A Degree Makes a Difference - or How I Learned To Stop Worrying About Drag and Concentrate on Gravity Losses

[Norcalplanner]

Based on an idea born in another thread, I decided to do some experiments in RSS (which I'm currently using as my career, with SMURFF - no RO) to look at TWR, drag losses, gravity losses, and ascent profiles.  I created a basic rocket with four Kickbacks, then simply thrust limited the Kickbacks to 50%, 75%, and finally 100%.  After an initial tip to the east, SmartASS would hold the craft on a surface prograde vector.  Once the navball switches to an orbital reference, SmartASS changes to hold orbital prograde until the Ap is 250 km.  MechJeb is then told to circularize at AP.

After conducting 18 test runs, I've created an album with the best run at each thrust level, posted below.

It was amazing that a single degree change in the initial tip can make a big difference - in one case, reducing gravity losses by over 100 m/s.  It was also eye-opening to see just how small the drag losses are compared to the gravity losses.  

Conclusions based on all this:

1. Launch with an initial TWR between 1.5 and 1.9.

2. Don't worry about drag - gravity losses are much larger and more important.

3. Ignore the flame effects.

4. At higher thrust levels, crank it to the east immediately after launching, and be precise about it.

5. Try to keep vertical velocity below 800 m/s.  If it's over 1 km/s, you're going to have noticeably higher gravity losses.

6. Getting a rocket to orbit in RSS for less than 9 km/s of delta V is very doable.  Using less than 8.6 km/s is a harder but still achievable goal.  

UPDATE: After making all these ascents in RSS, I redid the tests in stock with a slightly different rocket (but same concept) to see what was the same, and what was different,  You can find the stock version here.

UPDATE 2: After editing this OP, the embedded imgur album disappeared.  You can find the album here.

Edited September 20, 2016 by Norcalplanner
Added link to stock thread and link to imgur album

[SyzygyΣE]

Yeah you've made many good points here. I don't use SMURFF, but the principles are practically the same in my RO career. I try my best to crank up my starting TWR as high as possible without having difficulty controlling the vessel, within the 1.5–1.9 you recommend.

The biggest problem I find is that as the fuel tanks near depletion and mass is reduced, the TWR starts cranking up. I just got into orbit in my RSS and RO career but the first stage has a maximum TWR of almost 9, making it pretty difficult to control near burnout. How I manage to deal with this is to use an engine cluster of 5 instead of a single, powerful engine. When the TWR reaches about 5, I switch off all 4 of the outer engines and leave the central burning, which "resets" the TWR to 1. That way, upon burnout, the maximum TWR changes to a manageable 3.5 with only the single engine. Many of the engines cannot be throttled so I just kind of have to deal with it this way.

Not exactly sure this is the best way to keep a decent TWR at launch while keeping the max TWR low. I get a feeling I'm dragging 4 engines worth of useless mass after I switch them off so I can steer the rocket. If there is a better way anyone recommends me to do it, I would appreciate your input.

Edited July 10, 2016 by SyzygyΣE

[Mastikator]

Agreed, I've found that I always do better if I go faster earlier. You're only going too fast if you're burning up.

[Norcalplanner]

To give a better idea of where the title of the thread came from, here's some data from the runs at an initial TWR of 1.53.  All ascents went straight up until 30 m/s was reached, and then an initial tip to the east was dialed in with SmartASS.

A Degree Makes a Difference - 1.53 TWR

Initial Turn Angle	Drag Losses	Gravity Losses	Total Delta V to Orbit
5 deg	63.4 m/s	1,622.0 m/s	9,047.0 m/s
6 deg	69.6 m/s	1,441.4 m/s	8,855.4 m.s
7 deg	76.6 m/s	1,306.4 m/s	8,720.3 m/s
8 deg	84.2 m/s	1,208.3 m/s	8,627.4 m/s

 

[SyzygyΣE]

4 minutes ago, Norcalplanner said:

 

A Degree Makes a Difference - 1.53 TWR

Initial Turn Angle	Drag Losses	Gravity Losses	Total Delta V to Orbit
5 deg	63.4 m/s	1,622.0 m/s	9,047.0 m/s
6 deg	69.6 m/s	1,441.4 m/s	8,855.4 m.s
7 deg	76.6 m/s	1,306.4 m/s	8,720.3 m/s
8 deg	84.2 m/s	1,208.3 m/s	8,627.4 m/s

 

 

 

 

I see. So with a greater turn angle the drag loss increases, but that increase is paltry compared to the reduction of the gravity losses. Interesting. I was always taught to perform a 5° initial turn angle. Perhaps I'll try experimenting with some different angles.

Edited July 10, 2016 by SyzygyΣE

[Norcalplanner]

Just now, SyzygyΣE said:

I see. So with a greater turn angle the drag loss increases, but that increase is paltry compared to the reduction of the gravity losses. Interesting. I was always taught to perform a 5° gravity turn. Perhaps I'll try experimenting with some different angles.

Yep.  It was that initial difference between 5 degrees and 6 degrees which was the eye opener.  In exchange for losing 6.2 m/s to increased drag, gravity losses were reduced by 180.6 m/s, and the whole craft got to 250 km orbit for 191.6 m/s less delta V. 

[Mister Kerman]

That's pretty useful if it works on all sizes and shapes of craft.

[Norcalplanner]

2 hours ago, Mister Kerman said:

That's pretty useful if it works on all sizes and shapes of craft.

I don't think these numbers are universal, so I wouldn't put a lot of stock in them for other situations. For example, this craft is pretty clean aerodynamically, with no fins, solar panels, struts, or other radially- mounted equipment. What I think IS universal is that altering your ascent profile to minimize gravity losses instead of  worrying about drag losses is a good idea.

[Mister Kerman]

I'm going to try 8 degrees tonight. I'm building very simple rockets right now as I'm early my career and it's been a few updates since I've last played. (Pre-Unity) I've been pitching further than that so this sounds good. Gravity turn should handle the rest and it'll be nice to have a semi reliable starting point to start my gravity turn.

 

Eyeballing is a terrible way to do things. Getting reliable information from people who have already been there and done that to save on legwork is more my style. I think it's more or less why this community is so tight; common goals and cooperation for unified success. Comradery in a single player game is something you don't really see these days.

[jedensuscg]

I was thinking about doing an rss career and came across this. This beings up stuff I never even knew I had to know. Not even sure if any of this can transfer to non rss ksp. Might explode spectacularly a few times. In fact this is counter to how I launch in my non RSS with FAR career, where going to fast is dangerous and causes the craft to become unstable and doing any gravity turns to early is a nono. I have to limit some of my rockets to 14-16 m/s acceleration to prevent hitting too much speed to early and suffering from drag and dynamic pressure disasters(though I think FAR has a lot to do with this. I love far, makes for some far better launches if you understand your limits)

Going to save this for my toolkit when I start an RSS career. 

[Laie]

1 hour ago, Norcalplanner said:

What I think IS universal is that altering your ascent profile to minimize gravity losses instead of  worrying about drag losses is a good idea.

12 hours ago, Mastikator said:

You're only going too fast if you're burning up.

I agree to the first, and strongly suspect that the second statement is true. Though there's other consideration besides burning up; I for one find that I want aerodynamics to be pretty minimal when staging the first time; too many rockets have spazzed out in that brief phase of uncontrolled flight.

For what it's worth, my tipping regime is to use the +1 clickers on SmartASS to make a smooth turn until 80 degrees, then set it to follow prograde. That technique minimizes yaw deviations. The ultimate destination is 30@30: when I cross the 30km mark while pointing 30deg above the horizon, the original turn was performed well. If necessary I again use the +-1 clickers for minor adjustments, but strive to keep everything at zero after reaching mach-1, and until altitude >15km.

That's in RO, with RO-typical TWRs (for Titan-Gemini, 80degrees@130m/s is a good start; for SaturnV-Apollo you need to drag it out to almost 200m/s). 30@30 appears to be a good breakpoint for many (most?) rockets. It no longer works when the second stage has a TWR<0.8, or if the first stage burns longer than 150sec.

[moogoob]

2 hours ago, Mister Kerman said:

Comradery in a single player game is something you don't really see these days.

KSP is the only game where I've found players helping each other through charts, diagrams and spreadsheets. It's truly unique!

-As to the OP - fascinating! I think a big part of why your approach is novel to many is that many are used to launching on Kerbin - a planet with a tenth the size of Earth, but full gravity and atmospheric thickness. The atmo being a greater proportion of the planet's radius means our (stock solar system) launches are much more vertical than they would be on Earth.

Always wanted to try an RSS game... not sure RO was my speed but glad to see you can still orbit etc well using stock parts without modding anything in.

[wumpus]

I've been doing this for awhile (on Kerbin, not Earth).  The things I've noticed:

• Don't touch the throttle (including SRBs) unless you have control issues.  Gravity losses (for TWR<=2.0) are worse than aero.
• Control issues for TWR>1.5 can get nasty.  Expect your "gravity turn" to look more like beta and before than a post 1.0 gravity turn.
• Ignoring fairings doesn't seem to hurt that badly.  No idea if they've fixed the fairing bugs* to make me want to use them.

I'll have to look into keeping vertical velocity <1km/s.  I've always watched my ap during ascent, and will have to think about this tool.  I also haven't used mechjeb for just starting the gravity turn.  It always seemed to be an "all or nothing" thing that didn't work with such high powered rockets.

* the "fairing bug" I'm talking about was from 1.0.4 or so, and moved the center or pressure in front of the fairing.  This made everything with a fairing less stable and made fairings much less attractive in general.  Haven't heard if it has been squashed.

[Raptor9]

2 hours ago, wumpus said:

• Don't touch the throttle (including SRBs) unless you have control issues.  Gravity losses (for TWR<=2.0) are worse than aero.

I would challenge the accuracy of that statement in the lower atmosphere.  If your flying to fast in the thicker atmosphere layers you're actually burning more fuel than necessary due to excessive drag.  That's why some rockets don't throttle up until going through Max Q. The amount of fuel spent to accelerate and maintain higher speeds are wasteful since the amount of drag acting on the rocket negates the benefits of the quicker acceleration.  Once you get out of the lower atmo layer, drag will start dropping off rapidly enough that you can put the pedal to the medal again.  But yeah, it's a balancing act to prevent gravity losses from creeping up again.  Darned rocket science....

Having said that, I never reduce the thrust of any jettisonable boosters.  Otherwise they're expending most of their power just lifting themselves.

Edited July 10, 2016 by Raptor9

[Mister Kerman]

52 minutes ago, Raptor9 said:

I would challenge the accuracy of that statement in the lower atmosphere.  If your flying to fast in the thicker atmosphere layers you're actually burning more fuel than necessary due to excessive drag.  That's why some rockets don't throttle up until going through Max Q. The amount of fuel spent to accelerate and maintain higher speeds are wasteful since the amount of drag acting on the rocket negates the benefits of the quicker acceleration.  Once you get out of the lower atmo layer, drag will start dropping off rapidly enough that you can put the pedal to the medal again.  But yeah, it's a balancing act to prevent gravity losses from creeping up again.  Darned rocket science....

Having said that, I never reduce the thrust of any jettisonable boosters.  Otherwise they're expending most of their power just lifting themselves.

I've had to adjust my thrust output on some SRBs to make up for lack of stability from pushing through atmosphere too quickly. They last longer that way so it's not so bad.

[Raptor9]

2 minutes ago, Mister Kerman said:

I've had to adjust my thrust output on some SRBs to make up for lack of stability from pushing through atmosphere too quickly. They last longer that way so it's not so bad.

Yeah, in retrospect, that is a possible issue.  I suppose I just don't have any SRB-equipped rockets that are light enough to encounter that effect.

[Norcalplanner]

Thanks everyone for the kind words and interest in this.  Just to post a brief update with FAR - 

I've spent the last few hours with FAR installed, stock settings, and aerodynamic failures enabled.  I tried to replicate the results I was getting with the stock-ish RSS aero.  Every single ascent I tried with that craft ended in failure and rapid unplanned disassembly.  It seemed to "bobble" (for lack of a better word) while going transonic, drift off pro-grade, and disassemble.  (Full disclosure - I'm also running Kerbal Joint Reinforcement, and am not running Deadly Reentry, Real Heat, or anything else which might contribute to my rocket breaking up.)

However, I discovered an easy two-part solution, which works at least with this craft:

1. Add fins.  I added six of the delta fins around the base of the LFO core.  This turned the "bobble" into something which was barely noticeable.  To keep the weight approximately the same for the tests with fins added, I emptied out the ablator and monoprop.

2. Reduce the initial tip eastward by one degree.  This is enough to account for the greater drag down low imposed by FAR.

That's it!  With those two modifications, the craft got to orbit again just fine, with gravity losses only slightly greater than in stock RSS aero.  

I'll post some more images later tonight, along with additional data.  Maybe some of the FAR-types can help translate all the numbers and colored lines which are in some of the screenshots.

- - - - -

Edit - As suggested by THX1138, I added back in the ablator and monoprop.  The number of fins was reduced to four, and new runs were made.  See the post a few below this one for more info.

Edited July 11, 2016 by Norcalplanner
Update

[moogoob]

4 hours ago, wumpus said:

* the "fairing bug" I'm talking about was from 1.0.4 or so, and moved the center or pressure in front of the fairing.  This made everything with a fairing less stable and made fairings much less attractive in general.  Haven't heard if it has been squashed.

I have good news: it's been fixed. Use them all the time. Fairings work better than ever for me now!

There still may be some bugs involved with fairings, but the "blue arrow out in the air in front of your rocket" bug is gone. I mean, I still used them from time to time when the bug was in place, just with significantly more fins on my rocket.

[Norcalplanner]

6 hours ago, Laie said:

I agree to the first, and strongly suspect that the second statement is true. Though there's other consideration besides burning up; I for one find that I want aerodynamics to be pretty minimal when staging the first time; too many rockets have spazzed out in that brief phase of uncontrolled flight.

For what it's worth, my tipping regime is to use the +1 clickers on SmartASS to make a smooth turn until 80 degrees, then set it to follow prograde. That technique minimizes yaw deviations. The ultimate destination is 30@30: when I cross the 30km mark while pointing 30deg above the horizon, the original turn was performed well. If necessary I again use the +-1 clickers for minor adjustments, but strive to keep everything at zero after reaching mach-1, and until altitude >15km.

That's in RO, with RO-typical TWRs (for Titan-Gemini, 80degrees@130m/s is a good start; for SaturnV-Apollo you need to drag it out to almost 200m/s). 30@30 appears to be a good breakpoint for many (most?) rockets. It no longer works when the second stage has a TWR<0.8, or if the first stage burns longer than 150sec.

Laie, 

The problem you describe with a rocket losing control during staging is one of the reasons why I've gravitated towards using SRBs in nearly all my rockets.  It allows a larger first stage LFO core, so the gimbaled engine keeps firing and provides control during the first staging event (which is ditching the SRBs).  When that center LFO core finally runs out a minute or two later, the atmosphere is so thin aero is no longer an issue.

And what you're describing using the clickers with SmartASS is exactly what I'm doing.  I'll wait until the vertical velocity is where it needs to be, then I rapidly click the minus button for however many degrees of initial tip I need.  It smooths out the control inputs just a bit, as you mention.

And I fully acknowledge that this test rocket isn't the most realistic - it has a crazy powerful upper stage, with an initial TWR of 1.87.  Probably not the best example in hindsight, but it certainly helps keep the trajectory flat in the latter part of the ascent.

[Mister Kerman]

36 minutes ago, Norcalplanner said:

Laie, 

The problem you describe with a rocket losing control during staging is one of the reasons why I've gravitated towards using SRBs in nearly all my rockets.  It allows a larger first stage LFO core, so the gimbaled engine keeps firing and provides control during the first staging event (which is ditching the SRBs).  When that center LFO core finally runs out a minute or two later, the atmosphere is so thin aero is no longer an issue.

I do the same thing. Disposable SRBs with an LFO core for gimbal steering. Either that or reduce/disable gimbal on LFO radial boosters because they provide too much overcorrection with SAS on.

[THX1138]

1 hour ago, Norcalplanner said:

To keep the weight approximately the same for the tests with fins added, I emptied out the ablator and monoprop.

 

I don't think that's a fair test for the fins. The fins have weight and if you want to use fins you have to accept that so testing in a scenario where their weight is nullified isn't a fair comparison in my view.

[Norcalplanner]

1 hour ago, THX1138 said:

 

I don't think that's a fair test for the fins. The fins have weight and if you want to use fins you have to accept that so testing in a scenario where their weight is nullified isn't a fair comparison in my view.

I think you may be right.  I'll do a few more runs with everything back where it was.  TWR will be a little lower in the first stage due to the mass of the fins, but the upper stage will be identical.

[Norcalplanner]

Here are the final versions of the runs using FAR.  The only physical change made to the rocket is the addition of four small delta fins.  Mass of the underlying rocket is kept the same (unlike the initial tests I mentioned a few posts ago).

Data crunching was hampered by MechJeb not being able to capture the amount of drag losses with FAR installed, so all we're left with is gravity losses, delta V to orbit, and remaining delta V in the upper stage tank.  I've summarized things in a table below.

A Degree Makes a Difference - FAR edition

Initial TWR	Initial Turn Angle and Speed	Gravity Losses	Total Delta V to Orbit	Delta V Remaining
1.18	5 degrees at 70 m/s	1,577.4 m/s	8,990.4 m/s	1,101 m/s
1.53	8 degrees at 35 m/s	1,253.8 m/s	8,695.6 m/s	1,641 m/s
1.87	12 degrees at 10 m/s	1,129.3 m/s	8,594.5 m/s	1,823 m/s

While we don't have actual drag loss data, it's clear from the last row that drag is more of a factor with FAR.  12 degrees was the maximum tipover which made it to orbit, as compared to 14 degrees in stock RSS aero.  Nevertheless, the data show that at least for this craft, only minor modifications were needed to function adequately with FAR installed.  The 50% thrust ascent used identical settings as stock; the 75% thrust setting just needed 5 m/s of additional vertical speed before starting the tipover, and the 100% thrust setting required two degrees less tipover.

It was also enlightening to see just how much drag was being shown on each of these craft - I was able to grab a screenshot right at Max Q during each ascent so you can see just how much drag was being generated.

One final note - on a whim, I removed the two outermost SRBs, and gave the two remaining SRBs the slanting advanced nose cones.  With the SRBs operating at full thrust at an initial TWR of 1.57, this smaller and cheaper version was able to get to orbit with an initial tipover of 7 degrees at 38 m/s.  Doing so incurred gravity losses of 1,396.6 m/s, cost a total of 8,803.3 m/s of delta V, and made it to orbit with 1,097 m/s left in the tank.  These numbers are comparable to the 50% thrust option, at a lower cost and weight.

Edited July 11, 2016 by Norcalplanner
clarified things

[Norcalplanner]

On 7/9/2016 at 10:04 PM, SyzygyΣE said:

Yeah you've made many good points here. I don't use SMURFF, but the principles are practically the same in my RO career. I try my best to crank up my starting TWR as high as possible without having difficulty controlling the vessel, within the 1.5–1.9 you recommend.

The biggest problem I find is that as the fuel tanks near depletion and mass is reduced, the TWR starts cranking up. I just got into orbit in my RSS and RO career but the first stage has a maximum TWR of almost 9, making it pretty difficult to control near burnout. How I manage to deal with this is to use an engine cluster of 5 instead of a single, powerful engine. When the TWR reaches about 5, I switch off all 4 of the outer engines and leave the central burning, which "resets" the TWR to 1. That way, upon burnout, the maximum TWR changes to a manageable 3.5 with only the single engine. Many of the engines cannot be throttled so I just kind of have to deal with it this way.

Not exactly sure this is the best way to keep a decent TWR at launch while keeping the max TWR low. I get a feeling I'm dragging 4 engines worth of useless mass after I switch them off so I can steer the rocket. If there is a better way anyone recommends me to do it, I would appreciate your input.

Any chance you could post a photo of your craft?  It sounds like it's all LFO in the first stage.  If so, you may want to try reallocating the mass distribution between the stages so that the upper stage has more mass.  You could also try reducing your craft down to three or so engines, and adding some SRBs for the initial boost off the pad.

[Streetwind]

Thank you! I've been preaching this for forever over in Gameplay Questions, where most other people giving advice still tell newcomers to dial back their thrust, avoid aerodynamic effects and generally fuss a lot about drag losses. Heck no, people! That stopped being a thing with 1.0, get with the times!

What surprised me though was the giant difference between drag and gravity losses that you recorded. I always operated under the assumption that gravity losses were about five times drag on a typical ascent, but your numbers show more along the lines of 20 times. Perhaps that stems from how the stock atmosphere is scaled in RSS? After all it's tuned for FAR, which generally has more drag and also scales differently.