Poll - what rocket engine do you use the least / forgot about?

[GoSlash27]

56 minutes ago, Tweeker said:

The NERVA is often my choice in these stages. The engine costs more but it is cheaper to lift to orbit because the stage as a whole ends up being lighter,

Tweeker,

 In some situations, that is true, while in others it is not. For people who design with lots of DV in mind, the LV-N can be attractive. But as a practical matter, "more DV" isn't actually better once you have enough to do the job. I have found that I rarely need the kind of DV that makes LV-Ns attractive.

Best,
-Slashy

[Tweeker]

5 hours ago, GoSlash27 said:

Tweeker,

 In some situations, that is true, while in others it is not. For people who design with lots of DV in mind, the LV-N can be attractive. But as a practical matter, "more DV" isn't actually better once you have enough to do the job. I have found that I rarely need the kind of DV that makes LV-Ns attractive.

Best,
-Slashy

Slashy,

    Operating around 2,000 m/s is  on the low end of  the scale for the NERVA, but even here you can see it's benefits over the Poodle. Consider the following 2 stages. Both pushing the same 9460 kg payload.

The NERVA cost 9350 more, but the payload + transfer stage weighs 6,700 less, that's 26% less weight you have to lift to LKO.

.Lifting that extra weight will eat up the funds you saved by using the Poodle. 

    The same is true in swapping the Poodle for a Terrier. In this case you would gain ~130 m/s or about 6% more delta/v and save 910 funds. you also save 1.25 tones,  or. 5% less weight to lift to LKO.  IF your payload is a lander the terrier also serves there, meaning you can omit the descent engine from the paylod, for even more savings, in this case another 500 KG, which means more delta/v, If that is  too much extra delta/v, you can  cut the fuel in the transfer stage back to maintain the same ~2200 m/s from the example  you end up with a mass to LKO that is 12% lighter, and  costs 1,300 funds  less.

     To me it seems you can always gain something or save by swapping out the Poodle. It maybe immediately obvious such as the delta/v saving or harder to see such as the weight savings, which compound down thru the booster stages. 

    

       

[cantab]

For me, I tend to use the Poodle to push a lot more. 50 tonnes for an upper stage (0.5 TWR), and maybe up to 100 for deep space use (0.25 TWR). As far as chemicals go, for that kind of thing it's the right engine in my view. Four Terrier's are strictly inferior. The Reliant or Swivel have lower Isp and the reduced dry mass isn't enough to help. (It *used* to be, back when the Poodle was heavier). The Aerospike's an interesting alternative but some craft will miss not having gimbal. And four Nervs, well yes they probably will give you a lower mass stage but in career it starts to look really pricey and anyway "worse than the nuclear powered engine" isn't much of a criticism.

Edited February 19, 2017 by cantab

[Tweeker]

2 hours ago, cantab said:

For me, I tend to use the Poodle to push a lot more. 50 tonnes for an upper stage (0.5 TWR), and maybe up to 100 for deep space use (0.25 TWR).

I assume you mean the payload + Transfer stage weighs 50 tonnes.  How much payload? How much delta/V?

[GoSlash27]

Tweeker,

 Oh, I'm well aware. It's just that I rarely need over 2km/sec DV to do what I need to do. Taking advantage of gate orbits greatly lowers the DV cost to the destinations where funds are important. By the time I start using the Nerv in my designs, cost is no longer an issue.

Also... your comparison is not a direct one. Different t/w ratios, different DV. To make a direct comparison, you'd need to design both stages to the *same* mission requirement. Also keep in mind that I generally orbit payloads for around $1,000 per tonne, so even in this flawed comparison my launch cost would still be cheaper with the Poodle.

  As for trading down to the Terrier... I find that having a t/w below .5 can cause headaches with transfer burn timing and cosine losses, so I don't do it. I use the Terrier for lighter payloads. Just like everything else, I use the correct engine for my requirement. Very often, that winds up being the Poodle.

Best,
-Slashy

 

Edited February 19, 2017 by GoSlash27

[Aegolius13]

Puff would be my least-used, followed closely by the Spider. The Vector gets an honorable mention for weirdness - too good in Sandbox, too expensive in Career.  It makes sense for Eve ascenders, but that's about it. 

I would echo previous comments that the TwinBoar is probably the most OP engine in Career mode.  The game just gives you very little reason to use anything else for a load bigger than a Skipper and a few SRBs can handle .  If there's a reason not to use the TwinBoar, it's because it's too good and gets monotonous.  

[NCommander]

I rarely if ever use the Thud; its a relatively poor engine all around and even when it's the "best" radial mounted engine, I've never needed it. I tried it on a recent lander and found I was better off just building around terriers. 

Poodles get used extensively in my space program for upper stages and sometimes for transfer orbits depending on what I'm sending where. Ants and Spiders dominate my early probes, and I tend to use them right up until I get the puff and switch to monoprop only probes which are lighter at the cost of lower dV. Even then, I occasionally use them for something that needs to orbit separately; a recent Duna mission heading for an equatorial orbit dropped a SCANsat mapper powered by an Ant into a high polar orbit. Aerospikes get used on early space planes.

Dawns frequently get used when I send survey sats to Jool when playing with life support as I can send a few probes and have them orbit multiple bodies without issue.

What I usually don't use are the mid-range jets. I usually make due with the Junos for a long time, and I don't think I've ever used the 1.25M engine in career mode, or the panther.

[Whisky Tango Foxtrot]

The Panther's great for those "collect science above x altitude" contracts. On dry mode it's efficient enough to easily get where you need to go, at which time you can switch to wet mode to get to the necessary altitude. It doesn't have the ISP of a Wheesley in dry mode or the thrust of a Whiplash in wet mode, but it gets close enough to both of them to be a good all-around engine. It can be used for small SSTOs in early career games, too.

[Tweeker]

5 hours ago, GoSlash27 said:

Tweeker,

 Also... your comparison is not a direct one. Different t/w ratios, different DV. To make a direct comparison, you'd need to design both stages to the *same* mission requirement. Also keep in mind that I generally orbit payloads for around $1,000 per tonne, so even in this flawed comparison my launch cost would still be cheaper with the Poodle.

   Slashy,

           You are not going to get the same TWRs from these 2 engines, The only way that would happen is to  strap 3  NERVAS under this load, which *would* be a flawed comparison. The Delta/v are different because . the Poodle's delta/v moves in a very granular way. You can either have 2231 or 1686 m/s with this load.  1686 m/s is farther from the ~2000 you spec'd so  2231 it is.

        The NERVA can have either 2055 m/s or 2865 using only MK 1 fuel tanks.  I could have gone with 2055m/s and de-fueled Poodle to get a similar delta/v, But that wouldn't be a realistic. You could say that the poodles extra fuel is spent circularizing, or you could  adjust the tanks on the NERVA, which is what I elected to do. The closest you can get is 2070 m/s  with 2 MK1 + 2 MK 0. this fits cleanly under the 2.5M load, but does not seem like a configuration that someone would naturally use. 1 + 8 gives you  2068m/s, too low. 1 + 9 gives you  2170 which is close, but putting 3 sets of 3 tanks on seems like someone might do it, but it you need to do it in 3 steps, and it looks a little odd .  1+10 ..gives you 2270, but there is no 5X symmetry, and putting on a row of  6 and a row of 4 doesn't really seem like a natural choice either. So I went with 1+12 which fits under the 2.5m load, and feels like a realistic choice.  

          One thing I did not do was change out the 2.5 m de-coupler for a 1.25m one. This would be a natural choice but  I didn't want to get side track into how 2.5 m are heavier, or be accused of stacking the deck for the NERVA.

 

[GoSlash27]

16 minutes ago, Tweeker said:

You are not going to get the same TWRs from these 2 engines, (additional stuff)

Tweeker,

 Sorry, but that's how stages are designed; "Accelerate x payload at a minimum t/w of y for a total of z m/sec DV". Each job description is unique, and the ideal engine varies with the application.

For 2 km/sec DV, 9.46t payload, and minimum t/w of .5, they stack up like this:

Nerv: 21.3t, $21,650
Terrier: 22.2t, $3,930
Poodle: 23.5t, $3,700

The Nerv starts becoming competitive at higher DV, and the Poodle becomes dominant with larger payloads.

Best,
-Slashy

[Tweeker]

 

12 minutes ago, GoSlash27 said:

Tweeker,

 Sorry, but that's how stages are designed; "Accelerate x payload at a minimum t/w of y for a total of z m/sec DV". Each job description is unique, and the ideal engine varies with the application.

For 2 km/sec DV, 9.46t payload, and minimum t/w of .5, they stack up like this:

Nerv: 21.3t, $21,650
Terrier: 22.2t, $3,930
Poodle: 23.5t, $3,700

The Nerv starts becoming competitive at higher DV, and the Poodle becomes dominant with larger payloads.

Best,
-Slashy

Slashy,

    I really don't see any reason to spec a TWR of .5 for a transfer stage, except to knee-cap the NERVA or Terrier.  By insisting on this requirement you force an extra 3 tonnes on the NERVA powered stage, and an extra 10,000$. The only time your going to have to worry about losing delta/v burning off angle is leaving LKO, and that is easily solved by making your departure in 2 or 3 burns. .  

[cantab]

10 hours ago, Tweeker said:

I assume you mean the payload + Transfer stage weighs 50 tonnes.  How much payload? How much delta/V?

Yeah, total mass. Ratio between fuel and payload has varied and therefore delta-V too. For upper stage use in stock probably about 1000-1500 m/s, in RSS I might go up to 3500. I'll acknowledge that towards the upper end I might consider a Nerv cluster instead (once I unlock it).

Unlike a transfer stage, an upper stage really will give problems if it's much below 0.5 TWR. Lower can be done but gets extremely sensitive to the ascent profile.

Edited February 19, 2017 by cantab

[GoSlash27]

12 hours ago, GoSlash27 said:

I find that having a t/w below .5 can cause headaches with transfer burn timing and cosine losses, so I don't do it.

Tweeker, ^ see above.

 If you think .5G is asking too much, then you can always pick a different limit. The important part is to have one and compare engines that are actually doing the same job, not different ones.

Best,
-Slashy

 

[Cunjo Carl]

I like to keep my transfer stage TWR between 1-1.5, but hey that's just me!

One unsung super power of the aerospike is to attach radially to tanks. This comes in super handy sometimes (landers), and is often the reason I choose it over alternatives in v1.2.

[Tweeker]

20 hours ago, GoSlash27 said:

Tweeker, ^ see above.

 If you think .5G is asking too much, then you can always pick a different limit. The important part is to have one and compare engines that are actually doing the same job, not different ones.

Best,
-Slashy

 

Slashy,

For me one of the Poodle's big problems is it feels stuck in the middle between the 909 and the NERVA. It is not as tunable as the 909, because of the way it relates mass wise to the 2.5m fuel tanks. and capsules. this means it's delta/v and TWR moves in a very notchy manner.  And it is not as efficient as the NERVA. For me Either one is a better choice..  

[GoSlash27]

1 hour ago, Tweeker said:

Slashy,

For me one of the Poodle's big problems is it feels stuck in the middle between the 909 and the NERVA. It is not as tunable as the 909, because of the way it relates mass wise to the 2.5m fuel tanks. and capsules. this means it's delta/v and TWR moves in a very notchy manner.  And it is not as efficient as the NERVA. For me Either one is a better choice..  

Tweeker,

 I get that, Don't agree with all of it, but I get it, *But* since I don't design stages the same way you do, I find the Poodle incredibly useful. It's one of my most commonly used engines. Your mileage may (and clearly does) vary.

 Best,
-Slashy

 

[Chemp]

A very interesting  discussion to follow, indeed. I usually design my stages on the fly and rarely use the same one more than 3/4 times. From reading here I guess at least Slashy has made a neat spreadsheet. I really need to take the time to make one of those...

That said, there are a few engines I rarely use, the ant or spider for instance, because I tend to overengineer my craft.

 

[cantab]

For me a gripe with upper stage engines in general is they have too much performance for their node size. I often find myself putting Sparks under 1.25 m tanks and Terriers under 2.5 m ones, but in stock aero that means either adding the extra mass of an interstage fairing which is sometimes noticeable, or having horrible drag during launch. If only the game could handle engine shrouds and decouplers better. (At least when I use FAR I can simply clip the decoupler around the engine.)

[Tweeker]

2 hours ago, cantab said:

For me a gripe with upper stage engines in general is they have too much performance for their node size. I often find myself putting Sparks under 1.25 m tanks and Terriers under 2.5 m ones,

    It's not just the upper stage engines that suffer from this. the Rhino, Mammoth, and  Vector are all way OP vs their real life counterparts. The Mammoth for example should be a high efficiency  second stage/sustainer ~1700KN with mid/high ISP.

    SRBs on the other hand are greatly under powered. The kickback for example should be a heavy lift brute, 2.5M with ~2500 KN

  

[cantab]

Yeah. I did actually hypothesise why the Mammoth is so overpowered. The real Shuttle and SLS have hydrolox core stages which are large but relatively light, and solid boosters which have much denser fuel. More than half the launch weight of the Shuttle is in the SRBs! Hence the SRBs having high thrust and the orbiter, or for SLS the core, having relatively low thrust. But KSP only has one fuel which is more like kerolox in density, which means if you make a rocket that *looks* like an SLS then after you correct for 'Kerbal scale' the rocket will be heavier and so need more powerful engines. Though still, I reckon the Mammoth would be happy around 2000kN, if the Kickback and Twin Boar were improved to compensate.

[Tweeker]

  Really its more than that, the parts are just plain mis-sized. The twin boar for example is suposed to be a "Pyrios" liquid booster. Which is supposed to replace the SRBs later in the design process. However, the kickback isn't even close to it in size or thrust. 

     Each booster makes about 1.5-2x the thrust of the core stage.  The kickback to mammoth thrust ratio is 1:6 it should be ~ 1.75:1 

    I haven't messed around with re-specing the mammoth, but I have re-balanced the vector and kickback to make a shuttle with more real world performance. The kickback back ended up being 2,500 Kn and the vector 425 Kn. 

[foamyesque]

14 hours ago, Tweeker said:

    It's not just the upper stage engines that suffer from this. the Rhino, Mammoth, and  Vector are all way OP vs their real life counterparts. The Mammoth for example should be a high efficiency  second stage/sustainer ~1700KN with mid/high ISP.

    SRBs on the other hand are greatly under powered. The kickback for example should be a heavy lift brute, 2.5M with ~2500 KN

  

The Mammoth is meant to be an analogue of the Apollo 1st stage, in role if not in detail. The weedy SRBs we have are another problem altogether.

[Kerbart]

26 minutes ago, foamyesque said:

The Mammoth is meant to be an analogue of the Apollo 1st stage, in role if not in detail. The weedy SRBs we have are another problem altogether.

Actually the first stage engine of the SLS. Look at the shape! But your concept remains unchanged in that respect; it *is* the first stage engine of a large launch system.

[Tweeker]

2 hours ago, Kerbart said:

Actually the first stage engine of the SLS. Look at the shape! But your concept remains unchanged in that respect; it *is* the first stage engine of a large launch system.

    That is true, kind of. It is more like a 1.5 stage. it isn't the primary lifter however,.-- the SRBs are. Saturn had 35,000 KN of thrust in the first stage. The SLS should have 39,500 KN, but only ~7500 KN comes from the core stage that the mammoth is based on. To put it another way it has only about as much thrust as 1 of the F-1 engines on the Saturn 1st stage.   .

Edited February 21, 2017 by Tweeker

[TopHeavy11]

In all fairness, every engine is good for something. Ant/Spider engines are a good OMS for light satellites, the Dart is a great LFO engine for spaceplanes, the NERVA for Large Interplanetary Crafts (LIPCs), and the Poodle for high efficiency for short-range missions. Really, even the Dawn is good for something (that something is a thing I don't know about.) No engine is really "useless," because the Devs put them there for a reason. There is one forumer, @Raptor9, who had made great use of all engines for various purposes. DISCLAIMER: This is not sponsored by @Raptor9. He just makes a ton of cool stuff. Here is the link to his craft thread.