've seen it before, as have a few others I've discussed my data with. There were even some attempts at making more aerodynamic test weights, but given the new KSP drag model, apparently shape isn't really a factor, so there's only so much optimization we can do, outside of nuFAR's voxel model.

[NikkyD]

I just wanna add to this thread.

Terminal Velocity means the gravitational force equals the drag (force). The heavier your craft the higher its TV. Some crafts will be so heavy that their TV is above mach 1 at which you have increased drag.