Author Topic: Q&A: How fast does SLS's hydrogen fuel boil off on the launch pad?  (Read 1073 times)

Offline Shevek23

  • Full Member
  • *
  • Posts: 139
  • Liked: 43
  • Likes Given: 91
I think I did due diligence searching for this question being asked already somewhere, or answered by the way, given the limits of the search function as it seems to work for me (not well). Also I have searched generally on the Internet and this specific fact has not been specified anywhere I found.

Closely related would be hard facts about the Shuttle ET as they had the same diameter; the length of side wall exposed to atmospheric heat is greater, apparently by 25 percent or so, for SLS, and perhaps SLS insulation is significantly different for a number of possible reasons.

I have happened to stumble upon some remark saying STS Insulation ranges between 1/2 and 2 inches thickness, and masses about 4-5 tonnes of the total STS core structure of 85 tonnnes. I believe about 40 more of the core stage structure dry mass would be for the 4 RS-25 aka "SSMEs" and their thrust structure, plumbing etc, leaving about 40 for the actual tank itself which seems fairly reasonable given that this tank needs in-line strength the STS ET did not. Any fining down of these sorts of mass breakdown or details as to insulation distribution would be much appreciated too! I'm guessing the forward LOX section is mostly 1/2 inch foam with some thicker on the ogive nose cone to resist aerodynamic heating when ascending, and that the LH2 section behind the intertank station is covered in 2 inch thick foam.

I am aware of buzz about super-heroic efforts to "densify" the hydrogen with superchilling and so on, but what I am fishing for here is an empirical magnitude for rate of heat flow between the outer surface of the SLS tank specifically (again STS tank would be somewhat relevant if insulation performance is in the same close ballpark).

I presume that shortly before launch, the SLS hydrogen tank, on the bottom of the core tank structure as with STS, is topped off minus a bit of ullage volume for evaporated GH2 at cryogenic temperature or a bit higher. The tank will be full of LH2 at about 18-19 K, and its walls will be boiling as heat from the atmosphere seeps in due to the temperature differential.

If I know how much hydrogen total is boiling per second, minute or whatever time unit is used, either in liters, cubic feet or whatever of gaseous exhaust being tapped off to keep pressure within bounds, or vice versa mass flow which would be ideal, either of cold gas coming out or replacement LH2 being pumped in, on the ground just before launch, then I can estimate the wattage of heat altogether flowing into the tank from the air.

In these circumstances the main source of heat to boil off the hydrogen is through the sidewalls, from the atmosphere, which I suppose flows down the stage exterior rather briskly due to being chilled to the depth of some boundary layer well below ambient (it rarely gets down to freezing in Florida, whereas frost forms on the exterior of the STS and presumably SLS tank so we know that boundary layer is colder than 271 K) and thus bringing in much more heat by convection.

I want to know the boiloff rate on the pad mainly so I can then attempt to estimate what rate it would have in orbit, in vacuum in LEO. That would depend on how it is oriented to the Sun, the Earth, and whether there is LOX in the forward oxygen tank or not, but I think I can handle this moderately well if I have an empirical figure for temperature drop from outside to inside in a given circumstance, and launch from the lower atmosphere soup of Cape Canaveral seems like a good benchmark to know.
« Last Edit: 08/22/2021 10:14 pm by Shevek23 »

Online AnalogMan

  • Member
  • Senior Member
  • *****
  • Posts: 3350
  • Cambridge, UK
  • Liked: 1448
  • Likes Given: 44
Not sure if this will answer your question, but this relates to losses during stable replenishment of the shuttle ET:

Stable Replenishment Loss
After the ET is full, the cryogenic loading systems transitions into stable replenishment. The cryogen loss
during stable replenishment is due to the heat transfer through the ET insulating foam, heat transfer into the
cryogenic propellant transfer system, and continual SSME conditioning. Approximately 380 liters of LH2 and
475 liters of LO2 flow from the ground storage tanks every minute during stable replenishment. The length of
stable replenishment has varied over the SSP, and depends on the amount of work required to be performed
such as loading astronaut crew and tank inspections as well as the launch window. Ultimately, stable
replenishment occurs until a scrub is declared or until about five minutes before launch. The average total
stable replenishment is approximately 185,000 liters and 190,000 liters for LH2 and LO2, respectively.


This is from page 3 of the attached paper.

Offline Shevek23

  • Full Member
  • *
  • Posts: 139
  • Liked: 43
  • Likes Given: 91
Not sure if this will answer your question, but this relates to losses during stable replenishment of the shuttle ET:

Stable Replenishment Loss
After the ET is full, the cryogenic loading systems transitions into stable replenishment. The cryogen loss
during stable replenishment is due to the heat transfer through the ET insulating foam, heat transfer into the
cryogenic propellant transfer system, and continual SSME conditioning. Approximately 380 liters of LH2 and
475 liters of LO2 flow from the ground storage tanks every minute during stable replenishment. The length of
stable replenishment has varied over the SSP, and depends on the amount of work required to be performed
such as loading astronaut crew and tank inspections as well as the launch window. Ultimately, stable
replenishment occurs until a scrub is declared or until about five minutes before launch. The average total
stable replenishment is approximately 185,000 liters and 190,000 liters for LH2 and LO2, respectively.


This is from page 3 of the attached paper.

That's superb--probably close enough for non-government, freelance work anyway. I'm surprised the LOX rate is greater than the LH2 rate since 1) the temperature differential between outside and inside is far less and 2) the surface area exposed to atmospheric heat bath is less as well. It might come down to hydrogen having a much higher heat of vaporization (don't know if it does, I haven't looked it up yet, and if that is given specific to mass it is a lot less dense which is one reason I'd expect the volume flow for H2 to be higher not lower so heat of vaporization specific to mass would have to be a lot higher), might relate to the possibility that as I guessed for SLS, the LOX tank has much thinner insulation overall, or that the air encountering the LOX tank on top is a lot warmer than the pre-chilled stuff flowing down lower. Also, while 18 K is a hell of a lot colder than 90 K in many relevant applications such as achieving cryogenic re-condensation in space (much higher heat pump load, much more difficult waste heat rejection) the difference in temperatures between the exterior and interior is what matters here, in terms of heat flux through a given thickness of insulation, and say average 270K (on the LOX tank nose) minus 90=180 is not a lot less than say 250 (wild guess average of the cold air flowing along the lower hydrogen tank) minus 18= 232. Those differentials are pretty close actually.

In space, the absolute temperature makes a lot more difference because the input heat is coming from sources which can be shaded (no heat shield stops heat flow completely but it sure does trim it down a lot) and most of the sky is void at the background 3 K of the universe, so the main thing governing net balance of heat flowing in versus out is the heat sink temperature. If that is high enough so black body radiation out matches source radiation coming in, the temperature will not rise. On Earth the source is the mass of heat in the air, only somewhat moderated by boundary layer chilling, and pervasive. That's why I wanted to know about conditions on the pad, because it is a worst case scenario (short of launching our tank to a close encounter with the Sun or some such).

In space, if we could refill the tank there (that's the application I am aiming for!) I think that by turning the warmer LOX nose toward the Sun and putting a wide enough parasol over the nose to guarantee no direct sunlight on the trailing hydrogen tank, the two competing heat sources in LEO would be heat reflected from Earth in daylight and radiated by Earth' black body output when on the dark side, which are significant since Earth fills half the sky, and the 90 K LOX tank end radiating onto the 18 K hydrogen tank end; I suspect the latter might be higher and anyway one could put a secondary shade over one side of the tanks or actually we'd need on both sides, because if we have the nose facing the Sun always while in LEO, the Earth will appear relative to that body orientation to revolve all the way around once every orbit.

Boil off will happen in space, of the hydrogen anyway (we might be in luck with the LOX, if net input heat averages less than what the LOX tank radiates at 90 K over an orbital cycle). But surely at a rate far less than you cite and by golly I think I can work with those figures so thank you very very much!

 

Advertisement NovaTech
Advertisement SkyTale Software GmbH
Advertisement Northrop Grumman
Advertisement
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1