Problems compressing the helium tanks inside the F9 2nd stage LOX tank at the last minute?

A recent "Anomaly Update" dated September 23, 1:00pm EDT by SpaceX, includes the sentence:

"At this stage of the investigation, preliminary review of the data and debris suggests that a large breach in the cryogenic helium system of the second stage liquid oxygen tank took place."

This has been repeated in several news outlets, but spaceflightnow adds:

The Falcon 9 rocket’s upper stage liquid oxygen tank contains several composite helium vessels, each pressurized to about 5,500 pounds per square inch in flight. The helium is routed through the second stage’s Merlin engine, where the helium warms up and is injected into the rocket’s propellant tanks to pressurize the stage as the launcher burns fuel, keeping the tanks structurally sound.

While cryogenic helium was aboard the Falcon 9 at the time of the explosion, the mishap occurred around eight minutes before the rocket’s main engines were scheduled to ignite for the on-the-pad “static fire” test Sept. 1.

At that point in the countdown, the propellant tanks are normally not pressurized for launch.

I'm trying to understand the implications behind: "The Falcon 9 rocket’s upper stage liquid oxygen tank contains several composite helium vessels...". According to this answer "In the move to Falcon 9 v1.1 and the Merlin 1D engine they moved them into the LOX tanks on both stages. This de-cluttered the engine area a great deal." - mentioned in this answer as well - I can see that the helium tanks have been relocated inside the LOX tanks for several practical reasons including layout/arrangement and thermal management - inside the LOX tanks the temperature is much colder than ambient.

At a given maximum pressure, a lower temperature allows for more helium to be stored in the tanks, or a smaller/lighter tanks can be used for a fixed amount of helium, although I am not sure the actual reason. Boiling LOX is about 90K (only about one third of ambient absolute temperature), and SpaceX is (often/usually?) subcooling below that to pack even more LOX into the tanks.

1. Is the helium now substantially colder by moving the tanks inside the LOX tank, or is it simply easier to keep them cold - avoiding coolers and insulation?
2. If the helium tanks are pressurized only in the last minutes before launch, doesn't compressing this gas to roughly 400 atmospheres generate a huge amount of heat? If the tanks are fixed size, the pressure will increase steadily as it is filled, and that makes a heat source inside the LOX tank just before launch. I can understand adding the helium beforehand, letting it cool a bit after compression, and then filling with LOX, but wouldn't compressing the helium in the tank at the last minute be incompatible with sub-cooled LOX?
3. Or, can "...each pressurized to about 5,500 pounds per square inch in flight." be taken literally - the tank becomes pressurized in flight? The pressurization occurs in flight?
4. And does "At that point in the countdown, the propellant tanks are normally not pressurized for launch." refer to LOX and RP-1 but not helium? Is the helium pressurized but not the propellants, or does this also refer to the helium being not pressurized?

Consider for a moment the NORS 6000 psi oxygen tanks for delivery to ISS as described in this answer. According to that article:

Working inside the high bay of the Space Station Processing Facility at Kennedy, about a dozen people perform the operations to fill a single tank following two to three days of inspections and work stand preparations. They work about 40 feet from the stand with the tank and leave it alone for a day so it can cool down after the pressurization. (my emphasis)

The team performed several fills on other tanks to make sure the process could be handled safely and performed as planned.

"There's really no other place in the world that operates up to 10,000 pounds per square inch like this in an operational environment," Bigos said. "We've had to be exceptionally cautious. It's not yet normal day-to-day operations but we're getting there."

It doesn't sound like you'd want to pressurize a 6000 psi tank submersed inside a sub-cooled LOX tank just minutes before liftoff - due to the heat generation if for no other reason. The helium tanks may or may not be smaller, but this is not a location where you'd want to be generating much heat.

above: NORS oxygen tank from here. Credit: NASA, cropped/rotated.

I know it's just the color and general shape, but just looking at that makes me nervous - I think it's reminding me of these for some reason.

You can always top-off a boiling LOX tank at the last minute, but I don't think you can really re-sub-cool a LOX tank inside of a rocket. Heat introduced into sub-cooled LOX will raise its temperature and the LOX will expand and its density decrease, unless there are additional "chillers" in there as well.

However in any case I now understand that the pressure of the helium is still used outside of the LOX tank and that they were moved inside for other reasons.

• This space news report on their in-flight incident includes an explanation of the tank-within-a-tank arrangement. There are apparently multiple helium tanks within the LOX tank. – Anthony X Sep 24 '16 at 4:06
• The helium tanks are submerged in the LOX tank. This is not new, here is a picture of the helium tanks inside the first stage LO2 tank of the Saturn V scontent.fhou1-1.fna.fbcdn.net/v/t1.0-9/… – Organic Marble Sep 24 '16 at 4:23
• They are pressurized from the ground systems. The temperature in a LOX tank is nowhere near low enough to get helium to liquefy. – Organic Marble Sep 24 '16 at 4:24
• @OrganicMarble amazing photo - quite a reminder that not all rockets are 3.66 meters in diameter :-) – uhoh Sep 24 '16 at 5:01
• Liquid helium is a lot colder than liquid oxygen. (He 4.2 K, O2 90.15 K) If there is a tank for liquid helium inside the tank for LOX, the helium tank should be insulated, otherwise a lot of helium will boil of and the oxygen at the outer surface of the helium tank will solidify. To get a light helium tank that contains a lot of helium, liquid helium is superior to high pressure gaseous helium. Warming up the cryogenic helium near the engine could be the temperature difference from 4.2 K to about 90 K, from liquid helium to gaseous helium at the temperature of LOX. Heating the helium a lot – Uwe Sep 26 '16 at 13:52

1. It's much easier to cool the He by placing the tank inside the LOX tank. If you had a He tank outside, you'd have to insulate it, and provide active cooling. Both add weight and complexity.
2. Propellant tanks can be pressurized at the last moment: you fill them at atmospheric pressure and vent the boiloff (if any). For boiling-point LOX: when the tank is full, you close the vent valve until the boiloff pressurizes the tank to the value you want. For subcooled LOX you close the vent valve and use He to pressurize the tank. Filling at atmospheric pressure works because LOX is a liquid. You can't do this with a tank that contains gaseous He.

If they filled the He tanks during propellant loading, they'd introduce a large heat source into the subcooled LOX tank, which would lead to huge amounts of boiloff. Still, that's what they do, according to the user guide. "Early in the countdown" can be as little as 1 hour before launch.

Early in the countdown, the vehicle performs LOX, RP-1 and pressurant loading,

3. No.

4. He is not a propellant. It only refers to RP-1 and LOX.

This countdown sequence confirms: He and LOX are loaded simultaneously at T -30 minutes.

• So for item 2, if that were the case, they would be cooled to roughly LOX temperature during the filling to 5500psi? That means the pressure would triple if they reach ambient? (of course there'd be several overpressure release mechanisms so that shouldn't really happen). I'm just thinking that a couple of tanks filled to 5500 psi at <100K are somethings you'd really better keep cold once filled. After filling, while the rocket is moved and before fueling, there must be some reliable cooling mechanism. – uhoh Sep 24 '16 at 8:31
• Based on 3 and 4, maybe the line "At that point in the countdown, the propellant tanks are normally not pressurized for launch." is a bit of a red herring - the helium tank pressure may be the relevant pressure, not the pressure of the propellants. – uhoh Sep 24 '16 at 8:37
• Where did you find that - does it apply after the helium was moved inside? From this answer "In the move to Falcon 9 v1.1 and the Merlin 1D engine they moved them into the LOX tanks on both stages. This de-cluttered the engine area a great deal." – uhoh Sep 24 '16 at 9:06
• I just found this in spaceflight101.com/spacerockets/falcon-9-v1-1-f9r "Generator gas flows through a heat exchanger which heats up Helium gas for tank pressurization in flight before being dumped overboard through an exhaust." Helium is discussed in several places there. Perhaps helium is pressurized in flight after all, and not actually sitting static at 5500 psi. – uhoh Sep 24 '16 at 9:17
• The LOX and RP tanks are pressurized to something like 5 bar, not 370 bar. The He tank has to contain enough gas to fill the entire LOX tank at 5 bar. To keep the size of the He tank down, they'll want the pressure to be as high as possible. – Hobbes Sep 24 '16 at 9:54

The He is stored either as a very compressed gas (high pressure tank) at LOX temperature OR it is stored as a even colder 4K liquid (moderate pressure dewar).

The amount of He must be sufficient to expand or vaporise to fill empty LOX and fuel tanks to flight pressure. Remember that liquid He takes up a lot less volume than compressed He.

Either technique will work:

• if using a compressed He gas tank then service valve is closed until LOX/fuel tanks need pressure in flight, He filling can take place at any time and pressure is ready immediately. Some heat transfer may occurs during filling depending of speed and liquid/gas filling method. During flight the pressurised gas is heated in the engine to prevent the introduction of He into tanks at low temperatures caused by adiabatic (Joule-Thompson) cooling from causing LOX/fuel to freeze and block pipes.

• if liquid He dewar then filling can occur before, during or after LOX/fuel filling. There will be no heating when liquid He is filled, rather some cooling will occur as the inside of the dewar reaches 4K. The boil-off is vented until tank pressure is needed at which point vent is closed and boil-off will pressurise to LOX/fuel tank to working pressure. During flight the boil-off speed is maintained by directing the liquid He though the engine to boil it and maintain the pressure in the LOX/fuel tank vapour spaces.

The heating of the He will provide some welcome cooling of the engine for no additional cost.

Both options gain a benefit from placement in the LOX tank.

• The compressed He option gains from space saving as the geometric voids around a high pressure tank are filled with LOX and not wasted space.
• The liquid He option gains from having free primary temperature reduction around the dewar to LOX temperatures while on the pad and being able to use liquid He in a thin wall dewar because the differential internal pressure in the dewar is only that which is needed to feed the liquid He through the engine based heat exchanger.

The alternative of an external tank requires very strong tank for high pressure gas or strong well insulated dewar for liquid.

The low temperature of the expanding He would be a potential problem in the fuel tank as it would freeze around the dewar and pipes in flight and the benefits of the cryogenic LOX temperatures are lost if the liquid He option is used.

Many of the comments in this thread point to the fact that the He in the LOX tank is liquid and this would have the most benefits from moving the He storage inside the LOX tank.

EDIT:
It is clear that the Helium in the tank is initially colder than the LOX. It also appears that the He tanks are of complex structure with an aluminium internal liner surrounded by a composite strength member. The problem during filling seems to have been the cold rather than heat. The He tanks seem to be able to handle high pressures rather than large temperature differences. It appears that they are filled with liquid He and then hold pressurised He as they heat up to LOX temperature. The slower filling technique that has been proposed may use pressurised He to fill them causing MORE heat load rather than less but perhaps avoiding the suspected solid O2 formation. https://www.geekwire.com/2017/spacex-falcon-failed-helium-tank-iridium-jan-8/

• This answer seems to be entirely about the general benefits of putting a helium tank inside a LOX tank, but the question is specifically about SpaceX's particular practices and unusual countdown schedule. – Nathan Tuggy Jun 15 '17 at 23:10
• @NathanTuggy That is true, however there seems to be so much confusion in the comments about how the systems work that having the two options reasonably clearly defined will make all other comments and answers easier to read. It is also pretty clear that the High Pressure tanks are no longer required if there is a liquid He dewar in the LOX tank. It is possible that journalistic enthusiasm has confused the pressurised vs liquid He systems. – KalleMP Jun 15 '17 at 23:21
• Clarifications to a question do not belong as answers, whatever their merit otherwise. And anything that must be read before various answers can be properly understood is certainly not an answer. – Nathan Tuggy Jun 15 '17 at 23:40
• Downvoted until you provide a reference explicitly stating that liquid helium is used. – Organic Marble Jun 16 '17 at 0:58
• There is nothing intrinsically wrong with asking a question and then answering it yourself. You could consider asking a new question to which this is the answer, and then posting this as the answer there instead. Both would be subject to voting of course. Or you could just ask a question of the form "Has liquid helium been used in..." and that could generate a better space for exploring the question. That might be a really interesting way to go. – uhoh Jun 16 '17 at 3:25