TWR & Thrust limiting: SRBs vs LF engines

[KocLobster]

I was unable to find an answer or any suggestions for this question. Simply put, I'm trying to figure out if one is better than the other. When you have a bit too much TWR and you want to lower your thrust to more optimal levels, are there benefits from lowering the SRBs' thrust over limiting your LFE's thrust? Is doing one more efficient than the other? Generally, I will lower my LFE until it reaches the my desired TWR instead of the SRBs. I do this because it would use less LF and seems to be more efficient and conserving of dV on the surface. However, I don't have any research to back this up, it just seems like the most obvious answer to me. Sometimes, I will lower both.

The way I understand it, is that your SRBs are there mainly to get you up out of the soupy atmosphere without eating a ton of LF, like if it were left just to your liquid fuel engines to do so. By putting the brunt of the work on the SRBs, it conserves more fuel, and therefore more dV. This is why I usually let my SRBs always run at 100% thrust and lower my LFE(s) on that lowest stage. I find doing this is much more exact than leaving it at 100% and throttling down once in flight. Usually, once I ditch the SRBs in flight, I will slide the thrust limiter on the LFE back to 100%. I will also a lot of times throttle down a little though.

If this isn't ideal or you do it differently (I know a lot of people limit the thrust on SRBs instead), please let me know and why. Any suggestions would be great, with explanations of why even better.

 

Thanks KSP community.

[Snark]

One thing to bear in mind is that SRBs have a much crappier Isp than LF engines do.  For optimum dV efficiency, you generally want to burn your lowest-Isp fuel first.  Therefore, in general, burning the SRBs first is a good idea.  (For my own ships, I almost always launch on SRBs alone, and don't ignite the LF engines until the SRBs are all done.)

That said, SRBs can be awkward due to the inability to control their thrust once you ignite them.  They have a tendency to really slam your TWR hard near the end of the burn-- both because your mass is a lot lower, and because their thrust climbs about 20% as the air gets thinner.

So it's nice to have a way to smooth out the ride.  Some folks achieve this by running LF engines concurrently with the SRBs, and throttling the LF down as the SRBs burn.

My own preference is to achieve this by having multiple concurrent SRB stages with varying burn times.  For example, a big ship might have four Kickbacks and four Thumpers.  They all ignite together off the pad, but the Thumpers burn out and are jettisoned first, as the Kickbacks continue to burn for a while.

Another technique I like to do with SRBs is what I call "poor man's asparagus."  I attach them in two symmetry groups with different thrust limiters.  For example, let's say I'm launching a craft with eight radial SRBs, and I calculate that to get my desired TWR of 1.5 on the pad, I should set their thrust limiters to 85%.  So instead of making a single symmetry group of 8, set to 85%, I instead make two symmetry groups of 4 each.  The first one (call it group A) is set to 100% thrust.  The second one (call it group B) is set to 70% thrust.  The ship takes off using all 8  boosters together.  Group A burns out first and is jettisoned, while group B continues for a while longer, then I ignite my LF when group B burns out.

That's just the way I happen to like doing it; works well for me.    There are a lot of ways to slice that particular cake, though; there are plenty of other techniques that work well, too.

[Rhomphaia]

you get more DeltaV if you burn off the least efficient propellant first.  since LF engines are more efficient than SRBs you are correct that when you run both together you should limit the LF rather than the SRBs to your desired TWR.

[KocLobster]

So I do have the right idea by allowing the SRBs to take the brunt of the effort of getting off the ground then? I really like your idea of 'poor man's asparagus', that sounds quite efficient. I don't ever use as many engines as you do apparently, though. Usually I never have more than 4 SRBs; right now a ship I'm about to launch only needs two SRBs. From what you've said though, I think maybe I could improve upon my general designs. Maybe I should be using more SRBs, so that when I'm still in the very thick atmosphere and right after launching, I can launch my rocket purely off of SRBs, and not use the help of my LFE(s) at all. In general with my rockets, if I did that right now, I would have far too low of a TWR than necessary.

[Snark]

1 hour ago, KocLobster said:

I don't ever use as many engines as you do apparently, though.

MOAR BOOSTERS!

Well, yeah, I do like me some SRBs.  I love those things.  Would never, ever use them anywhere but the launchpad, but they're great there, as long as you don't have to worry about launchpad weight limits.  Dirt cheap, lots of bang for the buck.  Plus I like them for the purely irrational reason that I happen to really enjoy taking off on a roaring column of flame.

My typical booster designs, for different size ranges, are:

• Small:  A Swivel on a 4-ton LFO tank, bracketed by two radial Hammers
• Medium:  A Skipper on a 16-ton LFO tank, bracketed by 8 Thumpers in two groups of 4
• Large:  A Mainsail on a Big Orange Tank, bracketed by 8 Kickbacks in two groups of 4
• Huge:  A Mammoth on the giant tank, bracketed by 12 Kickbacks in two groups of 6

With some variation to fill in the gaps, but those are my most common configurations.

[Plusck]

My view on this is simply and extension of general principles: you want to spend as little time getting up to speed as possible, and dump excess weight as soon as possible.

Since overdoing the "get up to speed" bit makes it difficult to get an efficient gravity turn, so you end up going up too steep, the next best thing to "get up to speed quickly" is "get more useful mass up to speed". So by that first measure, SRBs should be burning as fast as possible, to get more LF up higher.

And SRBs are very easy and cheap to dump. So by that measure too, they should be burned and dumped as quickly as possible.

So I light everything on the launchpad, throttle down at about 200 m/s to ensure a good gravity turn, then throttle up again when the SRBs run out or once the gravity turn has gone far enough.

[Waxing_Kibbous]

I've had pretty good luck using SRBs in combination with a vectoring LF engine at launch- I keep the LF at low throttle to steer until the SRBs get ditched then throttle up. I also throttle down the thrust on the SRBs because I find I can get a better launch profile when they are mellowed out a bit- really though it depends a lot on the individual vehicle.

[KocLobster]

This is some great advice, thank you guys very much. I'm glad I was on the right track.

@Snark

That is an ungodly amount of rockets  The insanity! Kickbacks almost seem counterproductive because they are just so ridiculously heavy.

[numerobis]

The solution to TWR being too high for the last 20s of a kickback's flight and therefore sending you up too steeply is pretty simple: pitch over harder. If you launch kickbacks at TWR 1.5 and you're at 45 degrees pitch by 6km or a bit earlier, then you'll be about 30 degrees at 12km when they run out and you've got TWR over 3.

The solution to TWR being too high and thus your spacecraft falls apart is to add one (1) strut in exactly the right spot. I've got a spacecraft with 18 kickbacks on each side. Those two boosters each connect to the core stage with a small hardpoint and a single strut. High angle of attack or a lot of oscillation will tear the sides right off, but if you've got a nice stable spacecraft and you stick to prograde you'll be fine.

[Wemb]

19 hours ago, KocLobster said:

So I do have the right idea by allowing the SRBs to take the brunt of the effort of getting off the ground then? I really like your idea of 'poor man's asparagus', that sounds quite efficient. I don't ever use as many engines as you do apparently, though. Usually I never have more than 4 SRBs; right now a ship I'm about to launch only needs two SRBs. From what you've said though, I think maybe I could improve upon my general designs. Maybe I should be using more SRBs, so that when I'm still in the very thick atmosphere and right after launching, I can launch my rocket purely off of SRBs, and not use the help of my LFE(s) at all. In general with my rockets, if I did that right now, I would have far too low of a TWR than necessary.

I think so yes - thought I haven't run the numbers - but think, you're ditching them  - so the quicker you empty them and dump the weight, the better.  If your TWR is too high, this probably means that you're got too many SRBs - reduce the number or size of them, rather than lowering their thrust.  The problem with using too many SRBs is that they're utterly uncontrollable, and that's the last thing you want when you're trying to do a gravity turn or flying through an atmosphere.

Wemb

[Vegetal]

Just a note: Sometimes it's good to have a smaller LFO engine when you have SRBs, because when you ditch them, you only need enough TWR to keep your apoapsis in front of you.

I also use a little plugin by Crzyrdm that allows you to set the thrust limiter of SRBs both for start up and burn out, so you have decreasing thrust as the fuel burns up. It helps a lot in making your rocket more controllable.

[wumpus]

Generally speaking, you never want to limit the thrust of your SRBs.  If you are going to throttle anything (and unless you are leaving the ground with nearly 2.0 TWR you should do so only for control, aero losses will be made up by lower gravity losses) make sure it is your liquid boosters.

If you have too much thrust from your kickbacks, you can usually just use less.  If that doesn't work, things get weird (you might even have a second stage SRB.  It is roughly the extreme end of the "poor man's asparagus".  But that is an extremely rare situation).

I suspect that adjusting the thrust of SRBs is a leftover from the pre 1.0 aero-[and heat] model. Adjusting power was critical in those times, now without the souposphere it really only matters on places like Eve.  There might be plenty of advice left around for adjusting the thrust, but it is likely based on obsolete data.

[Harry Rhodan]

1 hour ago, wumpus said:

Generally speaking, you never want to limit the thrust of your SRBs.  If you are going to throttle anything (and unless you are leaving the ground with nearly 2.0 TWR you should do so only for control, aero losses will be made up by lower gravity losses) make sure it is your liquid boosters.

But sometimes you have to limit the SRBs to keep them burning long enough for your liquid engine to get high enough and to burn enough of its own fuel to carry itself.

[wumpus]

1 hour ago, Harry Rhodan said:

But sometimes you have to limit the SRBs to keep them burning long enough for your liquid engine to get high enough and to burn enough of its own fuel to carry itself.

Try it again with max thrust.  Unless the excess thrust gets in the way of your gravity turn (and it probably will), you are likely better off burning up and ditching the SRBs as soon as possible, and then coasting on the way up while burning your liquids until they it gets to TWR>1.

Consider the difference with two ordinary stages and how much more delta-v you get when burning out quickly and starting the next stage vs. starting the next stage at apoapsis.  It doesn't matter if you are burning with TWR<1, just as long as you are adding to the delta-v.

[g00bd0g]

Surprised I haven't seen it mentioned, but you do want to accelerate as fast as possible past 200m/s or so. If your initial TWR is low (<1.4 or so) you should use your LF engine, even just for a bit at the beginning to get to orbit with more DV left over.  You only want to stick to the "low isp fuels 1st" rule when you are not in a time or acceleration sensitive situation. When fighting out of gravity wells, TWR can overcome poor ISP.

[Blaarkies]

VariableThrustLimiter -by Crzyrndm. It gives a second thrust limiter to SRBs. Thrust drops linearly to whatever the second limiter was set to(or increases, depending on how you set your limiter). This mod blends seamlessly into stock and really makes SRBs useful

[Harry Rhodan]

6 hours ago, wumpus said:

Consider the difference with two ordinary stages and how much more delta-v you get when burning out quickly and starting the next stage vs. starting the next stage at apoapsis.  It doesn't matter if you are burning with TWR<1, just as long as you are adding to the delta-v.

Yes, but the difference is pretty marginal if you only reduce the thrust of the SRB to a value between 100% and 75%. Compared to the slightly better handling of the rocket I can really live with 50m/s less juice.

 

2 hours ago, Blaarkies said:

VariableThrustLimiter -by Crzyrndm. It gives a second thrust limiter to SRBs. Thrust drops linearly to whatever the second limiter was set to(or increases, depending on how you set your limiter). This mod blends seamlessly into stock and really makes SRBs useful

This sounds fun.

[Navy2k]

Keep a lookout to your TWR after Solid Booster separation. If its ok its best to lower the trust on the LF Motor rather than the solid.

[wumpus]

6 hours ago, Harry Rhodan said:

Yes, but the difference is pretty marginal if you only reduce the thrust of the SRB to a value between 100% and 75%. Compared to the slightly better handling of the rocket I can really live with 50m/s less juice.

True.  I always point out that while theoretical efficiency increases when increasing SRB thrust (to a fairly high point), handling is often a reason to not overdo it.  Note that if you are dialing back to 75% and have more than 8 SRBs, I'd often remove two and dial it back up and see if that works better (there are ways to remove one, but often take just too much time in the VAB).

[KocLobster]

On 4/5/2016 at 4:36 PM, Harry Rhodan said:

But sometimes you have to limit the SRBs to keep them burning long enough for your liquid engine to get high enough and to burn enough of its own fuel to carry itself.

I experienced this last night actually. I was right at the point where I needed to lower my SRB's thrust just a bit so my LFE had acceptable TWR once the SRBs burned out.

[wumpus]

18 hours ago, KocLobster said:

I experienced this last night actually. I was right at the point where I needed to lower my SRB's thrust just a bit so my LFE had acceptable TWR once the SRBs burned out.

It doesn't work like that.  Thrust adds linearly.

Burning your SRBs at maximum thrust means that you get your delta-v back as soon as possible.

Burning your SRBs and LFEs at the same time means that both engines are having their thrust divided by the mass of the combined rocket (F=ma).  If you dropped your SRB early and coasted, your LFE thrust will only be divided by the mass of the rocket, not the rocket+SRB.  You want to already have the delta-v of your SRB converted into velocity and coast on that, while getting the most from your LFE (even if you are steadily losing velocity overall).

If you can't get the thing onto a decent pitchover and are losing delta-v that way, then by all means reduce your SRB thrust.  Just don't reduce the thrust assuming that you are somehow saving delta-v.

[KocLobster]

@wumpus

I think you misunderstood, I was never talking about dV. I was referring to how by lowering the thrust of my SRBs so that they burn for a longer duration, it gave my liquid fuel engine more time to burn at maximum thrust, and thereby reducing the total mass of my rocket by the time that I ditched the SRBs than if I had had my SRBs burning at maximum thrust. I did this because when the SRBs burned at max thrust, I was being left with slightly too much mass for comfort when I ditched them, because the liquid fuel engine didn't have as much time to burn off fuel as when I had the SRBs set to a lower thrust ratio.

Hope this makes sense for you.

[numerobis]

With kickbacks it's often worthwhile to put fuel tanks on the SRB stage and feed it inwards with fuel pipes. That means you ditch empty tank mass with the spent SRBs.

When you talk about "acceptable" TWR, what do you mean exactly? I'm typically around 0.7 to 0.8 when kickbacks drop off. You don't need TWR > 1 at that point, you've got lots of time before you get to apoapsis.

[MaverickSawyer]

Personal preference with sidemount SRBs is to have a liftoff TWR of between 1.22 and 1.3. That way, I don't have to throttle back much, if at all, for Max Q, and my G loads rarely exceed 3Gs during SRB-assisted flight. Also, lowering core stage thrust will keep your maximum G loads (near core stage burnout) below 5Gs almost every time.

[Plusck]

3 hours ago, numerobis said:

With kickbacks it's often worthwhile to put fuel tanks on the SRB stage and feed it inwards with fuel pipes. That means you ditch empty tank mass with the spent SRBs.

Since the result looks so horrible and badly designed, I didn't expect anybody else to confess to doing this ; )

However, I'd add that it also has the advantage of ensuring the SRBs peel off nicely.