Newb question about Solid Rocket Boosters

[Giygus]

First, I am still pretty new to rocket design, but I have been noticing a trend lately.

Every time I try to add Solid Rocket Boosters to help out, I seem to make things worse.  Or at least, only make things a little bit better.

To clarify: by worse, I mean the rocket becomes a lot more complicated, more unstable, and I seem to get less far up into the atmosphere then without them.  And by a "little bit better", I mean I do get farther into orbit, but at what seems a very poor trade off. Something like 6 large boosters = just a noticeable amount of liquid fuel saved, never a lot.

I have only played in sandbox mode so far, but I do understand that SRB are a lot cheaper (in other modes), and are designed to be used at the start of the launch, when liquid fueled engines would be far less efficient.  That being said, I still feel like there is some trick to them, or some concept I am not grasping.

Any tips you're willing to share on using them?

[ForScience6686]

Srb are primarily used before liquid engines.  Better fuel efficiency for liquid engines the higher you go.  A picture would help but make sure you strut as well as proper placement on the radial decoupler.  Use the offset tool to place the Srb com below the decoupler.

[GoSlash27]

SRBs are cheaper, heavier, and less efficient than liquid fuels. Their best roles are 1) as a first stage from the pad where "cheap" is more important than "light" and 2) as a booster to get liquid fuel cores off the pad and into thinner air where they gain thrust.

"Moar boosters" is a running joke around here, but it's not actually a good design practice. If you take a bad design and add SRBs to help it, you won't gain much and occasionally make it worse.

 A good design is planned out before it's built. The best designs are derived mathematically.

We have excellent tutorials on the forum for how to do this. I recommend starting here:

Good luck and welcome to KSP

-Slashy

[Snark]

+1 to previous posters' excellent advice.

If you can't figure it out from the reference that Slashy gives, do post a screenshot-- it's very hard to give specific helpful advice without one, because there are so many ways to Do It Wrong™ that it's hard for us to know which particular one (or more) you might be bumping up against.

[HebaruSan]

Are you mounting them on radial decouplers so you can drop them once they're spent? If you continue dragging them along with you to space after they're empty, they'll cause extra air resistance and make the rocket heavier than it needs to be. (Yes, definitely post pics. Imgur is helpful.)

[Plusck]

I agree with HebaruSan - it does sound like they are being carried up for too long. Otherwise SRBs should, in all cases, provide at least a minor improvement.

This is, in my opinion, one of the difficulties with starting in Sandbox mode: there is a bewildering amount of options for things to stick on rockets, and it's easy to miss the basics.

[SalehRam]

For me, SRBs are good for probes, satellites and 1.25 - 2.5 meters designs with few stages. Large part count and large number of stages (for me) means no SRBs, and I can happily sacrifice more money in that case.

I still have some subassemblies for lifters with SRBs, but I rarely use them, as my standard lifters all now are a combination between orange tansk and Mainsails + Reliant engines...

My advise, use SRBs, but don't put them on everything, you can add them to support LF/O engines at launchpad, where you reduce the thrust of the engines and depend mainly on SRBs until they burnout, then set your LF/O engines on 100% thrust after you ditch the empty SRBs.

Also, SAS, fins help a lot when you have SRBs on your rocket.

[Blaarkies]

Terminal velocity, are you following it?
It sounds weird at first, but you want to go upwards at the same speed as your terminal velocity at that point(slow in deep atmosphere, faster in thin atmosphere)...this stops you from fighting too much drag, while also going fast enough so that you dont fight gravity that long.

Anyway, right-click on your SRB inside the editor(where you build the craft). See the green slider "Thrust Limiter"? Tune that down some until your TWR(Thrust / Weight) is less than 2...you will need to count and calculate this per stage(or get the KER mod, it give all the info engineers want)
Now, adding more SRB this way will get you further into orbit, because you are not overspeeding in thick atmosphere because of them

[radonek]

If SRBs make your stack unstable you are lacking control authority. Add some control surfaces. Same thing can happen if you use liquid fuel engine without thrust vectoring.

[Giygus]

Wow, thanks for all the interesting tips.  I didn't realize that this would be as complicated a question as it was.  A mistake I seem to keep making.

I'll do some tests with pictures a bit later, see if you can see anything I am missing.

[Wemb]

As others have said, you -can- use SRBs as a main engine, and they can be okay for probes and satelites - but for a big ship, you probably want to use them as sparingly as possible. 

Their primary purpose is to give your ship a good kick when it's low in the atmosphere and heavy because it's full of fuel.  Because they're completely uncontrollable, you'll generally want to ditch them as soon as you can - so it's almost never a good idea to add 'moar boosters'.  Fundamentally - bad things can happen if you a) Go too fast in the atmosphere b) have a reduced ability to steer your rocket c) can't reduce thrust, or stage at the right time - and a poor choice of SRB can easily cause all three of those at once.

If you're using them, try and

* Ensure you only use them where you need therm - i.e. have them run out of juice when your TWR is high enough to sustain flight without them
* Ensure you have plenty of control when they're running - that could be with a more steerable engine, or with bigger control surfaces, or with bigger reaction wheels, or RCS
* Don't go nuts with the TWR at launch - making it too high will just make your life difficult.  Remember, when you launch your TWR should rise as you burn fuel. If your TWR is too high, use smaller SRBs rather than limiting their thrust. A high TWR probably means you don't need so much SRB support.

Wemb

 

[Geschosskopf]

14 hours ago, Giygus said:

First, I am still pretty new to rocket design, but I have been noticing a trend lately.

.......

Any tips you're willing to share on using them?

First off, welcome to the madness forums 

Anyway, as you've no doubt realized by now, KSP is woefully lacking in basic, critical in-game information that is necessary to design successful, efficient rockets.  For a rocket (regardless of engine type) to get to space, it must have sufficient delta-velocity (aka delta-V or dV) to get from the ground to orbit, and it must have sufficient thrust-to-weight ratio (TWR) to overcome Kerbin's gravity.  It takes about 3400-3600m/s dV to achieve orbit and a TWR > 1.0 to get off the ground in the 1st place, but KSP doesn't tell you that, nor does it say what your rocket is capable of doing while you're building it.  So you have 2 options.  You can do a lot of hand calculations on paper outside the game to figure your dV and TWR from the stats of the parts (which the game DOES show to some extent) or you can use an instrumentation mod like Kerbal Engineer Redux or MechJeb, which do the math for you.  But anyway, you need to know where you stand vs. what you need or you'll never get to space, or you get there but waste a lot of money on more rocket than you need.  So I really recommend getting 1 of those mods if you haven't already.

But there's 1 other thing to worry about---going too fast during the ascent in Kerbin's lower atmosphere.  If you do that, your rocket will either disintegrate or flip out shortly after launch as it approaches Mach 1.  This means that too much TWR is a serious problem, so you want to make sure your TWR is < 2.0 to start with.  I generally go for about 1.5 for the lower stages and 1.5 to 2.5 for the upper stages, which works just fine.  Understand that a stage's TWR will increase as it runs due to fuel consumption making the rocket lighter, so you can start out pretty low and still end up with  high TWR.

And this, finally, is where we get into SRBs.  SRBs are used for the 1st part of the flight, from the ground up through the lower, thicker part of the atmosphere.  SRBs have a pretty high TWR individually and if you're using them at all, it's because you want them to provide a significant amount of the dV needed to reach orbit.  That means you'll have multiple SRBs on the ship, which means that the ship's initial TWR will be VERY high, too high to survive in the lower atmosphere.  So what you do is, on the 1st stage, add however many SRBs it takes to get the dV for that stage.  Then you right-click on one of the SRBs and drag its thrust limiter slider down until your initial TWR is about 1.5.  Now you're good to go.

To put this all together, you always design rockets from the top down and give each stage the TWR and dV it needs for the part of the flight it does.  So you start with the payload, say a small probe that will do a Mun flyby.  So you've got a small wad of electronics on top, and it needs a small engine with about 900m/s dV to get from low Kerbin orbit to Mun.  TWR doesn't matter for this engine as it will be in space and the light weight of the probe itself will automatically give it a very high TWR.  So that's your total payload.  Now you need to get this to orbit, which means everything from here down on the rocket needs about 3500m/s total dV and a TWR that starts out about 1.5.  You can usually make a relatively small, cheap, and simple LFO stack here that will have about 2500m/s and a TWR of about 1.5-2.5.  If you're going to use SRBs, that's all you need.  This stage will get from wherever the SRBs burn out to establishing a low Kerbin orbit.  This stage will not be running at launch---you launch on SRBs alone.   So then you make the bottom (or 1st) stage out of SRBs.  The SRBs need about 1000m/s of dV---add SRBs as needed to achieve that.  Now limit their thrust so that they proved about a 1.5 TWR at ignition.  In the stock game, limiting thrust has no effect on stage dV, so you're all set.

It is, of course, possible to have SRBs and the LFO core all running at once on launch (like the Space Shuttle and many other real rockets).  But this is only necessary if the SRBs alone can't provide a TWR of about 1.5, and that's generally not a problem in KSP with stock parts.

[Gojira1000]

Early in career you can hit orbit and save money with a couple of staged Thumpers and a small LF final stage with a Terrier, but once you have access to better gear, SRBs should be relegated to "initial launch kick" and that's it.

If you do use them as a primary launch system, there are 2 rules:

1) Put steerable fins on them

2) Tweak thrust to less than TWR 2 in the VAB (1.7 seems like a happy number for me)

[CSE]

Welcome to KSP!

As well as considering your rocket design, it's worth practicing your launch profile. In general, the closer your trajectory is to a gravity turn - the more you can keep your heading inside the ring of the prograde marker on the navball without hitting either outer space or the ground - the more benefit your rocket will get from its boosters. A search on the forums for "gravity turn" should find a lot of advice; recent (KSP 1.0+) is better.

For me, a lot of the KSP builds and techniques I was reading didn't really work until I'd had some launch practice and was flying a half-respectable pitch program.

[Giygus]

OK, back with pictures to try and illustrate.  First, the rocket I am using for these examples is a modified design from Vanamonde

The other ones I had cobbled together myself were deleted awhile ago after they kept failing, but I think these will be good enough for you to see where I am going wrong.

Also, today is the first day I have used RoboJeb and can see the info it is showing, bear that in mind.

Alright, this one got me to the moon with a larger lander then Vanamonde had on it:

http://imgur.com/gH7WFi9

and this is how it did getting to Low Orbit

http://imgur.com/8GiNoee

Not bad, and just about what I was hoping for.

 

This is the same rocket with SRB on it:

http://imgur.com/LyNEpWS

And how far it got with the boosters: 

http://imgur.com/7iTelMq

That is what I mean by "just a little bit better".  I add four giant boosters, and I get about 1/3 of a tank of liquid fuel saved.

 

Now this design down here was a failure no matter how I tried to modify it.  Maybe you can give some tips on what I did wrong.  Again, this is just Vanamonde's rocket with a little touch of newbness added by yours truly:

http://imgur.com/1VOgul5

with SRB:

http://imgur.com/mt1yFpV

Specifically, this one was too wobbly, and suffered from that "flipping" thing when I detached the 3 liquid fuel tanks.  Oddly, with the SRBs added, the main engine in line would, I think, be hit by something after detaching the 3 liquid tanks, and refuse any commands, including clicking on it.  Not sure whats up with that.

Album with all pictures: http://imgur.com/a/cMlhh

Edited March 2, 2016 by Giygus
pictures messed up

[SalehRam]

I would leave the more detailed stuff for the other better guys than me... however one thing I would like to note...

I would not use a Skipper in my lifter stage while I have already unlocked the Mainsail... maybe this is your main problem, try replacing the Skippers with Mainsails and adjust their thrust.to get your 1.5 TWR and I bet the whole rocket will be much better than it is now... Skippers are good in vac. but if you got an atmosphere stage with a place for 2.5m engine, use the Mainsail...

 

Compare the Isp for both and you would see the difference. Isp is the thing that will make you choose an engine over another. Plus the Mainsail has better thrust than the Skipper already...

 

Edited March 2, 2016 by SalehRam

[Kerbart]

At launch, your TWR (Thrust to weight ratio) can be close to 1. That means that your rocket will spend a relative large amount of time not gaining speed. SRB's can help you over that "first bump" and basically give you literally a little kick that gets you going.

But once those SRB's burn out, and they do burn out pretty quickly, I'm back at square one!

Ah! Not quite! First of all, you've already burnt off some fuel, so that alone will have improved your TWR compared to the launch pad. Second of all, even if your TWR is still close to 1, you're at least moving which will get you closer where you want to be (although you really need the speed at one point, of course). And #2 does get you, eventually, out of the lower atmosphere with the high air resistance, meaning that you'll get more bang for your thrust-buck. Finally (although not advisable with low TWR ratios) it allows you to have transitioned to a flatter trajectory -- the horizontal thrust component doesn't get eaten by gravity, like the vertical thrust component (although it helps to go "fast enough to miss the ground" at that point)

[wumpus]

40 minutes ago, Kerbart said:

At launch, your TWR (Thrust to weight ratio) can be close to 1. That means that your rocket will spend a relative large amount of time not gaining speed. SRB's can help you over that "first bump" and basically give you literally a little kick that gets you going.

If you do this, I can only recommend enough SRBs to get a significantly higher than 1.0 (1.5 will do nicely).  If not (which *will* work, and gives a little better control) I would at least recommend side drop tanks (attach fuel tanks to decouplers on your main fuel tank, then attach fuel lines from your drop tanks to your main tank) so you don't have to drag the empty tanks with you while you gain acceleration.  To take this further (typically done only in sandbox because it costs too much) put engines on those "drop tanks" and discover the old "asparagus staging" method.

Note: have you installed Kerbal Engineer yet?  It is virtually necessary to determine rocket performance (I think more boosters is a joke left over from earlier aero models which were vastly more forgiving).  If your trust to weight ratio (TWR) is much more than 1.5 (especially more than 2.0) you will certainly see that adding SRBs (or any means of adding more thrust) isn't going to help at all.

But if you are doing fine without SRBs, then don't bother with SRBs.  Some of us like them, but I suspect it has more to do with slavishly reducing costs and learning with more limited part selection (when using a lot of basic thumpers made sense).

[GoSlash27]

Giygus,

 In both examples, you already had a t/w of 1.6 and enough DV to get the job done. All adding SRBs was going to do for you is increase thrust, which isn't going to help you. Now... if you had added fuel so the SRBs had something to lift, you would've noticed a dramatic increase in DV. Or if you had left the first stage turned off and used the SRBs exclusively as the first stage.

 As you can see, just tacking on "Moar Boosters®" as an afterthought does very little to help, and can easily cause problems that render a design useless.

The central problem is that you're following an incorrect process for designing a lifter.
1) You're taking a working design, adding payload to it, and then trying to get it to work again with additional boosters.
2) You're neglecting aerodynamics, which is actually important in 1.0 and beyond.

Foxster's tutorial addresses both of these problems and walks you through how to design a lifter the right way.

Best,
-Slashy

 

[Giygus]

OK then.  I have a lot to learn about Dv it would seem.

Thanks to everyone trying to give help, but I am going to have to keep practicing until I get it.

I do have one last question on topic here, this picture here: Chart

Does the Dv values shown have some leeway?  For example, it says you need 3200 to leave Kerbol, could you ever design one that needed less? (I would assume an inefficient design would need more?)

[Plusck]

9 minutes ago, Giygus said:

OK then.  I have a lot to learn about Dv it would seem.

Thanks to everyone trying to give help, but I am going to have to keep practicing until I get it.

I do have one last question on topic here, this picture here: Chart

Does the Dv values shown have some leeway?  For example, it says you need 3200 to leave Kerbol, could you ever design one that needed less? (I would assume an inefficient design would need more?)

There is virtually no leeway in that chart - almost all of the values are "most efficient".

However, there is leeway if you can use gravity assists. For example, timing your burn out of Kerbin's system to take you screaming past the retrograde side of the Mun at low altitude will significantly increase your orbital velocity, saving you a hundred or so m/s dv.

Similarly, if you can get to Jool then you have a multitude of options to use Jool or its moons to slow you down or speed you up. So yes, with a 2km/s burn from Kerbin to Jool (950 to edge of Kerbin SOI and 980 from there to Jool, plus inevitable inefficiencies), you should be able to get a gravity assist out of Kerbol system.

And finally, going from moon to moon within a planet's SOI does not require you to go down the branch and back up, as it were. The map cannot possibly show all the different possibilities of transfer orbits within each system.

[FullMetalMachinist]

1 hour ago, Giygus said:

Does the Dv values shown have some leeway?  For example, it says you need 3200 to leave Kerbol, could you ever design one that needed less? (I would assume an inefficient design would need more?)

Not really any leeway, no. Those values are absolute best case scenarios, and don't take into account inclination changes* and other troubles. But you can get some of those values for less by using gravity assists. 

Edited March 3, 2016 by FullMetalMachinist
*Apparently they account for worst case inclination changes. Learning is Fun!

[GoSlash27]

1 hour ago, Giygus said:

OK then.  I have a lot to learn about Dv it would seem.

Thanks to everyone trying to give help, but I am going to have to keep practicing until I get it.

I do have one last question on topic here, this picture here: Chart

Does the Dv values shown have some leeway?  For example, it says you need 3200 to leave Kerbol, could you ever design one that needed less? (I would assume an inefficient design would need more?)

The surface to orbit numbers are only estimates for bodies with atmospheres. You can do better or worse than the listed values depending on design and flight efficiency.

Launch and landing numbers for airless bodies are theoretical minimums, rounded up to the nearest 5 m/sec. You're not going to beat them, and you're not liable to match them.

Interplanetary transfers are an average Hohmann transfer assuming perfectly circular orbits and identical planes. You will get generally identical results on an average transfer, not counting plane changes (the map includes a worst- case for plane change).

If you get slick and use gravity assists, integrated plane changes, bi-elliptic transfers and the like, you can dramatically undercut these values.

But for general mission planning, these maps are good ballpark numbers for what you will need. Just keep a reserve to cover the variations and pack extra DV for landings on airless bodies.

*edit* just as with MJ, the DV maps are limited AFA what you can do with them. If you create a spreadsheet to model the DV budget for transfers, you can find shortcuts that allow you to undercut the map.

For example, a return from the Mun to Kerbin should cost a little over 310 m/sec DV according to the map, but if you calculate it out, you will find that it's only 269 m/sec.

Best,
-Slashy

Edited March 3, 2016 by GoSlash27

[Wemb]

On 02/03/2016 at 1:34 PM, Giygus said:

Wow, thanks for all the interesting tips.  I didn't realize that this would be as complicated a question as it was.  A mistake I seem to keep making.

Well, it's not rocket science, is it? Oh, wait... :-)

Wemb

Edited March 8, 2016 by Wemb

[Giggleplex777]

On 3/1/2016 at 2:30 PM, SalehRam said:

I would leave the more detailed stuff for the other better guys than me... however one thing I would like to note...

I would not use a Skipper in my lifter stage while I have already unlocked the Mainsail... maybe this is your main problem, try replacing the Skippers with Mainsails and adjust their thrust.to get your 1.5 TWR and I bet the whole rocket will be much better than it is now... Skippers are good in vac. but if you got an atmosphere stage with a place for 2.5m engine, use the Mainsail...

 

Compare the Isp for both and you would see the difference. Isp is the thing that will make you choose an engine over another. Plus the Mainsail has better thrust than the Skipper already...

 

Mainsails are twice as heavy, only slightly more efficient in the lower atmosphere, and much more expensive than Skippers. Looking at his setup, the Skippers and SRBs should work just fine.