Sigma Dimensions

[Sigma88]

7 hours ago, ThorBeorn said:

What am I doing wrong here?

A couple of things:

1 - you are not changing the radius, you are adding a second "radius" parameter which doesn't get used

changing the radius is a bad way to resize the planet anyways, (otherwise making SD would have been a pretty huge waste of time).

to get kerbin's radius to be 6371000 you should use

Resize = 10.6183333333

in the SD settings

in alternative, if could look at Planet Specific Changes (see README.txt in GameData/Sigma/Dimensions)

2- for the atmosphere I am supposing you copied the whole RSS "Atmosphere" bit from Earth cfg

that's fine, but try doing it this way:

@Kopernicus:FINAL
{
	@Body[Kerbin]
	{
		!Atmosphere,* {}
		Atmosphere
		{
			// FROM RSS
		}
	}
}

this should work

[ThorBeorn]

Thanks a lot. I must have been tired yesterday, I looked at the how-to-apply-planet-specific-changes section three times

[Bahamut]

I've been experimenting with a 6.4x system and have been experiencing some difficulty with re-entry. By searching the forum, I've implemented the physics modifier from RSS, and boosted ablator quantity in some heat shields.

To test, I have a rocket that launches a small probe with one pie of universal storage science and a 1.25m heat shield. Re-Entry from LKO is OK. However, re-entry from returning from a Mum/Minmus trip results in a destroyed probe within ~15 seconds of entering the atmosphere with speeds of >7500 m/s.

With stock KSP, I could re-enter from such an oblique orbit and not burn up. I'm sure that the high speed is the source of my problem, but not sure if:

1) This is to be expected for realism from a 6.4x or RSS type system, and I should use very slight atmosphere entry with retrograde burning to slow myself down more before attempting re-entry.

2) The heating physics is still too severe and that I should be able to do a single-pass re-entry when returning from the moon.

Any pointers would be great!

[Sigma88]

9 minutes ago, Bahamut said:

I've been experimenting with a 6.4x system and have been experiencing some difficulty with re-entry. By searching the forum, I've implemented the physics modifier from RSS, and boosted ablator quantity in some heat shields.

To test, I have a rocket that launches a small probe with one pie of universal storage science and a 1.25m heat shield. Re-Entry from LKO is OK. However, re-entry from returning from a Mum/Minmus trip results in a destroyed probe within ~15 seconds of entering the atmosphere with speeds of >7500 m/s.

With stock KSP, I could re-enter from such an oblique orbit and not burn up. I'm sure that the high speed is the source of my problem, but not sure if:

1) This is to be expected for realism from a 6.4x or RSS type system, and I should use very slight atmosphere entry with retrograde burning to slow myself down more before attempting re-entry.

2) The heating physics is still too severe and that I should be able to do a single-pass re-entry when returning from the moon.

Any pointers would be great!

I think it should be expected to see those extremes when returning straight from mun.

I am not really an expert on the topic, but I think Apollo missions did something like this:

https://en.wikipedia.org/wiki/Skip_reentry

have you tried less aggressive manouvres?

[Bahamut]

Actually yeah, I just did some more testing after changing the atmosphere height factor to 2. I used a mostly similar ship (though I beefed the ascent stages to account for the higher atm height). I tried to use a mild aerobrake maneuver using my vacuum stage cryo engine in atmo. The target perapsis was 100 km out of a 140 km atmosphere height.

1) Don't use vacuum stage cryo engine's in atmosphere... that part of the rocket blew up in about 10 seconds
2) Though after that disaster, the probe/science/heatshield part of the rocket survived the first re-entry. Though by this point my solar panels blew up and my battery definitely would not last till the next pass (RT installed). 
3) With no control, the rocket still made it halfway through the 2nd pass before blowing up.

I think from this, I definitely need to be VERY careful with designing my rockets. I'm going to need to design a re-entry stage that has some kind mono-prop or radial-based propulsion with a heatshield to both slow myself down without blowing up everything I need to keep control and flying for the true landing. Then come out from the aerobrake close enough to LKO that I can then use a traditional landing.

What is going to result looks very similar to a skip re-entry though to make it a single pass, it's going to have to be powered and I'm going to have to test to see if I can reliably build something to survive that.

Wow... 6.4x is challenging, in a way that I haven't felt since first learning to fly in stock KSP.

[Bahamut]

3 hours ago, Sigma88 said:

https://en.wikipedia.org/wiki/Skip_reentry

have you tried less aggressive manouvres?

OK, yeah the combination of a one pass for aerobraking and using a powered rocket was able to get my test probe back to Kerbin safe and sound. Started with 375 ablator before the first pass, had 18 left by landing... that was too close, lol. Picture album shows the final configuration that was able to make the landing. Also had to cheat a little by using forcing local control since I didn't have a satellite relay for RT (lost control during landing). Towards the end, the probe would tip over, so on my successful run was able to use spin stabilization to keep the probe facing the right way until my drogue and then main parachutes could deploy.

Overengineered, but totally worth it for the first successful landing.

Edited May 6, 2016 by Bahamut
Adding additional info.

[Sigma88]

6 hours ago, Bahamut said:

OK, yeah the combination of a one pass for aerobraking and using a powered rocket was able to get my test probe back to Kerbin safe and sound. Started with 375 ablator before the first pass, had 18 left by landing... that was too close, lol. Picture album shows the final configuration that was able to make the landing. Also had to cheat a little by using forcing local control since I didn't have a satellite relay for RT (lost control during landing). Towards the end, the probe would tip over, so on my successful run was able to use spin stabilization to keep the probe facing the right way until my drogue and then main parachutes could deploy.

Overengineered, but totally worth it for the first successful landing.

If I understand this correctly you are using RSS physics modifications right?

Could I see the cfg you use?

Have you tried using smurff to reduce the mass of parts? If I understand it correctly stock parts are heavier that what they should be and thus slow down less while reentry

[Bahamut]

Spoiler

// Borrowed from Realism Overhaul.
@PHYSICSGLOBALS:FIRST // done on first so anything else will override this.
{
    // AeroFX
    @aeroFXDensityScalar1 = 0.00004
    @aeroFXDensityExponent1 = 0.7
    @aeroFXDensityScalar2 = 0.0036
    @aeroFXDensityExponent2 = 2
    @aeroFXStartThermalFX = 2.5
    @aeroFXFullThermalFX = 3.5
    @aeroFXVelocityExponent = 4.0
    
    // CONVECTION
    // Convection areas
    @fullConvectionAreaMin = 0.1
    @fullToCrossSectionLerpStart = 0.8
    @fullToCrossSectionLerpEnd = 1.5
    
    // Newtonian Convection
    @newtonianTemperatureFactor = 1
    @newtonianConvectionFactorBase = 4.0 // approx real number here.
    @newtonianConvectionFactorTotal = 1.5873 // was 1.0, but 1.0.5 has a .63x factor internally.
    @newtonianDensityExponent = 0.5
    @newtonianVelocityExponent = 1.0
    
    // Newtonian-To-Hypersonic transition
    @newtonianMachTempLerpStartMach = 2.0
    @newtonianMachTempLerpEndMach = 4.0
    @newtonianMachTempLerpExponent = 3
    
    // Hypersonic Convection
    @machConvectionDensityExponent = 0.5
    @machConvectionVelocityExponent = 3.0
    @machConvectionFactor = 2.0
    @machTemperatureScalar = 7.5 // ~6000 at 7.3km/sec
    @machTemperatureVelocityExponent = 0.75
    
    // Turbulent convection (total guesses)
    @turbulentConvectionStart = 100
    @turbulentConvectionEnd = 400
    @turbulentConvectionMult = 25
}

The file is copied above. Yeah, I'm using SMUFF at 0.5 lever already. 

[Sigma88]

2 hours ago, Bahamut said:

  Hide contents

// Borrowed from Realism Overhaul.
@PHYSICSGLOBALS:FIRST // done on first so anything else will override this.
{
    // AeroFX
    @aeroFXDensityScalar1 = 0.00004
    @aeroFXDensityExponent1 = 0.7
    @aeroFXDensityScalar2 = 0.0036
    @aeroFXDensityExponent2 = 2
    @aeroFXStartThermalFX = 2.5
    @aeroFXFullThermalFX = 3.5
    @aeroFXVelocityExponent = 4.0
    
    // CONVECTION
    // Convection areas
    @fullConvectionAreaMin = 0.1
    @fullToCrossSectionLerpStart = 0.8
    @fullToCrossSectionLerpEnd = 1.5
    
    // Newtonian Convection
    @newtonianTemperatureFactor = 1
    @newtonianConvectionFactorBase = 4.0 // approx real number here.
    @newtonianConvectionFactorTotal = 1.5873 // was 1.0, but 1.0.5 has a .63x factor internally.
    @newtonianDensityExponent = 0.5
    @newtonianVelocityExponent = 1.0
    
    // Newtonian-To-Hypersonic transition
    @newtonianMachTempLerpStartMach = 2.0
    @newtonianMachTempLerpEndMach = 4.0
    @newtonianMachTempLerpExponent = 3
    
    // Hypersonic Convection
    @machConvectionDensityExponent = 0.5
    @machConvectionVelocityExponent = 3.0
    @machConvectionFactor = 2.0
    @machTemperatureScalar = 7.5 // ~6000 at 7.3km/sec
    @machTemperatureVelocityExponent = 0.75
    
    // Turbulent convection (total guesses)
    @turbulentConvectionStart = 100
    @turbulentConvectionEnd = 400
    @turbulentConvectionMult = 25
}

The file is copied above. Yeah, I'm using SMUFF at 0.5 lever already. 

You do that on :FIRST

this means that SD will add changes to those values

Try doing it on :FINAL

[Walker]

I'm currently testing sigma dimensions (6.4 resize and rescale with 1.285 atmosphere). Reentry seems to be the only problem. I've Deadly Reentry installed. While returning from Mun, entering atmosphere with periapsis greater than 38 km, I run out of ablator (2.5 heat shield, 800 ablator units). If I go with lower periapsis (32-35 km), there will be some ablator left, but there are very high g-forces.

[Sigma88]

16 minutes ago, Walker said:

I'm currently testing sigma dimensions (6.4 resize and rescale with 1.285 atmosphere). Reentry seems to be the only problem. I've Deadly Reentry installed. While returning from Mun, entering atmosphere with periapsis greater than 38 km, I run out of ablator (2.5 heat shield, 800 ablator units). If I go with lower periapsis (32-35 km), there will be some ablator left, but there are very high g-forces.

to be honest I am not sure how much this behaviour should be considered "wrong" and how much it should be expected.

I decided not to change parts precisely because someone might want to use, let's say, realistic parts on a 10x kerbin

and if I change the "dissipation power" of the resource ablator, it might result in having the realistic parts messed up.

plus, this is not a part rescale mod, is an universe rescale mod (with few exceptions)

so, let's start from the easy issue: g-forces

SD has nothing to do with g-forces, you have a craft in orbit with x speed (relative to the surface) and want to end up landed (0 m/s speed relative to the surface)

and you want to do that in n seconds

so, you should expect a minimun g force of  x/n m/s^2

if that's too high, you either need to lower x (lowering you Ap once close to kerbin) or increase n (changing your descent profile)

 

the ablator issue:

you still burn off too much ablator, which is probably to be expected since the 2.5m shield is designed to work on kerbin and not on "earth"

you might consider buffing the ablator capacity of the shields or the energy dissipated by 1 unit of the ablator resource

I could consider putting a parameter for this, but it would be turned off by default because of the reasons I explained earlier

 

the physical parameters:

after considering the previous points I have to admit that the physics fix of SD can probably be improved, I plan to do this once I have updated all my mods to work with KSP 1.1.2 and kopernicus 1.0.4

If any of you can help me with the math behind those parameters, feel free to contact me via PM or post in this thread

keep in mind tho, that I am not interested in the numbers themselves but rather in the math I need to do in order to make reentries work with any resize value

[Bahamut]

12 hours ago, Sigma88 said:

You do that on :FIRST

this means that SD will add changes to those values

Try doing it on :FINAL

OK, I'll try that tonight and see if results are different. 

EDIT: Had to rebuild my mods list again, but changing the physicsconfig file :FINAL seemed to have helped a lot. Still burning a lot of ablator, but I was able to do a direct entry from 7 km/s without feeling like I was going to burn up. This was with me changing the atmosphere height multiplier back to 1.285, which is closer to expected atm height. 

Edited May 7, 2016 by Bahamut

[Walker]

@Sigma88

Thank you for detailed information. I agree that this problem is caused by using parts that are not balanced for resized system; I think that physics themselves work correctly. The same problem applies to other parts, like engines and fuel tanks, but is easier to solve. For example I use Real Fuels with sotck alike configs. With "useRealisticMass = true" parameter, I get similar mass to orbit ratio in 6.4 as I would in vanilla game. There is also SMURFF. I'll try to buff the ablator capacity in Deadly Reentry - although I will be doing it with with "trial and error" method.

 

 

 

[JebThrillMaster]

I've been noticing that there seems to be a sort of 'natural' decay once in orbit on 6.4 scale. Just wondering if this is intended and would like to know if the InverseRot timewarp fix ever got an exact number (using 91000 currently).

[Sigma88]

18 minutes ago, JebThrillMaster said:

I've been noticing that there seems to be a sort of 'natural' decay once in orbit on 6.4 scale. Just wondering if this is intended and would like to know if the InverseRot timewarp fix ever got an exact number (using 91000 currently).

SD sets that parameter to the atmosphere max height

You should not need to change that.

If you are seeing the issue aftrr hyperediting a ship then just save and reload

If you are not using hyperedit then maybe it's the rotation of the planet that is too fast?

I would need a bit more info to know for sure

[Sigma88]

Sigma Dimensions v0.5.0

With this update the mod now works on KSP 1.1.2

Requires Kopernicus 1.0.4 and the latest modulemanager (download them separately)

 

all the links are available in the OP

 

If you want to follow the development of my mods: 

If you want to buy me a cup of coffee:   

 

Changelog:

v0.5.0 - Updated to KSP 1.1.2 and Kopernicus 1.0.4 - Added the parameter scanAltitude - Tweaked inverseRotThresholdAltitude on atmospheric bodies - Fixed AtmosphereFromGround rescaling - Fixed orbital rescaling of asteroids

 

Edited May 8, 2016 by Sigma88

[Bahamut]

@Sigma88

Thank you for detailed information. I agree that this problem is caused by using parts that are not balanced for resized system; I think that physics themselves work correctly. The same problem applies to other parts, like engines and fuel tanks, but is easier to solve. For example I use Real Fuels with sotck alike configs. With "useRealisticMass = true" parameter, I get similar mass to orbit ratio in 6.4 as I would in vanilla game. There is also SMURFF. I'll try to buff the ablator capacity in Deadly Reentry - although I will be doing it with with "trial and error" method.

 

 

 

I found that buffing the ablator amount by 2 - 2.5x (and rounding up to nearest round number) was a good balance for 6.4x. For example about 350 ablator for 1.25 heat shields and around 1500 ablator for 2.5 m heat shields.

Edited May 8, 2016 by Bahamut

[Bahamut]

Does Sigma Dimensions affect the altitude at which high orbit is set at? In more testing, I've found that while low space starts at 90K, even with 6.4x rescale, high space still starts at 250k. Any ideas if Sigma changes that or if I have a conflict with another mod that is overriding that?

[Sigma88]

2 hours ago, Bahamut said:

Does Sigma Dimensions affect the altitude at which high orbit is set at? In more testing, I've found that while low space starts at 90K, even with 6.4x rescale, high space still starts at 250k. Any ideas if Sigma changes that or if I have a conflict with another mod that is overriding that?

flyingAltitudeThreshold which is the threshold between flying low and flying high is rescaled using the "Atmosphere" parameter.

spaceAltitudeThreshold which is the threshold between in space low and in space high is rescaled using the "Rescale" parameter.

 

Unless I've messed up my parameters

 

I've checked my calculations and they work like this:

in stock kerbin high space is at 250000 meters. this means you need a circular orbit with SMA 850,000m to get there

since you have "Rescale = 6.4" you will need a circular orbit with a SMA 6.4 times bigger (5,440,000 meters)

this means that the "high space altitude" is 5,440,000 minus the new kerbin radius

so 5,440,000 - 3,840,000 = 1,600,000

if you are not seing this numbers it means there is something wrong

Edited May 10, 2016 by Sigma88

[JebThrillMaster]

On 07/05/2016 at 1:10 PM, Sigma88 said:

SD sets that parameter to the atmosphere max height

You should not need to change that.

If you are seeing the issue aftrr hyperediting a ship then just save and reload

If you are not using hyperedit then maybe it's the rotation of the planet that is too fast?

I would need a bit more info to know for sure

Here is my current Settings.cfg contents. (Used settings from the 64k settings from Paul Kingtiger).

Let me know if you need more info. Just saw Procedural Parts got updated so I'll give that a go and see if that was the issue.

// Base Settings

SigmaDimensions
{
	Resize = 6.4
	Rescale = 6.4
	Atmosphere = 1.33
	dayLengthMultiplier = 2
}

// Advanced Settings

@SigmaDimensions
{
	geeASLmultiplier = 1
	landscape = 0.125
	atmoVisualEffect = 1
	resizeScatter = 1
	CustomSoISize = 0
	CustomRingSize = 0
	atmoASL = 1
	tempASL = 1
}

 

[JebThrillMaster]

According to Kerbal Engineer with SAS and RCS off along with engine shut down a force of 0.00 to 0.003kNm goes up and down constantly likely causing the change in Apo and Peri heights. Just my guess though.

[Sigma88]

13 minutes ago, JebThrillMaster said:

According to Kerbal Engineer with SAS and RCS off along with engine shut down a force of 0.00 to 0.003kNm goes up and down constantly likely causing the change in Apo and Peri heights. Just my guess though.

did you hyperedit the craft in orbit from landed?

[JebThrillMaster]

8 minutes ago, Sigma88 said:

did you hyperedit the craft in orbit from landed?

No hyperedit installed. If you want I can provide a list of the mods I am using.

[Sigma88]

Just now, JebThrillMaster said:

No hyperedit installed. If you want I can provide a list of the mods I am using.

no need for now, could you try with dayLengthMultiplier = 4 ?

[Bahamut]

9 hours ago, Sigma88 said:

so 5,440,000 - 3,840,000 = 1,600,000

if you are not seing this numbers it means there is something wrong

Yeah, I'm guessing something is wrong because I wasn't seeing that number. Any suggestions as to how I can figure out what is affecting that value?