Engine rebalance - improved figures included + reasoning

[thereaverofdarkness]

I've suggested this before and I'll keep suggesting it until some of these unbalanced engines are either balanced or given an excuse for why they are out of balance. Even just moving them up or down the tech tree would make a positive difference, though if that's going to happen, it would be nice to see lower and higher tech versions of the same basic engines. However for the stock space program as early in development as the game is, I think it makes sense for all engines to be on par with each other, so that they all have uses in sandbox and after unlocking the whole tech tree.

Quick reference: section 1 is about lander engines, section 2 about lifter engines, and section 3 about radial engines.

Section 1

It can be difficult to find base values for engines, a "par" if you will that the other should measure up to, when there are so few engines to choose from. But I am seeing a few patterns in here. For starters, the "lander engines", the 390 Isp LV-909 and Rockomax Poodle have about 10:1 TWR on Kerbin. They aren't very good as lifter engines to get to Kerbin orbit because they lose Isp more strongly in atmosphere and don't have the kind of TWR you want to have, but they can work as lifters if that's how you want to use them. The Poodle has inferior attributes to the LV-909 because it has less TWR (~9:1) for the same Isp. The 909 is already a bit unpopular because its substantially lower TWR isn't fully made up for by its only slightly better vacuum Isp (390 in comparison to 370 of the LV-T30/45). A lot of people choose it for being smaller and lighter than the T30/45 when they aren't really needing to be picky other than choosing an engine that doesn't take up a lot of space or mass. Speaking of space and mass, the massive space-hog of the Poodle is quite a bit taller than the large landing legs and it also has a poor TWR. Due to this, it is unpopular and mostly used by people who just want to slap a single engine on rather than deal with trying to mount multiple smaller engines (which is more powerful and efficient even with the added coupler mass) or they want to mount something 2.5m underneath and don't want to deal with the poor attachment of a flipped coupler underneath multiple engines. But I don't think that's any excuse for the Poodle to have such pathetic attributes. I think the Poodle should be both more powerful (to match its size) and mroe efficient (to match the LV-909). So we know that the LV-909 certainly has the TWR necessary to make a popular lander engine for airless landings (usually low-G), and really the only thing holding it back from being a loved engine is that it doesn't have enough of an Isp margin to make its efficiency stand out and make up for the lost TWR. But before we knock it up to 400+, I think I can provide some reasoning why the vacuum Isp of lander engines could be decreased without actually hurting their ability to lift off. Before that, however, I will cut the threadnought with my proposed adjusted attributes for the "lander engines". I haven't changed the LV-909--it'll shine along with my other suggestions.

Lander Engines................................................................................

LV-909:

Mass: 0.5 t

Thrust: 50 kN

Isp: 300 - 390 s

TMR/10: 10:1

Rockomax "Poodle":

Mass: 3.0 t

Thrust: 300 kN

Isp: 280 - 390 s

TMR/10: 10:1

-----------------------------------------------------------------------------------

Section 2

In this section I will be discussing lifter engines. The lifter engines are the Mainsail, Skipper, LV-T30/LV-T45, and the 48-7S. These engines have a high TWR which is important for getting rocks out of Kerbin's intense gravity well and out into space. They are reasonably efficient as well, for they have to fight against Kerbin's thick atmosphere, but are not as efficient as a less powerful lander engine. Here is a comparison of the approximate Kerbin TWR and Isp of these engines: (I'm using TMR/10 instead of TWR for easier math)

Engine: TMR/10 // Isp air - vacuum

Mainsail: 25.0:1 // 280 - 330 s

Skipper: 16.3:1 // 300 - 350 s

LV-T30: 17.2:1 // 320 - 370 s

LV-T45: 13.3:1 // 320 - 370 s

48-7S: 30:1 // 300 - 350 s

There is a bit of a trend here toward the more powerful engines being less efficient, though it is imperfect. The Mainsail has a better TWR than the Skipper but is less efficient. Likewise, the 48-7S has more TWR than LV-T30/45 but is less efficient. But the Mainsail and the 48-7S clearly shine as the top TWR lifters yet the 48-7S is not only the greater of the two (in terms of TWR) but has better efficiency too. The Skipper has the same middle efficiency as the 48-7S but lower TWR like the more efficient LV-T30. The LV-T45 doesn't have much more TWR than a lander engine. We know that it's pretty easy to get decent payloads into orbit with the LV-T45, likely due in part to its high efficiency. It wasn't even difficult back when its thrust was lower.

Now since the LV-T30 is the only engine here without thrust vectoring, we want to base its attributes off the finalized LV-T45 which we want to balance first. Right off the bat, I think it could be made into more of a lifter engine and leave the lander engine functionality to the 909 by reducing its Isp to the middle value: 300-350s of the Skipper and 48-7S. Then we can increase its TWR to something more like the LV-T30 or Skipper. It will certainly be a great engine with this rebalancing, and will be the mainstay of 1.25 meter lifter engines with thrust vectoring (if not the only one). We could accomplish this either by decreasing its mass, increasing its thrust, or a composite of both. Here's a figure I like: 1.25 tons and 200 kN thrust. That brings its TMR/10 to 16.0, right next to the Skipper. Now the next problem I have with the Skipper is that it is too weak for how large it is. It takes up 4x the ground space of the LV-T45 and is taller, it should be over 4x the mass and thrust. Otherwise you get a smaller and more powerful engine set out of a quad coupler and 4 LV-T45s. But here's another thought: I think the Skipper is supposed to be a more efficient middle-stage engine, combining decent thrust with decent efficiency. After all, it says that in the description. Its bell is also more narrow than most lifter engines. So we could make it into a lower TWR higher Isp engine. If we keep its current thrust of 650kN and shoot for a TWR more like the existing LV-T45, we can adjust its mass to 5 tons and get a TMR/10 of 13.0. Then we can give it the 320-370 Isp of the T30/45, or even put it midway in the trend of lander engines losing efficiency in air faster than lifter engines, and give it 300-370 Isp. Now the Skipper is both powerful and efficient! With this setup, it would actually be rather popular as a mid-stage engine, either for final boost to orbit or for launching out of orbit on an Hohmann transfer intercept to another planet, taking full advantage of the Oberth effect while not losing much to lifter engine inefficiency.

Now on to the really powerful lifters, the Mainsail and 48-7S. I haven't forgotten about the LV-T30 or the Isp thing I was going to talk about earlier. I'm going to bring up the LV-T30 in a moment in fact. So the Mainsail has both incredible power and incredible TWR. Its thrust is nearly 8x as great as an LV-T45, which matches its greater size. It's not nearly 8x as heavy though, lending to its really high TWR. Part of me feels its mass should be increased just because of its incredible size. I've figured in the past that 9 tons would make more sense than the 6 tons it is now (1.25x8=10). If we didn't adjust its thrust, its TMR/10 would become 16.67, about where the Skipper is. So lets increase its thrust by 50% as well, to 2250 kN. That's a big and powerful engine! It's quite a bit less efficient than the Skipper to balance it out. It's too powerful you say? I disagree, and I think the reason you might say that is due to the lack of a high-power low-efficiency engine in the 1.25m range. Which brings me to the LV-T30. Also, the most powerful engines should lack thrust vectoring I think, since fron an engineering standpoint having positive control of the engine bell means you didn't build it as big as you could have. So lets take thrust vectoring off the Mainsail and make the LV-T30 into a smaller version of it. If we keep the 1.25 ton mass of the T30 and give it a TWR of about 25 like the Mainsail, we can set its thrust to 325, giving it a TMR/10 of 26.0.

Now the 48-7S can be fixed by nerfing its Isp to the 280-330 of the Mainsail and our new LV-T30. It still has a significantly higher TWR than the mainsail or T30 but it's more reasonable. We could adjust it further but I wouldn't say it's out of whack anymore, at least not in relation to the other heavy lifters. Then again, the Isp difference between 280-330 and 300-350 is a difference of about 6-7% while the TWR difference between our heavy lifters and the medium lifter LV-T45 is over 50%. So lets go with something more reasonable, like a TWR of 20. I think this is better than decreasing their efficiency because they need some efficiency to get out of Kerbin's atmosphere, also nobody likes an engine that basically dumps fuel out the back rather than burning the stuff. Without changing the mass of the rockets, we can set the Mainsail's thrust to 1800 and the LV-T30's thrust to 250. So they are both still more powerful than the old versions, but with either reduced TWR or Isp. The Mainsail should lift the same size rockets as before, if not bigger, so I don't think anyone will complain that its mass makes more sense for its size now. And I bet people would love the new medium-mount heavy-lifter as well as the greater distinction between the T30 and the T45. Finally, to bring the overpowerful Rockomax 48-7S in line with the changes (currently 30kN and 0.1 tons), we can either cut its thrust to 20kN or raise its mass to 0.15 tons. I'm leaning toward the second one because its a 0.625m rocket. The size of its base is a quarter of that of the 1.25m rockets, and it is a lot shorter. If we put its attributes in-line with being a heavy lifter 1/8th the size of the LV-T30, that means it should have about 0.15 tons mass and 30kN of thrust. And then it should be taller than it is and have a bigger engine bell.

Now before I present my finalized lifter engine figures, I want to discuss changing the Isp a bit. Above, I already put the Isp values lower for our more powerful lifter engines. Only the Skipper has 370 Isp in vacuum, but even 350 isn't that far behind the top 390 of the lander engines. My idea to make the lander engines shine even more at their job is to lower the top efficiency of lifter engines while leaving their Kerbin launchpad efficiency the same or possibly increasing it. Instead of 280-330, we could have 290-320, which will leave them about as effective as lifter engines as before, but will increase the vacuum Isp margin between them and the lander engines. It will also increase their efficiency in Eve launches, as they will have a greater resistance to the high air pressure. They won't take away the job of aerospike engines in any way, but can add other options to make degining Eve launch vehicles a bit more fluid. Now the greater launchpad efficiency doesn't actually make these engines more efficient at launch, because that air pressure drops off quickly and engine efficiency is almost at vacuum level by 10km while your rocket is still getting slowed a lot by the air. But having the efficiency start a bit higher should make up for it going a bit lower further into the launch.

Now, without further ado, lifter engine final attributes:

Lifter Engines................................................................................

Rockomax 48-7S:

Mass: 0.15 t

Thrust: 30 kN

Isp: 290 - 320 s

TMR/10: 20:1

LV-T45:

Mass: 1.25 t

Thrust: 200 kN

Isp: 300 - 350 s

TMR/10: 16:1

LV-T30:

Mass: 1.25 t

Thrust: 250 kN

Isp: 290 - 320 s

TMR/10: 20:1

Rockomax Skipper:

Mass: 5.0 t

Thrust: 650 kN

Isp: 300 - 370 s

TMR/10: 13:1

Rockomax Mainsail:

Mass: 9.0 t

Thrust: 1800 kN

Isp: 290 - 320 s

TMR/10: 20:1

And we could even turn that pathetic ant engine into a tiny yet effective lander engine, without even increasing its thrust:

LV-1:

Mass: 0.015 t

Thrust: 1.5 kN

Isp: 320 - 390 s

TMR/10: 10:1

-----------------------------------------------------------------------------------

Section 3

In this section I will be discussing radial engines. We have the Rockomax Mark 55, the Rockomax 24-77, and the LV-1R. All of these are inefficient engines and I think they're supposed to be. It's so easy to stick engines on the side that it should come at a cost. Radial engines are a band-aid fix to a rocket that doesn't have enough thrust with its in-line engines, or they allow you to build rockets in a huge variety of shapes. Well the efficiency of the engines seems fine, but the Mark 55 isn't all that powerful. Clustering lots of them around a 2.5m rocket doesn't add that much thrust. You can fit 8-9 of them around a 2.5 meter rocket base, and at 120 thrust each, together all of them add 960-1080 kN of thrust to your rocket. It seems weak to me for how much space they take up. Given their poor TWR and fuel inefficiency, I find that on several designs that need a bit more thrust, adding those wont help because they reduce the efficiency too much for how little thrust they add. Maybe they are too weak, maybe they are too inefficient. Well let us address that.

Its current TMR/10 is 13.33, similar to the currently existing LV-T45 and my Skipper proposal above. Its Isp is 290-320, same as my proposed Isp for heavy lifters. It is clearly lacking in TWR but perhaps that's reasonable for a radial engine. Lets increase its power to make it actually lift heavy things. If we increase its mass and thrust by 50% to 1.35t and 180kN, you can get as much as 1440-1620 thrust out of these. So if you jam-pack your large rocket base full of these, you can get around as much power out of them as a Mainsail, albeit with less mass efficiency. They will have a combined mass of 10.8-12.15 tons. Seems fair to me.

The 24-77 has a TMR/10 of 22.22, which is even higher than our heavy lifters. It's not very fuel efficient, with an Isp of 250-300. I use these a lot because they are small and easy to mount. I wish they had better efficiency and I wouldn't mind giving up some of that thrust or making them have more mass. But I think they're balanced more or less, considering benefit/deficit ratio.

The LV-1R is absolutely pathetic. It has a weak TMR/10 of 5:1 and a terrible efficiency of 220-290 Isp. It's pretty much just dumping unburnt fuel out, and not very quickly at that. It's basically a keg tap. Underneath my lifter engine proposals I put a proposal for the LV-1 as a tiny lander engine. Perhaps the LV-1R could be altered the same, as at current its attributes match the LV-1. I would then go a step further and increase its mass to make up for the increased materials needed in installing it on the side of a rocket.

So here's my radial engine proposals--the 24-77 hasn't been changed:

Radial Engines................................................................................

LV-1R:

Mass: 0.02 t

Thrust: 1.5 kN

Isp: 320 - 390 s

TMR/10: 7.5:1

Rockomax 24-77:

Mass: 0.09 t

Thrust: 20 kN

Isp: 250 - 300 s

TMR/10: 22.2:1

Rockomax Mark 55:

Mass: 1.35 t

Thrust: 180 kN

Isp: 290 - 320 s

TMR/10: 13.3:1

So there's my proposed figures, and why I think these changes make sense. I'd love to hear everyone's responses to my suggestions, any adjustments you'd make, or arguments against anything I have said here. And please, Squad, consider rebalancing the rocket engines. It's annoying that some are clearly better than others.

[nholzric]

The 909 is already a bit unpopular because its substantially lower TWR isn't fully made up for by its only slightly better vacuum Isp (390 in comparison to 370 of the LV-T30/45).

Probably interesting! If you included some graphs or charts it would make it much easier to read/skim. Especially when you make statements like "isn't fully made up for" - because I'm interested in what you're trying to say - I want you to show me how you translate the utility of TWR to ISP... preferably with a graph.

[Strikerklm96]

I agree with almost everything you had to say, and I hope these changes, or something similar are implemented.

However,

I always saw the 48-7S as a Lander Engine, not a Lifter Engine. Look at the ISP and shape of it, i'm 99% sure it was supposed to be the small Lander, with the 909 being the medium, and the poodle the large.

Also, for any situation where an object (no matter what it is) has a larger cousin, the larger cousin should always be every so slightly better, so that people are encouraged to build somewhat realistically. For instance, the fuel canisters get more and more sturdy as they get bigger(I always thought they should have a little more fuel). My point is, the Mainsail should be very slightly better than the LV-T30, probably its TMR.

And for those who say "The incentive is that its less parts so it puts less strain on the computer.", thats an incredibly stupid game incentive. Do FPS games encourage one weapon over the other because one has laggy textures? Do board games encourage one action over the other because action 1 will take 5 minutes while action 2 will take 5 hours and waste your time? Game balance shouldn't have anything to do with anything outside the game world, like your CPU.

Also, one thing i've never been sure of is whether the BACC SFB is better than 2 or 3 staged RT-10 SFB's, maybe they are already balanced.

Edited February 6, 2014 by Strikerklm96