Jump to content

blowfish

Members
 View Profile  See their activity

  • Posts

    4,559

  • Joined

  • Last visited

 Content Type 

Everything posted by blowfish

  1. blowfish
    At supersonic speeds, you need to have a terminal shock - oblique shocks reduce the mach number, but the flow will stay supersonic. In an ideal inlet, the flow will be only slightly supersonic when it hits the terminal shock, minimizing the pressure loss. I've seen some theoretical designs for compressors that can operate on supersonic air, but so far no one has actually built an operational engine out of one, and in any case, the efficiency is likely to be less than subsonic compressor stages. The helium loop bleeds heat from the preburner. This is the HX3 in the diagram. Yeah, I think hydrogen embrittlement is the main issue. True that you need to run hydrogen through something, but I think the main point is that the main precooler is large and complex, so you want to minimize wear on it / use of exotic embrittlement-resistant materials to maximize reusability and minimize mass. Plus, you'd need additional turbines in the hydrogen flow, which are also vulnerable parts. I have no idea what that would take or if it's possible.
  2. blowfish
    The smallest total pressure loss is achieved by a large number of low-angle oblique shocks before the normal (terminal) shock. The terminal shock is already far inside the engine (at what's called the intake "throat"). Some intakes are adjustable to adjust the amount of air that they capture - the SR-71 has retractable intake cones, and many ramp intakes can adjust their angles to only capture a certain amount of flow. But if you're trying to match through a large number of mach numbers, then the adjustment range is going to be quite large. That's how non-conservative forces work.
  3. blowfish
    I have studied jet engines rather extensively, and all the research I've done suggests that this is false. Cite your source.
  4. blowfish
    A few things: LH2 takes up a huge amount of volume but isn't very dense. Look at the vessel mass with stock fuels, and try to match it with real fuels. If you're filling the same tank volume with less dense fuel, you're going to have less fuel. The SABRE is a very unique engine cycle that doesn't really allow it to be classified as a rocket or a jet. A jet is probably a better bet in air-breathing mode. I've tried to come up with a better solver for the SABRE, but it has issues so it's not ready to see the light of day yet. REL has very inconsistent data, but it's possible that the SABRE's Isp is underestimated by a factor of 2. I might consider changing it. Isp in rocket mode should be set by RF Stockalike, not AJE. I added changes to RF Stockalike that increase it to 460s in a vacuum, but I'm not sure if those changes have been released yet. hope that helps The code that determines how fast engines spool has been in there for a long time and hasn't changed. I'm not opposed to changing it, but it has to be based on hard data on how long it takes real engines to spool, and how to extrapolate those values to engines that don't have data. I'm not going to change it to something that just "feels right." It may be too high, but again, without hard data I'm very hesitant to change it. And it's better than how it was before, where many engines would idle at negative thrust.
  5. blowfish
    @sevenperforce Airflow is determined by the intake and nozzle throat diameter. At supersonic speeds, conditions inside the engine cannot affect how air flows outside the engine, so there is a maximum airflow determined by the intake. So if the nozzle doesn't allow as much airflow as the intake, you can simply resize the nozzle. Now, there is a caveat to this - if you match the intake an nozzle at a particular airspeed, they will not match at other airspeeds. In particular, if the nozzle is sized for maximum airflow at mach 5, then there will be more air entering the intake than the nozzle can take at lower mach numbers (above mach 1). The SABRE solves this by allowing the excess air into a series of ramjet bypass ducts around the engine - they aren't shown in the diagram I linked, but plenty of engine cutaways show them. The bypass ramjets burn the air with hydrogen - since there is no compression beyond whatever ram compression happens in the intake, the bypass jets are not as efficient as the engine core, but it's better than simply bleeding the air overboard.
  6. blowfish
    Am I missing something? I see a mention of actively cooling the wings during re-entry, but no indication that this is done with excess fuel. The user manual states specifically that excess propellants are dumped from the main tanks before circularizing (section 2.3.1). There might be some hydrogen left in the auxiliary (OMS) tank, but not much.
  7. blowfish
    Air inside a jet engine is subsonic regardless of airspeed. The reason why supersonic engines have fancy intakes is to minimize the total pressure loss when decelerating the air to below the speed of sound. But even the best intakes are not perfect - you loose some pressure, and the fan/compressor has to make up that pressure in order to even produce positive thrust. That is why low pressure fans don't really work well at supersonic speeds. @sevenperforce I'm still not sure why you're obsessed with the turborocket idea. The oxidizer is there, entering the engine, you're only going to increase efficiency by using it. For those unfamiliar, here is a diagram of the SABRE's engine cycle: It's a little daunting at first, but the underlying concept is pretty simple: the helium loop acts as a heat engine, extracting energy from the temperature difference between the cryogenic fuel and the superheated air entering the engine, and using that energy to drive the compressor and fuel pump. This guarantees that (1) very high pressures can be used, which directly increases the efficiency of the engine, and (2) Energy that would otherwise go to waste is extracted to do useful work.
  8. blowfish
    There's no limit, but I think there's a bug on FXModuleAnimateThrottle where it will only look at the first Animation component it can find in the part. I think you can attach all your animations to one component, but I've never tried it myself.
  9. blowfish
    Logs. I have no clue what you have installed. But you should also probably use the downloads on the releases page unless you know what you're doing.
  10. blowfish
    That's not really correct. Subsonic jets would overheat if run at higher speeds, but even if they didn't they loose thrust very quickly (this comes from months of working with AJE's simulations BTW).
  11. blowfish
    Low pressure, high flow devices loose thrust and efficiency very quickly above mach 1. It's why supersonic jets all use low-bypass turbofans or turbojets. I don't know exactly how they plan to start it. Since it really only needs to start once per flight, they might just vent a bunch of pressurized helium through it to start.
  12. blowfish
    Has already, @linuxgurugamer It might be a good idea to update the OP.
  13. blowfish
    You're not going to get anywhere close to the Isp of the SABRE with that. Think about it - a jet engine or turborocket takes energy from combustion to impart energy to the air flowing through it. So in order to impart the same amount of energy to the air, you really need about the same amount of combustion mass. And most of that combustion mass is oxidizer BTW, the SABRE doesn't really function the same way as other rockets/jets as far as the pumping/compressing goes. Both the compressor and fuel pumps are driven by the helium loop, which gets its energy from extracting a temperature difference between the fuel and the incoming air. At lower speeds, the incoming air isn't very hot so it also needs a heat exchanger after the preburner, but there are no turbines in the actual propellant/air flow path. Once you get up to higher mach numbers, it's not actually tanking any energy from the combustion process.
  14. blowfish
    There's nothing special about setting up KM_Gimbal_3 compared to the stock module. Just make sure that the thrust transform (and any nozzle parts you want to move with the gimbal), and make sure that the animation doesn't touch the actual gimbal mover.
  15. blowfish
    They were split into a separate pack. Download the HX pack (it's in the same place). And welcome to the forums
  16. blowfish
    @Nibb31 Source on hydrogen cooling the skin on re-entry? I recall reading in the user manual that all excess propellant in the main tanks is dumped before even circularizing the orbit.
  17. blowfish
    If you're using BahamutoD's AnimatedThrust module, it's not going to work with a MultiMode. It only looks for one engine module, so it has no knowledge of the other. But at this point you're better off using the stock FXModuleAnimateThrottle, which accomplishes the same things and does support multi-modes. You just have to set the preferMultiMode flag so it knows to look for the MultiModeEngine module rather than either of the individual engine modules.
  18. blowfish
    I don't think this is true. Air is about 23.2% oxygen by mass, but you only need 1 kg of hydrogen for every 8 kg of oxygen, so you're really getting most of the mass you're combusting from the air. In other words, you'd need to have several times the fuel flow to get the same thrust if you carry the oxidizer. Density does affect drag, but ultimately on a single stage vehicle you need all the efficiency you can get, and methane doesn't really cut it. And a higher takeoff mass means larger wings, which means more drag anyway, especially at higher mach numbers. Everything you're mixing in the combustion chamber should be at the same pressure, otherwise you're loosing efficiency. And the precooling is not really about temperatures after combustion so much as temperatures in the compressor - the SABRE has no moving parts in the flow after the compressor, the compressor itself being driven by the helium loop. This, combined with the precooling, means that the air can be compressed much more than in traditional jet engines, increasing the specific impulse significantly.
  19. blowfish
    Could you be specific? What files did you download and what did you install from them? And for the third time, we wouldn't be having this back and forth if you would provide logs. If I could see your logs I could identify the problem immediately. Signature is under account settings.
  20. blowfish
    Okay, what exactly did you install? You seem to have the old Mk2 parts (part of the legacy pack) but not the new ones (part of the core pack). And again, logs would probably reveal the issue immediately.
  21. blowfish
    Logs would be helpful. See the first link in my signature if you're unsure of how to get them. Re: FSfuelSwitch - all of the non-legacy parts have already been moved to a custom switching module ModuleB9PartSwitch. Only the legacy parts still use FSfuelSwitch. I considered using IFS but (1) It doesn't provide all the functionality I want and (2) It, like FSfuelSwitch, is very painful to mod with.
  22. blowfish
    @Jimbodiah Okay, clearly there is some discrepancy here. My upper stage tank contains 7110 units of LH2 and 474 units of oxidizer, and masses 0.43t dry (this is a 4.0875 x 1.875m tank with the flat mount). The mass numbers that MechJeb is reporting are slightly different than those that RCS Build Aid and the Engineering Report are saying, but of any of those I'd trust MechJeb (and it seems to be right - the discrepancy is in the engine but MJ seems to report it correctly). The reported delta-v and TWR are consistent with the stats of the engine (27 kN and 450s Isp). Does anything here not match up with what you see? EDIT: here is my test craft, for what it's worth. Maybe you could load it and see if it produces any discrepancies with the numbers MechJeb shows in my screenshot. I could try the same for yours.
  23. blowfish
    And why would I want the shortest tank? I was initially about 5-10 degrees above prograde, but that actually proved excessive. Your time to apoapsis just has to be about half your burn time.
  24. blowfish
    Rescaling the solar system. Parts remain with their default balancing, including the LH2 patches. Delta-V to orbit is about 5.5 km/s, so I believe the scale is being done correctly. You can see the craft, including its delta-v and TWR distribution here, and the final orbit here.
  25. blowfish
    @Jimbodiah I just modified my rescale settings to 3.2x and tried getting the Soyuz capsule to orbit. And honestly, using the soyuz launcher and a single RL10-A-3 upper stage, I was having trouble making it small enough. Even with a high liftoff TWR I ended up with over 1 km/s of leftover delta-v after making a 180 km orbit around Kerbin. So I'm entirely unsure of why you think more thrust is needed. E: Oh and @Shadowmage I think the Soyuz service module could have its thrust cut by at least a factor of 2 - It's like 0.66 with the orbital and descent modules as it is which is way more than necessary for orbital operations.
×
×
  • Create New...