Starship is Go...

GW Johnson · 2021-08-12 11:49:05

The ullage thrust for free-surface tank designs must be applied for the entire duration of the transfer event, plus the interval it takes for all the globules to settle and coalesce in one end of the tank.

Yes, it will affect the orbital path of the vehicle. That's why the ullage motors on the Saturns pointed out the tail: (1) to settle propellant into the tail of the tank where it was when sitting on the launch pad and during the ascent burn, and (2) add to the delta-vee of the ascending vehicle (although a fairly negligible contribution; we are talking about forces for roughly 0.001 gee's acceleration).

For on-orbit refueling, what you do is orient the ullage thrust to the side instead of along the orbital path. That way your perigee and apogee are unaffected; only your orbit plane changes, and by an almost unmeasurable amount (it'll be in the parts per million range because of the high value of orbital velocity). If you alternate refill ullage thrust directions by 180 degrees between tankers, that unmeasurable plane change really is zero.

Guys, Musk himself says he's not thinking much about this issue until he starts getting Starships successfully to orbit. If he's not thinking about it, neither are his Spacex team. They'll figure this out as they go. That's been their pattern. All of this is angels-on-the-head-of-a-pin speculation, until Spacex is forced to face up to, and deal with, the issue. They haven't yet reached that point.

GW

Last edited by GW Johnson (2021-08-12 11:52:32)

tahanson43206 · 2021-08-12 12:52:30

For GW Johnson ... re #1426

Thanks for the reminder of the power of plane change!

SearchTerm:ullage thrusters for refueling per GW Johnson apply to plane change
SearchTerm:refuel technique for side-by-side Starships

Nice!

Looking ahead a bit, and taking into account the potential of ion thrusters to perforum ullage duty, the much vaunted concept of orbital refueling depots might come down to specialized satellites designed to provide ullage services.

That prospect is ** certainly ** a business opportunity for someone(s).

(th)

Oldfart1939 · 2021-08-12 13:52:49

Many of these "self-styled experts" are new guys just looking for clicks. I really only follow 2-3 of the better pundits: Scott Manley, Tim Dodd, and Felix Schlang. These 3 seem to put a lot of work in their presentations and aren't just a computer generated voice over some stock footage. Marcus House also does a fair job in his accuracy.

Last edited by Oldfart1939 (2021-08-12 13:55:53)

Calliban · 2021-08-12 15:33:21

An alternative that wouldn't involve constant thrust, would be to lift a counterweight into orbit, attached to a long cable. Both Starships would be attached at one end of the cable, the counterweight at the other. The Starships and counterweight would then rotate around a common centre of gravity. The counterweight could be a lot lighter than the combined mass of the Starships and could be reused hundreds of times.

Alternatively, each fuel tanker could carry its own counterweight, which could be reeled out several hundred metres on the end of a long, thin cable. So long as the fuel tanks of both vessels are kept beneath the axis of rotation, then one tank can drain into the other by (artificial) gravity. It all comes down to what is most mass-efficient and what works best for mission reliability.

Last edited by Calliban (2021-08-12 15:39:00)

GW Johnson · 2021-08-12 15:47:47

The cable spin thing was done at around 0.001 gee on one of the Gemini missions that docked with an Agena target, back in the mid 1960's. It sort of worked, but they never spun it any faster. The problem is you really need to accelerate both weights tangent to the circles they will follow, simultaneously, and at proportional acceleration levels.

You have to do this on a cable with no slack in it while freely floating in zero-gee. Otherwise you get oscillations in the cable; half the oscillation cycle you are "pushing on the string", which is a recipe for failure. They had no way to meet the start-up and spin-down criteria during the Gemini mission, which is why they restricted that experiment to about 0.001 gee.

Spin dynamics with cable connections is horrifically hard and complicated. With rigid bodies, it is relatively easy. And you have two solutions to choose from: spin axes are stable for both the min and max moments of inertia (just not the intermediate one if it is distinct). All you have to do is implement enough symmetry in your shape to zero-out (or at least minimize) the cross products of inertia (there are 6 of them).

GW

Last edited by GW Johnson (2021-08-12 15:53:01)

Calliban · 2021-08-12 16:17:09

GW, that sounds very complex and I find it hard to visualise. One thing that does seems unavoidable with a tether and solid counterweight idea is that centre of mass will shift as fluids are transfer from one tank to another. That will shift the velocity of both components around the spin axis as momentum is conserved. It this where the oscillations come from?

Maybe a rigid arm attached to a permanent orbital counterweight is a better idea. One could even design a trim system within the counterweight to ensure that centre of mass remained constant as fluid drained from one tank to another.

One thing I like about the idea of an orbital counterweight rig with a rigid arm, is that it could be designed to produce rotation without any expenditure of propellant. Just attach an electric motor and dumb mass to the centre of gravity and spin it using solar electric power. Conservation of angular momentum will spin the rig and attached Starships in the opposite direction. When propellant transfer is complete, apply a brake to the motor and spin of all components is reduced to zero as angular momentum is conserved.

Last edited by Calliban (2021-08-12 16:19:45)

GW Johnson · 2021-08-13 07:46:43

Calliban:

It really is complex and hard to visualize. Don't feel bad about that. I was referring to spin-up and spin-down transients. But you are correct to worry about the effects of cg shift, too.

The transients I was worried about take the form of oscillations that jerk on the cable. When the masses respond, cable tension goes up and down, and if large enough oscillations occur, the tension can zero during part of the cycle. The resulting sharp jerks will break something. The stiffer the cable, the worse the jerks. The lower strain capability the cable has, the more likely it will break. But the attach point structures are also at risk.

GW

louis · 2021-08-15 13:10:50

https://twitter.com/elonmusk/status/1426715232475533319

Elon confirms we're on: "First orbital stack of Starship should be ready for flight in a few weeks, pending only regulatory approval".

SpaceNut · 2021-08-17 14:49:01

Looking at the staging time and boiloff for a mars window of opportunity to launch. Since we want 2 cargo ships to land a full mars cycle ahead of another pair with a crewed ship ignoring the crew size thats quite a total number of launches to make it happen.

First off, boiloff is a lot lower in interplanetary space than in planetary orbit, as there is no nearby warm body radiating IR, so for most of the trip you will see a much lower boiloff rates. Secondly, the boiloff figures you quote are for hydrogen in LEO.
For LOX and LM there are ways to reduce boil off maybe to near zero. By orientating the craft so the engines or crew capsule faces the sun the boil off rate could be dramatically reduced because the amount of solar radiation absorbed would be greatly reduced and the heat radiated to space would be maximized.
But by default liquid methane is very close in temperature to liquid oxygen since it’s boiling point is -161.5 °C and it becomes a solid at -182 °C. Cryogenic Propellant Temperatures Due to the closeness in boiling points, we’ll likely see SpaceX chill their methane down closer to -180 degrees Celsius as they continue to squeeze performance out of that beast.
We don't want the CH4 to freeze, so the LOX should be allowed to rise to the freezing point of CH4 -182.5°C =90.65K .

http://www.unece.org/fileadmin/DAM/tran … 02-06e.pdf

The measure for the boil-off is the amount of vapours per unit time. It can be an absolute measure – kg/h, kg/day or a relative measure – % vaporized from total amount per unit time.

https://space.stackexchange.com/questio … ip-to-mars

The atmospheric boiling points of the chosen propellants are as follows: Oxygen 90K (-183C, -287F) Methane 111K (-161C, -258F) Compare Hydrogen 20K (-253C, -423F) Space is a place of temperature extremes: roasting in the sun, but pretty cold in the shade.

https://www.reddit.com/r/SpaceXLounge/c … ions_uses/

Note that simply painting Starship with the right white paint could reduce the solar absorptance from 0.39 to 0.09, which would reduce boiloff by a factor of 0.23, to 0.62 percent/day for LOx and 0.72 percent/day for methane, even if Starship is oriented in the worst possible way.

https://cryocooler.org/resources/Documents/C17/056.pdf

So since the star ship is going to have tile its got to spend its time facing away from the sun. If its got solar wings its got to be able face them towards the sun. The other side of the ship needs to be white to reflect heat. Since its seems that we are going to launch no quicker than a month appart that leaves days between refueling for it to boil off even with all of the above being taken into account.

Lets ignore the ration and difference of fuel boiloff and use the lower 0.62 %

At month 1 we are basically zero in the main tank with the header tank holding 100Mt and at the next
1 month launch that fuel load will have boiled off to become 80Mt in the header tank.
We fill the main tank with 100Mt and wait for the next flight to happen.
Of course the header tank will drop to 68Mt and the main tank will be at 80Mt of fuel due to the boil off staying consistent.
This is why we need to fill faster and having the extra ships of cargo makes the refueling levels even higher to achieve.
So at that time of 2 months end you have a ship that has been in orbit and has just 150Mt for the ship to use.

louis · 2021-08-17 17:34:46

Felix's latest video:

https://www.youtube.com/watch?v=yBmbkK043nw

Interesting details on the catching system and issues with the heat tiles.

Felix says 3-6 weeks for the first orbital launch . "Solid chance" of a September launch. I agree but that is of course subject to FAA approval, and that will be essentially a political decision in this case.

louis · 2021-08-17 18:01:06

Not sure if it's realistic but Space X talk in terms of having a Starship undertake 3 orbital flights within a 24 hour period. Not clear how that relates to refuellers but it shows ambition!

I think everything about Space X suggests they are aware of all these boil-off issues and have resolved them in a theoretical sense.

A lot of these "barriers" are proving less difficult over time. Previously a lot of people made out a retro rocket landing on Mars was impossible - now nobody says that.

My view is if they have a Starship rocket that can get humans safely to orbit and land back safely on Earth, they essentially have the rocket to get humans to Mars because the interplanetary part of the journey is the easiest. When I say land back safely on Earth, I think they will eventually try test rock-field landings.

SpaceNut wrote:

Looking at the staging time and boiloff for a mars window of opportunity to launch. Since we want 2 cargo ships to land a full mars cycle ahead of another pair with a crewed ship ignoring the crew size thats quite a total number of launches to make it happen.
First off, boiloff is a lot lower in interplanetary space than in planetary orbit, as there is no nearby warm body radiating IR, so for most of the trip you will see a much lower boiloff rates. Secondly, the boiloff figures you quote are for hydrogen in LEO.
For LOX and LM there are ways to reduce boil off maybe to near zero. By orientating the craft so the engines or crew capsule faces the sun the boil off rate could be dramatically reduced because the amount of solar radiation absorbed would be greatly reduced and the heat radiated to space would be maximized.
But by default liquid methane is very close in temperature to liquid oxygen since it’s boiling point is -161.5 °C and it becomes a solid at -182 °C. Cryogenic Propellant Temperatures Due to the closeness in boiling points, we’ll likely see SpaceX chill their methane down closer to -180 degrees Celsius as they continue to squeeze performance out of that beast.
We don't want the CH4 to freeze, so the LOX should be allowed to rise to the freezing point of CH4 -182.5°C =90.65K .
http://www.unece.org/fileadmin/DAM/tran … 02-06e.pdf
The measure for the boil-off is the amount of vapours per unit time. It can be an absolute measure – kg/h, kg/day or a relative measure – % vaporized from total amount per unit time.

https://space.stackexchange.com/questio … ip-to-mars
The atmospheric boiling points of the chosen propellants are as follows: Oxygen 90K (-183C, -287F) Methane 111K (-161C, -258F) Compare Hydrogen 20K (-253C, -423F) Space is a place of temperature extremes: roasting in the sun, but pretty cold in the shade.
https://www.reddit.com/r/SpaceXLounge/c … ions_uses/
Note that simply painting Starship with the right white paint could reduce the solar absorptance from 0.39 to 0.09, which would reduce boiloff by a factor of 0.23, to 0.62 percent/day for LOx and 0.72 percent/day for methane, even if Starship is oriented in the worst possible way.
https://cryocooler.org/resources/Documents/C17/056.pdf
So since the star ship is going to have tile its got to spend its time facing away from the sun. If its got solar wings its got to be able face them towards the sun. The other side of the ship needs to be white to reflect heat. Since its seems that we are going to launch no quicker than a month appart that leaves days between refueling for it to boil off even with all of the above being taken into account.
Lets ignore the ration and difference of fuel boiloff and use the lower 0.62 %
At month 1 we are basically zero in the main tank with the header tank holding 100Mt and at the next
1 month launch that fuel load will have boiled off to become 80Mt in the header tank.
We fill the main tank with 100Mt and wait for the next flight to happen.
Of course the header tank will drop to 68Mt and the main tank will be at 80Mt of fuel due to the boil off staying consistent.
This is why we need to fill faster and having the extra ships of cargo makes the refueling levels even higher to achieve.
So at that time of 2 months end you have a ship that has been in orbit and has just 150Mt for the ship to use.

SpaceNut · 2021-08-17 18:37:02

The star ship orbital landing is a tall order to be able to duplicate each time with perfection but the one coming back from Mars is even taller as the heat shield must stand up to the higher speed.

3 flights means you have 3 launch pads with no errors in flight....

This is precisely why we need a fuel deport with active cooling and sun-shade on orbit for this task as we eliminate the time cycles to load and go with out all of the lost fuels.

tahanson43206 · 2021-08-17 20:32:07

For SpaceNut re #1437

There is so much going on, it is easy to miss developments.

The plan published recently for the SpaceX proposal to NASA uses a single tanker as a fuel depot.

All the flights of tankers from Earth will be to fill up that dedicated tanker.

When the dedicated tanker is full, regardless of any boiloff, the crew vessel will launch and refuel at the dedicated tanker. There is NO issue of boil off. None! The dedicated tanker will receive fuel and oxidizer until it is full. When the tanker is full, the crew will launch.

(th)

SpaceNut · 2021-08-17 21:03:56

That means no cargo landers are going to mars with a crewed vehicle with only 1 depot ship in orbit and the filling of that cycle does matter since there is only a 3 month typical window for launch to mars possible approximately every 2 years and 7 weeks.

There is still boiloff as the star ship can not get filled when the time to fill the depot keeps getting longer and if a monthly flight does not happen the fuel it will hold is even lower in the depot to fill a starship with.

The depot needs active cooling, paint for reflection of heat, a power system that is isolated and radiators to get rid of heat.

SpaceNut · 2021-08-18 18:42:46

Would some one check values of boiloff for starship even thou we know that each would be a different number and volume we still are close to assume that they will track together.

I did a few cycles just to see a timeline events and amounts that would remain and its not what we think.

Its that number of days between fueling to the next which will be the issue when adding up all of the cargo ships and crew for a journey.

kbd512 wrote:

SpaceNut,
Will do.
Regarding your LNG boiloff calculation:
LNG Boiloff Rate Calculation by Dr. Mihai Ursan for the LNG Task Force, Brussels, 2011

tahanson43206 wrote:

For kbd512 ... re #1497 ... thank you for the the link to the LNG boiloff paper ... I appreciated seeing the comparisons between storage sizes and type, and the explanation of coping mechanisms.
For SpaceNut ... as you think about the boil-off problem for Elon's dedicated fuel depot tanker, please add consideration of the coping methods as described in Dr. Ursan's paper.
My guess (only a guess at this point) is that Solar Power (on the outside of a large heat shield) ought to be more than sufficient to provide power for the coping mechanisms.

No solar panels like on the dragon truck section covering the heatshield as they would need to be shed for re-entry into the atmosphere which would leave the anchor connection exposed...

lets redirect to On orbit fuel depot

GW Johnson · 2021-08-18 19:59:54

The propellant ullage problem as it relates to propellant transfers was being discussed about post 1420 and several subsequent. I have looked rather closely at the ullage thrust idea, the bladdered tank idea, and the displacing piston idea. I posted those results on "exrocketman", as an article titled Propellant Ullage Problem and Solutions, dated 18 August 2021.

The solution that is historically proven, and supported by technologies that are ready to apply, is ullage thrust applied to a vehicle with free-surface tanks. That is the one I recommend. The ullage thrusters will have to be fired for adequate duration each time there is a free-fall engine ignition, propellant transfer, or periodic boiloff vapor-venting event.

This not only applies to the Spacex Starship and its tanker versions, it applies to anybody else's vehicles, and it applies to orbiting fuel depots. All are in free-fall.

GW

SpaceNut Adding link
https://exrocketman.blogspot.com/2021/0 … tions.html

Last edited by GW Johnson (2021-08-18 20:01:12)

kbd512 · 2021-08-18 20:35:04

ULA's small inline six cylinder engine is the best technology available for providing power and tank pressurization control in an orbital environment. A piston engine is both simple and durable. The solar panels need to be of the thin film roll-out variety that NASA is actively working on. That origami nonsense is fine for a single deployment event, but there's no practical way to re-stow those types of solar arrays many times, because that was never a design consideration for satellites or interplanetary probes or space stations. In contrast, the thin film arrays are designed to be rolled out and rolled back up again, and thus represent a better design optimization. They're also much lighter per Watt of output power, if also much larger in total surface area for a given level of power output when compared to silicon wafers, but weight is an important consideration for what is essentially the upper stage of an orbital rocket.

louis · 2021-08-19 18:47:52

Matt Lowne's update:

https://www.youtube.com/watch?v=mw-ISkwCfE8

SpaceNut · 2021-08-19 19:09:24

yes using the boiloff to create power is a plus but with out the added heat exchangers, radiators and pumps the boil off problem will still exist.

GW Johnson · 2021-08-21 09:31:48

You have to remember that this is a prototype test. Only the things they want to test are included.

They need the quick-disconnect arms on the tower, but they need no "catch" arms (if indeed they ever do it that way at all). There are no landing legs on either stage, since both are intended to "land" in the sea.

They have to test the rocket engines (all 29 together in the first stage, plus 3 sea level and 3 vacuum in the second stage), they have to test the booster tanks and airframe, they have to test the stage separation, they have to test the second stage heat shield, and they have to test the hypersonic entry into the belly-flop maneuver. None of those things have been all-up tested yet. All are fatal problems if they don't work right.

They can get away with "only" 29 engines in the booster, because the stripped-down prototype second stage Starship is under-weight compared to what they eventually want to fly. Only 29 engines are needed to reach the approximately T/W = 1.5 they need for acceptable launch kinematics. (Once things are nearer operational configurations, and weights, you will see 35+ engines in the booster, probably 37 or 39 for a max weight flight.)

"Pending regulatory approval" is code for "we don't have a permit yet, and are at the mercy of the bureaucrats to get one". If they fly without that permit, they will never fly again from US soil. And likely no other country would permit them to fly, either. Musk knows that. So should all of you.

You usually have to look through all the hype to see what really is. With all of them, not just Spacex. The recent exception is Boeing's Starliner, whose faults became pathetically obvious despite any possible hype.

GW

Last edited by GW Johnson (2021-08-21 09:36:18)

Oldfart1939 · 2021-08-21 12:24:52

GW-
Musk has recently stated that there will probably be 33 uprated Raptor engines on the fully loaded and operational booster vehicle. The thrust and efficiency seems to be rising rapidly on the newer serial numbered engines, and is probably why he's willing to throw the older ones in the drink instead of attempting to recover them.

Last edited by Oldfart1939 (2021-08-21 12:29:39)

GW Johnson · 2021-08-21 18:43:58

Hi Oldfart1939:

I hope you're doing better now.

I had figured the sea level Raptor for 2 MN thrust at 4400 psia chamber pressure, on a 40:1 expansion ratio, at average 15 degrees half angle. To increase the thrust, either they have increased the chamber pressure or added to the expansion ratio. Or both. I dunno, haven't seen anything new about these engines that I can use as data, in a couple of years now.

The best data I have say the Superheavy will be near 250 metric tons inert mass, with a 3400 ton propellant capacity, and whatever the ignition mass works out to be for the Starship second stage.

The best estimates I have for Starship are 120 metric tons inert and a 1200 ton propellant capacity. My best spreadsheet-based estimates say about 150 metric tons of payload to 300 km circular LEO is consistent with recovering the booster at the launch site. That puts the ignition mass of Starship at 1470 metric tons, and thus the ignition mass of Starship/Superheavy at just about 5120 metric tons.

The Earth weight of 5120 metric tons is 50.21 MN. My original sea level Raptor estimates had max thrust 2 MN per engine at 4400 psia chamber pressure. 29 engines is 58 MN thrust for launch T/W = 1.155.

31 engines is 62 MN thrust, for launch T/W = 1.234

33 engines is 66 MN thrust, for launch T/W = 1.314.

35 engines is 70 MN thrust, for launch T/W = 1.394.

37 engines is 74 MN thrust, for launch T/W = 1.473.

39 engines is 78 MN thrust, for launch T/W = 1.553.

Musk has said in one of his public presentations that Spacex wants about T/W = 1.5 at launch. Which is why I thought 37 to 39 engines is just about right for the thrust estimate that I had for sea level Raptors.

If the thrust rating is going up, the number of engines can go down. But this is also quite sensitive to the stage inert masses they really are achieving.

GW

Last edited by GW Johnson (2021-08-21 18:44:53)

Oldfart1939 · 2021-08-21 22:43:07

GW-
I'll try to find the exact number that Musk stated about the thrust produced. It's a function of chamber pressure and it's up substantially over the earlier serial numbers. The Raptor isn't a mature design and is undergoing continuous upratings and weight per engine mass reductions. As he often states, the best part is no part. A lot of the plumbing appears more compact and smaller now.

Oldfart1939 · 2021-08-22 09:27:18

The interview of Musk by Tim Dodd is my source of the number of Raptors that will be used in later rockets being 33--or more. This was towards the end of either segment 2 or in the much shorter 3rd installment. Elon also states that the chamber pressure is quite high and results in high thrust.

GW Johnson · 2021-08-22 10:03:22

OF:

My data is about 2 years old now. Most of it came from earlier iterations of their website, and has disappeared from their site now. More recently, there has been useful data on Wikipedia. Not as reliable, though.

I did the interior ballistics and nozzle analyses for chamber pressures from 20-to-100% of Pc = 4400 psia. I did it from c* data for LOX-LCH4, with a best-estimate c* = K Pc^m model to get the increase from 1000 psia data to 4400 psia. I did it with a generic gamma = 1.2. It's pretty close. That was where I estimated max thrust at just a hair under 2 MN for a sea level Raptor, at their indicated exit size, and expansion ratio. My numbers rounded to 2 sig fig matched what they said at that time for max thrust.

I saw they were revising the rat's nest of plumbing. It also seems likely that they are uprating engine thrust, too, especially since that's what every rocket engine developer does. I don't know how they are doing it, though.

An increase in chamber pressure at the same throat area, with higher flow rate would do it. Raising the pressure allows you to increase the expansion ratio a little bit without risking backpressure separation, which also raises thrust. I have seen nothing that tells me what they are doing, though.

GW

New Mars Forums

Announcement

#1426 2021-08-12 11:49:05

Re: Starship is Go...

#1427 2021-08-12 12:52:30

Re: Starship is Go...

#1428 2021-08-12 13:52:49

Re: Starship is Go...

#1429 2021-08-12 15:33:21

Re: Starship is Go...

#1430 2021-08-12 15:47:47

Re: Starship is Go...

#1431 2021-08-12 16:17:09

Re: Starship is Go...

#1432 2021-08-13 07:46:43

Re: Starship is Go...

#1433 2021-08-15 13:10:50

Re: Starship is Go...

#1434 2021-08-17 14:49:01

Re: Starship is Go...

#1435 2021-08-17 17:34:46

Re: Starship is Go...

#1436 2021-08-17 18:01:06

Re: Starship is Go...

#1437 2021-08-17 18:37:02

Re: Starship is Go...

#1438 2021-08-17 20:32:07

Re: Starship is Go...

#1439 2021-08-17 21:03:56

Re: Starship is Go...

#1440 2021-08-18 18:42:46

Re: Starship is Go...

#1441 2021-08-18 19:59:54

Re: Starship is Go...

#1442 2021-08-18 20:35:04

Re: Starship is Go...

#1443 2021-08-19 18:47:52

Re: Starship is Go...

#1444 2021-08-19 19:09:24

Re: Starship is Go...

#1445 2021-08-21 09:31:48

Re: Starship is Go...

#1446 2021-08-21 12:24:52

Re: Starship is Go...

#1447 2021-08-21 18:43:58

Re: Starship is Go...

#1448 2021-08-21 22:43:07

Re: Starship is Go...

#1449 2021-08-22 09:27:18

Re: Starship is Go...

#1450 2021-08-22 10:03:22

Re: Starship is Go...

Board footer