Jump to content

Erik3003

Members
 View Profile  See their activity

  • Posts

    10

  • Joined

  • Last visited

 Content Type 

Posts posted by Erik3003

    [WIP] Nert's Dev Thread - Current: such nuke, wow

    in KSP1 Mod Development

    Posted

    4 hours ago, Nertea said:

    It's really important to note the specific impulse too. If you only balance on TWR you negate the fact that a number of these engines have 20-30s better impulse than their LF counterparts.

    I didn't take ISP into account, because I was balancing the methane engines relative to each other, since they all have an almost equal ISP of roughly 330 seconds at sea level. That is just 12% more than the (stock) Vector, though that is a bad point of comparison, so lets look at more reasonable average LF/Ox engine ISP of 280, making it a good 18% more. Now lets look at LH2/Ox launcher engines for a point of comparison. The ISP at sea level is around 370 seconds here, which is roughly 32% more than LF/Ox, so quite significant.

    4 hours ago, Nertea said:

    Ah ok see that I need to disagree with. Methalox mass ratio is currently set at 7.5, so very close to LF/O spaceplane tanks at 7.8 and not that much worse than LF/O rocket tanks at 9. Density and number of fuel tanks aren't relevant at this. I would also suggest that boiloff be ignored for launch engines which you seem to be aiming at here - on the scale of a launch, or even on the scale of a mission to the Mun/Minmus, boiloff is very small.

    Obviously, the fuel types aren't really directly comparable, because of differing densities and boiloff. But since boiloff isn't that significant for a launcher, we can take a look at comparing their densities. Looking at your figures and at the code, LCH4/Ox is ~83% as dense as LF/Ox, LH2/Ox is just ~0.45 as dense as LF/Ox. This means LH2/Ox requires ~124% more fuel tanks than LF/Ox and LCH4/Ox requires just 20% more for the same dV, in a theoretical perfect environment at equal ISP and infinite thrust that is. (This is ofc reliant on my logic, math and physics being right, as well as no stupid errors sneaking in, so no guarantees, I'm certainly not an expert at all on any of these things)

    So, more fuel tanks mean more dry mass! Well, no, upon trying to calculate the densities with ingame weights, I stumbled upon some inconsistencies with the data in the code and upon further inspection I discovered: The dry mass of tanks change when the fuel type is different, so that the same weight of fuel is accompanied by the same dry tank mass for any fuel.

    Well, this means the performance across different fuel types is actually quite comparable and that you are right and my balance has been build on false assumptions. The dV increase through higher ISP is actually "free", and not as I thought significantly reduced by the added dry mass, which makes the Iguanodon obviously a bad choice as a new balance baseline. The Vector in hindsight is also a terrible choice for a point of reference, since it is quite overpowered in the vanilla game and there is actually a patch to balance it into something else provided for CryoEngines.

    So a better move would be to orient the balance around the Deinonychus, as it is a quite good engine which sets a theme of methane engines having very high thrust, excellent (but not ludicrous) TWR and good ISP. With the new TDW to aim for being ~250 kN/T, the Iguanodon could be nerfed to weigh 4 T (~240 kN/T TWR) and I'd still give the Allosaur a weight and thrust increase but to 2200 kN ASL and 9 T (~244 kN/T TWR). An argument could made to nerf them further, since they don't possess such huge drawbacks as hydrogen engines, but I think it would be better to try to regulate them trough means other than nerfing their performance, like cost or appearing late in the tech tree.

    And now for something completely different:
    The naming schemes for engines is a bit, well strange. Stock/Restock LF/Ox engines are "Liquid Fuel Engines", CryoEngines LH2/Ox engines are "Cryogenic Rocket Engines" and CryoEngines LCH4/Ox engines are "Liquid Rocket Engines". "Hydrogen Fuel Engines" and "Methane Fuel Engines" would be more consistent imo. Also, the CryoEngines Restock Patch changes the fuel type of some engines, but still leaves them named as "Liquid Fuel Engines". This may be an oversight, when not a limitation of MM Patches.

    [WIP] Nert's Dev Thread - Current: such nuke, wow

    in KSP1 Mod Development

    Posted

    23 hours ago, Nertea said:

    They're in the development branch now if you're brave, though you'll have to go get all the updated dependencies and waterfall dev yourself. 

    Gotta say, the engines look really beautiful ingame. Though they may need a little more balancing imo. The Deinonychus, while trust is already high compared to other 1.25m engines, does still loose quite substantially against the Vector when it comes to clustering. The Vector seems like an unfair comparison, since it is fairly high up the tech tree, but it is also 1.25m and doesn't suffer the same problems as methane engines (boiloff, less density). The Iguanodon however competes much better against the Vector, offering slightly higher thrust and ISP, while keeping a very similar albeit slightly larger profile when in compact mode, making it the new king of clustering. The Allosaur seems especially underpowered to me, being just a bigger and slightly more efficient Mailsaw.

    I think the best way to try to balance them is to look at the TWR of the engines:
    The Vector has a TWR of ~234 kN/T (ATM), making it the close second best TWR in the vanilla game (Mammoth being the best, which are 4 Vectors missing a ton). The Deinonychus has a TWR of ~256 kN/T, the Iguanodon has ~320 kN/T and the Allosaur has ~231 kN/T. There is an argument to be made that methane engines should have higher TWR, since they also need more fuel tanks to accommodate the lower fuel density and sometimes extra electrical systems to cover the boiloff. The bigger, less clusterable engines should also generally have a slightly higher TWR, to keep clustering from potentially being always superior.

    Personally, I would therefore give the Deinonychus more thrust, ~500 kN ASL, increasing the TWR of it to ~312 kN/T. The Iguanodon should get a slight mass increase to 3.1 T, lowering the TWR to 310 kN/T ASL. Since the Deinonychus has a slightly higher ISP than the the Iguanodon while at equal TWR, the Deinonychus is now generally slightly more preferable, but the Iguanodon still has the edge when it comes to size vs thrust, making it still more viable on bigger rockets. The Allosaur is quite huge, although in the 2.5m class, its scale is a lot more comparable to (non-cluster) engines of the 3.75m class than a Mailsaw, so it should probably get a big trust upgrade like 2200 kN ASL and a slight weight increase to 7 T, giving it a record TWR of 314 kN/T. This thrust upgrade also makes sense to bring it more in line with the other methane engines, which have generally much higher thrust compared to other engines of their size class.

    I'd also change the tech progression in the vein of what lemon cup said, though I'd even go a step further: Methane vacuum and hydrogen lifter engines get shifted one node to the right compared to LF engines, while methane lifter and hydrogen vacuum engines get shifted a whole two nodes. This kinda makes the CTT a soft requirement, but I image most people using CryoEngines using it anyway.

    [WIP] Nert's Dev Thread - Current: such nuke, wow

    in KSP1 Mod Development

    Posted

    11 hours ago, Nertea said:

    Revised content list (names are not really final), with approximate completion levels. 

    0.625m

    • [80%] Compsognathus: ESA Prometheus
    • [80%] Hawk: NASA RS-18

    1.25m

    • [80%] Deinonychus: SpaceX Raptor
    • [80%] Buzzard: Avio M10

    1.875m

    • [40%] Dromaeosaur: Blue Origin BE-4
    • [80%] Harrier: Landspace TQ11/12

    2.5m

    • [95%] Allosaur: Methalox STBE concept (that I originally built for NFLV)
    • [80%] Kite: SpaceX Raptor Vacuum. 

    3.75m

    • [60%] Tyrannosaur: 7x Raptor cluster in a cylindrical boattail
    • [30%] Vulture: BE-4U with extensible nozzle

    Instead of Dromaeosaur, Achillobator would fit better imo, since it is actually bigger than a Deinonychus (unlike the Dromaeosaurus), doesn't have the saurus suffix and is actually in the same subfamily of Dromaeosaur (Dromaeosauridae Dromaeosauridae).
    Allosaur could also be replaced by Baryonyx or Suchomimus, if wanting to get rid of the suffix. The Tyrannosaur name should probably stay, as the beefiest methalox engine (cluster) deserves to be named after the heaviest and most famous theropod.

    [1.12.x] SystemHeat - a replacement for the CoreHeat system (October 9)

    in KSP1 Mod Development

    Posted

    51 minutes ago, Nertea said:

    I'm currently doing some work to leverage the 1.11 inventory system for reactor repair and refuel. Repair kits are now needed to repair reactors - depending on the size of the reactor, you will need more repair kits to repair damage.

    For fuel, you will need your Kerbal to equip the new nuclear fuel canister inventory item:

    unknown.png

    I modeled it before I learned how useless the model is in the current implementation. It won't be shown anywhere, really! Squad made all these assets (eating a lot of memory) for just icons in the part picker. Oh well...

    Anyways, with a cylinder equipped, kerbals have the options to collect and store nuclear fuel and waste from reactors and containers. Slightly more fun.

    This seems like a really cool and logical way to differentiate and balance NFE fission reactors from FFT fusion reactors, if fission reactors can only be refueled manually by kerbals. The amount of radiation the kerbals will need to endure refueling is obviously not great, but not terrible either. Currently I feel like fission reactors are just superior in most cases, since fuel tends to last longer and can be easily reprocessed or mined, while fusion fuel is slightly harder to get and runs out far quicker.

    Also, I shall greet you from Dúnedain and Mathijs of the Age of the Ring mod team, whose dwarven visual style is still stongly reminicent of your work on TDH even after the very recent overhaul of that faction.

×
×
  • Create New...