Jump to content

SpaceX Discussion Thread


Recommended Posts

2 hours ago, Scotius said:

No! Oh, come on. Delta IV Heavy is awesome. It's power personified :D. At least let's wait until the first successful flight of Falcon Heavy to proclaim  it as the new Vice-King of Rockets. King of course being Saturn V.

The last time something tried to usurp the King of Rockets, it exploded 4 times.

Link to post
Share on other sites
2 hours ago, DMagic said:

The total mass to LEO has always seemed like an academic argument. I've heard lots of times that the fairings are about the size of a regular bus, which sounds big, but consider that an actual bus only weighs 10-15000 kg.

From when I calculated a Uranus probe (hehe) that used 85% of Falcon Heavy’s payload it barely fit in the fairing. And it was around 40t of propellant. It was an NTR/NH3 propellant probe and I think it had like 13 km/s dV.

edit: found my old post: 

details: prop mass: 48t, dV: 11.7 km/s, fairing margin (w/o probe): 1.8m

Edited by Racescort666
Link to post
Share on other sites
5 minutes ago, Scotius said:

When we'll get to send a probe to that planet, i predict a mulchstorm of bad jokes, awful puns and lewd comments flooding the rocketry part of the Internet :sticktongue: And Reddit... *shudders*

I generally pronounce it “oo—RAN—ohs” and people look at me funny. Of all the possible names, I would prefer “King George” the best.

Link to post
Share on other sites
3 hours ago, DMagic said:

The total mass to LEO has always seemed like an academic argument. I've heard lots of times that the fairings are about the size of a regular bus, which sounds big, but consider that an actual bus only weighs 10-15000 kg.

They could make an larger fairing if an customer wanted an oversize payload. 
Cheaper than flying an falcon heavy in disposable mode, no need to make it before someone want to pay for it. 
Even if SpaceX facility is unable to make larger fairings they could outsource it. 

Link to post
Share on other sites
55 minutes ago, Racescort666 said:

I generally pronounce it “oo—RAN—ohs” and people look at me funny. Of all the possible names, I would prefer “King George” the best.

I tested "Ouranos" once and it was received pretty well. However, I think the best option would be to convert to the Greek "Caelus". It would be a diversion from the typical Roman planet naming scheme, but it's towards a just cause. It also sounds a lot less out of place than "George". Perhaps we should petition the IAU?

10 minutes ago, magnemoe said:

They could make an larger fairing if an customer wanted an oversize payload. 
Cheaper than flying an falcon heavy in disposable mode, no need to make it before someone want to pay for it. 
Even if SpaceX facility is unable to make larger fairings they could outsource it. 

There are aerodynamic issues that come into play, you can't just put a bigger fairing just like you can't just put moar boosters on the side. In some cases it works out but I don't think FH can take much more fairing size. It's long been talked about as a issue with the system.

Link to post
Share on other sites
17 minutes ago, cubinator said:

There are aerodynamic issues that come into play, you can't just put a bigger fairing just like you can't just put moar boosters on the side. In some cases it works out but I don't think FH can take much more fairing size. It's long been talked about as a issue with the system.

MOAR... fins, maybe? :/

Link to post
Share on other sites
1 hour ago, cubinator said:

I tested "Ouranos" once and it was received pretty well. However, I think the best option would be to convert to the Greek "Caelus". It would be a diversion from the typical Roman planet naming scheme, but it's towards a just cause. It also sounds a lot less out of place than "George". Perhaps we should petition the IAU?

There are aerodynamic issues that come into play, you can't just put a bigger fairing just like you can't just put moar boosters on the side. In some cases it works out but I don't think FH can take much more fairing size. It's long been talked about as a issue with the system.

Yes an wider fairing might well have issues, an common KSP issue is that you can not drop first stage early as you need the long stage to keep your oversize fairing aerodynamic stable. an stubby upper stage would flip in atmosphere. An longer fairing should not generate much problems, neither would an 1 meter wider one, you could even have an more relaxed tapering at bottom.  
An more serious issue with falcon heavy is the second stage, can it handle 50% more load?
 

Link to post
Share on other sites
44 minutes ago, magnemoe said:

Yes an wider fairing might well have issues, an common KSP issue is that you can not drop first stage early as you need the long stage to keep your oversize fairing aerodynamic stable. an stubby upper stage would flip in atmosphere. An longer fairing should not generate much problems, neither would an 1 meter wider one, you could even have an more relaxed tapering at bottom.  
 

Longer fairings tend to cause problems too IRL, I think...

Link to post
Share on other sites
1 hour ago, cubinator said:

I tested "Ouranos" once and it was received pretty well. However, I think the best option would be to convert to the Greek "Caelus". It would be a diversion from the typical Roman planet naming scheme, but it's towards a just cause. It also sounds a lot less out of place than "George". Perhaps we should petition the IAU?

Caelus is Roman. Ouranos is Greek. Uranus is... Latinized Greek, I guess.

I do agree, though, Caelus would work pretty well instead of Uranus.

Link to post
Share on other sites
13 hours ago, magnemoe said:

They could make an larger fairing if an customer wanted an oversize payload. 
Cheaper than flying an falcon heavy in disposable mode, no need to make it before someone want to pay for it. 
Even if SpaceX facility is unable to make larger fairings they could outsource it. 

If there is a bigger fairing, there is much more drag too, if the shape is the same

 

Very hyped for FH :sticktongue:

Link to post
Share on other sites

There's an Amazon joke and also a Barn comment in there somewhere. :) I shall resist.

Hope Chris B has it right and that the fit checks are going to plan!

Link to post
Share on other sites
On 12/28/2017 at 11:44 AM, michal.don said:

If all goes to plan, there should be a fueling test in a few days, and a static fire in about a week.

Brace yourselves gentlemen, looks like it's finally here :)

look again its already gone.

 

Link to post
Share on other sites
2 hours ago, CatastrophicFailure said:

Nah, based on @tater‘s retweet it’s just having a nice lay down after all that Heavy lifting. :D

Reason for not going to space today number 246: Rocket was getting a bit cranky and needed a nap.

Link to post
Share on other sites
20 hours ago, magnemoe said:

Yes an wider fairing might well have issues, an common KSP issue is that you can not drop first stage early as you need the long stage to keep your oversize fairing aerodynamic stable. an stubby upper stage would flip in atmosphere. An longer fairing should not generate much problems, neither would an 1 meter wider one, you could even have an more relaxed tapering at bottom.  
An more serious issue with falcon heavy is the second stage, can it handle 50% more load?
 

I would make the points that rockets that want to flip over, in my experience have other problems. I have posted pictures here of very large structures (space factories, telescopes launched in the game that I have launched whole into space). Although KSP does not model entirely the aerodynamic forces properly (see other threads) I would point out that when I have had tip-overs there have been problems with the rockets generally that would have been problematic in real life.

And . . . . . . . . Take a look as some of the Russian rockets, they don't often use interstage shells.

There is a benefit to doing that, you don't need to rescale the fueltank or engine if the engine does not fit under a flush mounted fairing, (and with some of the deep space engines the nozzel diameter to thrust ratio is high-ish.

When we talk about fairing we first have to address the question what is Max Q and where does it apply. If the users familiarized themselves with this problem they would have less problem in the game (Vanamonde this is about spaceflight not KSP as I point out some of what people think are game issues are design issues minimizing the consequences of physical forces)

46ddd1684c1211cf11375559548759ac191c47b7

Max Q is created by the fact that as a rocket presses to orbit its free-air relative velocity magnitude increases as T*a + T*da/dt  where a = acceleration and p = f(dy) and since at the beginning of the flight is  were Max Q is experiencedthe Y component unit vector of velocity is close to one. and thus the dy at any point is given by  dy = 0.5aT^2 + 0.125(da/dt)T^3 - 0.5gT^2, note that the T^3 component is a function of time, so that you can control MaxQ by controlling time. The equation is flawed however as on approaches 275 m/s the dynamic pressure is follows the above equation, however above 275 the mach effect takes over and just over Mach (varies with temperature) its about 3 times the anticipated pressure. To state the otherwise the boundary layer determines the Area in which pressure is applied. At very low velocities the boundary layer is connected to the surface it is the area around the aircraft where the speed of the air relative to the craft changes most rapidly with distance. Its where air has to accelerate and since air is massive and accelates over distance it creates drag.

So lets take an Tractor trailer going down a highway, a pedestrial on the side of the road will feel a moment of air pressure and then the flow of air forward as the truck passes. So a passenger car passing a tractor trailer on the highway gets drawn into the truck as gets close to the front, and then passes into this shock and is pushed away and is released as it finally passes. Behind the truck the a constant power vehicle will speed up.

Lets talk about objects on the truck, the mirrors, the horns the roof lights, the top front edge of the trailer all experience increased pressure. As speed increases to Mach speed the pressures vector on all parts will change, obviously, but some will change direction almost to the opposite direction (that's why steering craft at mach speed can be problematic). If you speed the truck up to Mach speed the a small sears-haack shaped car  passing the truck would all but experience a vacuum behind the truck and would be sheered by the shock wave as it passed the truck. This is a laymans view of Mach Effect.

Getting back to boundary layers, its not simply that pressure increases. If we consider a parts Mass as all the all mass closer in radius to the axis of motion than the boundary layer distance to that axis then the mass of the parts has increased. So you have Pressure and Mass increases and these apply to all parts that have a net + δpressure vector that is pointing into the part. Suffice to say that these vectors are not constant, and in fact can change rapidly passing through the Mach barrier

So the next thing that you want to know is were the pressure and mass increases the most and which changes in the direction are beneficial and detrimental to spacecraft. Why are we asking this question, the reason is this.
During the vertical portion of flight the spacecrafts gain in acceleration relative to the earths surface is (TWR - 1) * 9.8. As TWR is increases relative to 1 then less of the thrust is wasted hoovering over the launch site. The drag is generally a small fraction of acceleration, so it serves the craft to accelerate as rapidly as possible when drag is low, but relative to an increasing a from a rocket, drag will almost always be lower because altitude is always increasing with speed, and the scale factor is 1/ek so that the faster one goes vertically the faster that pressure drops; were it not for the Mach barrier Q would not be much of a problem. But by the same token we don't really need to get away from the launch site at maximum acceleration when the velocity is already high. The air we need to pass in vertical portion of flight is only about 20 to 30km and 27km / 275 m/s is 100 seconds. But if we discount the 11km needed to get to 275 thats only a minute of flight. dV of most rockets can live with that although somewhat wasteful. If you had a stage at this point that had enough thrust to turn significantly horizontally while still providing net positive acceleration upward, it might be worth the cost (depending on the fragility of the payload). For most rockets as long as the speed stays below 275 you are going to survive if you don't turn to quickly.

There is also a difference between the game and RL (actually RSS): that to orbit in the game requires only 3800 dV, in RSS or RL its 9000 to 10,000. As a consequence for any given size payload there is alot more Mass below the payload in RSS than in KSP. And in the game you eventually cannot build higher you have to build that mass out such that it creates a steering effect for the rocket. IOW in any size atmospheric planet that only has 3800 dV need to reach orbit,you can have a similar problem unless you design the rocket to go far beyond the planets orbit (like back to earth).

So specifically we want to ask the question what specific areas of the space craft would suffer damage if pressure vectors altered direction along with pressure increases while under 300 m/s or below 30km.  The obvious first is the nose of the space craft, the second is the any side tanks, and finally an wind structures like fins. Obviously nose cones have to be durable to support the added mass of air and the pressure on top of that. However if the fairing is fairly large then any parts directly below that are relatively protected up at least to 275 meters per second with an aerodynamic space craft. 

So that if a tank immediately above and segment is set in slightly from the previous section it should not be affected up to 275 and air should flow laminar-like between separated sections.

So from 275 m/s to 320 m/s is the problem area. When the ship approaches mach speed the air coming off the nose's boundary layer separates, flows outward like the tractor trailer but then flows inward. As speed increases that arc's endpoint moves down the rocket and finally passes the engines. The rockets motion (and acceleration translated to motion and then back to areodynamic force) acting with a net radial air motion with a vector that increases rapidly at Mach speed. As the boundaries air-relative speed slows (and looses momentum)  the pressure of the atmosphere forces it back to the craft increasing the -radial motion, this creates a momentary net pressure on the craft. If this counter shock waves passes in between two unprotected parts (i.e.) an engine cavity the pressure is not downwards but against both parts. So if that connection requires two parts to constantly press to each other that could create a momentary problem. The amount of momentum that air is going to have depends on the pressure. Once the craft is well above Mach or temperature falls, the bow shock recombines so far behind the craft that is no longer of concern. 

So that in the equation above V = f(t) + f(t2) of a and altitude is a f(t2) + f(t3) of a. is there a way to arrange a = f(t) such that alt is high and Q is lower.

Solutions

For spacecraft that want to tilt.
Are you using the right drag cube and in RL/KSP does your nose cone have a Sears-Hack shape. If you don't think that it is behaving properly why not change the Max drag coefficient (lower). Sears-Haach shape offer major increases in performance. However you may want to use interstage fairings, those counter shocks hit close home at higher velocities the more aerodynamic the shape. My space factory that was 30M in diameter had  forward section shaped like a bullet. Never tried to lift it in RSS but have the DV to do it. A second problem is fishing pole behavior, is the upper stage secure or is it pivoting back and forth. Another problem is segmental flexibility, can you redo the section with fewer vertical joints. Can you increase the torque values in the config (the falcon heavy problem) and also be fair and increase mass. Standouts and struts can transfer torque down the craft to the center of mass.

Another problem with tilting craft is having a solid core booster at the center of the craft. These make great power peices, but you cannot control power and, worse, it make the center of mass climb quickly and, see above, the altitude has a 3rd function of time if da/dt is very fast you will break Mach at a dangerously low altitude.  if the SFRB burns out after reach Mach, the craft is likely doomed with out adequate steering fins in the second section(and these are exposed to Mach effects also in RL). Its better to boost a core of LFOx and release before MaxQ.

To deal with the specified problem, something that happens in the game that might be better controlled in RL (with hydrolics tuned to the craft) is separation angular velocity. Again this can easily be circumvented by having side tanks that are stage to release in pairs while the core is mostly powered. (Space shuttle SFRB like). The gimble needs to be tuned to the first stages inertia, the large the inertia the larger the gimble limit. Who said rocket science was easy.

Have side tanks that burn to 275 m/s and then are kicked off slowing acceleration. This is good, but the time spent at mach speed is a problem. But lets say that at 275 you drop the tanks and acceleration = specific (gravity - drag) then the craft is not accelerating relative to earth surface (which lest we not forget is being pushed up at 9.8m/s2) at all, its simply moving and as drag force falls it will go faster. This reverses pretty much all a added in the early flight, and then begins accumulating a again. But the metric is very different. Alt = d + VoTsep + 0.125(da/dt)T3. In KSP make sure that the detachment point is higher on the tank side than lower. Don't use a radial separator that is too powerful. 

So that this craft will slowly begin to accelerate to MΘ slow down. The performance through mass could be improved by throttling down for a few seconds past 290 and then full thrusting. Another solution is to tilt the craft since horizontal acceleration is not deprecated by gravity, verticle acceleration will slow but horizontal acceleration will speed more rapidly.

Thats  solution that will both remove mass from the core and slow down. It solves the tilt problem and the cavitation problem. To solve other cavitation problems.

Another solution is simply to trottle down and throttle up.
A third solution is just to make sure that intersections are reinforced.
A forth solution is to varigate the nose cone causing the reformation of the boundary layer to spread out over a wide area. (this increases mass and form drag  on the top).
Another solution would be to have microwave stations heat the air in front of the space craft increasing the speed of the mach barrier.
Microwave generators under the nose cone could cause altenating hot and cold areas over the nose causing them to cross mach speeds a different times (disrupting the shock wave).
One could feed air from the nose cone into the intersection and along the form pressurizing the area under the bow shock. This would lower the momentum of the counter shock.
And alternative is to simply bleed air off the edge of the cone of the payload fairing and spread it out under the bow shock (this has the advantage of blowing back engine vortexes. The bleed holes could be gated and open up only when speed reaches a certain speed and close once the speed window is passed. This actually benefits the tilt problem also, because once a craft starts to tilt the lack of pressure on the -verticle side of the craft increases the likelihood it will continue to fall.
Add a reinforced fairings.

The solution in KSP is to create more powerful struts, which is badly needed in the game (apologies to banana).

Give me any payload and I can get it into space, right now my limit is 2 kT,  I dont care about the shape, it could be like a satellite antenna and I can still get it into space.

 

 

 

 

39 minutes ago, KSK said:

Reason for not going to space today number 246: Rocket was getting a bit cranky and needed a nap.

It need to rest its feet after a long day of walking down the trackway.

Link to post
Share on other sites
  • Dman979 locked and unlocked this topic
  • Dman979 locked and unlocked this topic
  • Dman979 locked this topic
  • Dman979 unlocked this topic

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...