Adjustable SRB thrust curves

[Hannu]

Real solid boosters can have complex thrust vs. time curves depending of mission profile. For example Shuttle's SRB:s behaved like this:

http://en.wikipedia.org/wiki/Space_Shuttle_Solid_Rocket_Booster#mediaviewer/File:Srbthrust2.svg

They have reduced thrust around max Q (maximum aerodynamic stress), then thrust is increased until it decreases again when ship gets lighter to prevent too high acceleration and stresses.

I think that simplified linear decrease would fit well in KSP. You could adjust separately initial thrust and end thrust relative to initial value and thrust would decrease linearly over burn time. End value could be for example 100% - 70 % of initial value. If default was 100 % they who do not like this could just left it untouched and get what they expect.

Edited January 14, 2015 by Hannu
Image did not work.

[chaos_forge]

Honestly I think just giving all SRB's a constant TWR thrust curve would be good enough

[CaptainKipard]

What's an easy way to consistently work out realistic curves?

[Capt Snuggler]

this would be realistic and I think it would add an extra layer of depth for those who want it.

SRBs use many core shapes:

if you ever try making sugar rockets (with caution) you can get wildly differing results from the different burn rates.

[CaptainKipard]

Oh cool! It's like the surface area determines the reaction speed?

[vexx32]

That's exactly how it goes, aye. More surface area = more fuel burnt = more thrust at that particular instant. I've yet to see a design that can continuously provide low thrust, though; as far as I can tell, it seems that sooner or later, a high thrust point must exist... or something. Number 2 in the image Snuggler posted defies that by spreading the thrust out and plateauing, but it's still not 'low' thrust.

[DBT85]

Am I right in recalling the Shuttle SRBs using the first variant there? Also iirc it burnt from the top down?

[Bill Phil]

I support this!

[CaptainKipard]

I've yet to see a design that can continuously provide low thrust, though

How about a thin spiral?

[basic.syntax]

This is a great idea, something deep in physics, that could be exposed to players with a simple drawing app window in KSP:

Let us DRAW our own SRB in profile, (an SRB tweakable option) like the image shown by Capt Snuggler, and then observe the burn-rate results of our various doodles, patterns and shapes.

[vexx32]

How about a thin spiral?

Hmm. All exposed surface of the spiral would burn simultaneously, though. If I had to guess, I'd say that would burn with a very high thrust, burning almost all the fuel in the first handful of seconds.

[CaptainKipard]

What I had in mind was the fuel spiral being separated by metal (or whatever) an all sides, except the thin edge of it. I don't know if that's realistic, I'm just using my imagination.

[vexx32]

Oh, I see what you mean. Well, it's one way to go, for sure. The only other way that makes much sense would be simply to burn from bottom to top (instead of having all vertical sections burn simultaneously). However, that would cause the SRB's CoM to shift as it burns, which is less than desirable here and there.

[Wallygator]

http://forum.kerbalspaceprogram.com/threads/67574-Tweakable-SRBs-via-selection-of-fuel-core-mandrel-profile

I remember bringing this up last year. So naturally, I agree with the suggestion. ;-)

[prophet_01]

Has been suggested and it's still an idea I like a lot. If it gets implemented together with wider boosters I would be rly happy. And I would most likely scream like a little girl if we get stackable srb's aswell. How about a "adding moar boosters!" for the update's name?

Btw, why would you want srb's with continuous thrust? Since you burn fuel over time the twr would still increase. Did I missunderstod you? I would expect that profile with the cross in the midle to be very useful, as it should keep the twr at a more or less constant level

[vexx32]

The only advantage to wider boosters is increased burn time, assuming similar grain profiles. To make higher-thrust SRBs, typically they are simply extended in length.

[Hannu]

That's exactly how it goes, aye. More surface area = more fuel burnt = more thrust at that particular instant. I've yet to see a design that can continuously provide low thrust, though; as far as I can tell, it seems that sooner or later, a high thrust point must exist... or something. Number 2 in the image Snuggler posted defies that by spreading the thrust out and plateauing, but it's still not 'low' thrust.

Overall thrust can be adjusted by changing burn rate of the material. But I think that it eats ISP, which is low in any case. There should not be much use for SRB which fires several minutes at low thrust (at least in real spaceflights).

The only advantage to wider boosters is increased burn time, assuming similar grain profiles. To make higher-thrust SRBs, typically they are simply extended in length.

Disadvantage is that whole SRB is one large burn chamber. Whole structure must handle the chamber pressure. It is easier and more efficient to make small chamber with high pressure than large diameter chamber with low pressure. Therefore real SRBs are quite thin compared to liquid fuel tanks and as you said, more thrust and dv is achieved by lengthening of SRB instead of widening it.

[prophet_01]

So change the grid. The current boosters have differen thrust levels aswell, so it's not rly a good reason against them. The current sizes don't work that great with large lifters unless you spam them and attach boosters to boosters... not only does that look strange it's also not aerodynamic, one large booster causes less drag. At least that's how it's supposed to work, stock aero is something else atm

And it also helps with designing a rocket. Maybe it's just me, but I'm rly tired of adjusting the thrust of up to 4 sets of boosters over and over again just to get the twr about right. With more tweakables for boosters you have to set even more stuff or copy them. It's just not as comfortable as it could be. Small boosters also don't fit bigger payloads that good, again unless you stack em and use like 4 as a center rocket, which limits radial attachements :/

you are right they, lose some efficiency. But at some point there is a limit of high you csn build with thin boosters snd a bigger rocket should help here

On another note, grid profiles would rly shine with stackable segments I think, but either way they can be used for many different tasks and would rly help srb's in late game. I think they could use some more flexibilty. I personally only use them on small and medium launchers. If I have to use more than 8 at the first stage, I usually just use a skipper instead

Edited January 15, 2015 by prophet_01

[lajoswinkler]

This is a great idea, something deep in physics, that could be exposed to players with a simple drawing app window in KSP:

Let us DRAW our own SRB in profile, (an SRB tweakable option) like the image shown by Capt Snuggler, and then observe the burn-rate results of our various doodles, patterns and shapes.

Although this is interesting, it's for a rocket engine simulator (computing speed resource consuming), not something for a game where you want to put accent on universal parts which are special (policy of Squad, it seems).

[PlonioFludrasco]

Although this is interesting, it's for a rocket engine simulator (computing speed resource consuming), not something for a game where you want to put accent on universal parts which are special (policy of Squad, it seems).

Yeah, not something that has to be included in stock KSP. But it's a nice idea for a mod, I don't know if some already includes that.

[Hannu]

Although this is interesting, it's for a rocket engine simulator (computing speed resource consuming), not something for a game where you want to put accent on universal parts which are special (policy of Squad, it seems).

I think that it would be possible to get interesting results with very simple assumptions. Most simple would be that fire consumes a material at a constant number of millimeters per second. It would be easy to calculate how the shape of the propulsion mass would evolve during burn. It could be precalculated before ignition in couple of milliseconds. For example one value per second and linear interpolation between them would take couple of kilobytes and some tens of clock cycles per physics time step.

But I guessed too, that Squad is not interested in any realistic physical models. Therefore I asked linearly decreasing model. It would be the most practical to keep acceleration in limits when ship loses its mass and it takes less than an hour to implement game and does not need any mathematical or technical interest.

[lajoswinkler]

I think that it would be possible to get interesting results with very simple assumptions. Most simple would be that fire consumes a material at a constant number of millimeters per second. It would be easy to calculate how the shape of the propulsion mass would evolve during burn. It could be precalculated before ignition in couple of milliseconds. For example one value per second and linear interpolation between them would take couple of kilobytes and some tens of clock cycles per physics time step.

But I guessed too, that Squad is not interested in any realistic physical models. Therefore I asked linearly decreasing model. It would be the most practical to keep acceleration in limits when ship loses its mass and it takes less than an hour to implement game and does not need any mathematical or technical interest.

What about a compromise? Instead of designing your own solid fuel geometry, you choose between few examples? That would be a great mod.

[asmi]

RealFuels mod supports SRB thrust curves for quite some time. It doesn't have GUI to edit them thou - NathanKell told me a while ago that it's in works, but it's not there yet. In the mean time you need to use config files to edit it (Bob Fitch provides a good guide on how to do it in one of his videos).