We're all aware of the recently aborted SpaceX Demo-2 launch... While talking with friends we discussed the venting of LOX pre-launch.

I presume that the reason for this venting is largely to achieve (or maintain) a "full" tank of Liquid Oxygen. In part, I presume it's likely easier to permit some to boil off and replace it with additional cooled LOX, rather than try to cool and pressurise a closed tank "first time" - which blatantly isn't going to happen within the design's constraints.

Is this even a somewhat correct presumption?

Either way, the fact that some LOX is vented is undeniable.

Do we know how much is pumped beyond the ~287,430 kg capacity of the core and booster's tanks? or in other words: "How much is wasted?"

I appreciate that this figure will likely be significantly different to other models (for example, those that don't use sub-cooled LOX), and I'd be happy with any an answer, not necessarily specifically related to this launch.

During the conversation, I toyed with a ~130% figure (i.e: ~30% wasted), but honestly I pulled that out of thin air. If this isn't something that we know (perhaps it isn't public information), then I'd appreciate an informed estimate.

• See the related question. – Uwe May 28 '20 at 20:50

Since "I'd be happy with any an answer, not necessarily specifically related to this launch." here are some shuttle numbers.

Once the External Tank (ET) was filled with LO2, it went into replenish mode in which the system replaced boiled off propellant. These charts show that the nominal replenish time was about 4 hours (including a built-in one hour hold) and the nominal replenish flow was about 1.5 lbm/s, giving ~21,600 lbm of LO2 boiled off.

This is out of a nominal load of ~1,400,000 lbm LO2.

Source: personal notes (MPS Integration work in Shuttle SE&I office)

• Very interesting, thank you! Your numbers would indicate ~15% of load was spent again during the replenish... but if I'm reading this correctly, I came to ~16,200 lb boiled off (~3 hours, at ~1.5 lb/s ... not-quite-T-3h to not-quite-T-0 = ~3h ... 60 * 60 * 3 * 1.5), and your figure is 10x what I came to for 4h...? ~16,200 lb (~21,600 lb) boiled off vs. noninal load of 1.4e6 lb is ~1.15% (~1.5%), which sounds much smaller than I'd expected. I imagine it would be somewhat higher for this launch, but not nearly ~30%. – Attie May 28 '20 at 21:39
• @Attie The graph is a bit confusing because it shows T-minus numbers which are countdown clock numbers. Not L-minus numbers which are wall clock numbers. So there is an additional hour in there - you can see the black arrow pointing at it and saying "1 hr hold". You have to account for that hour as well. But you are right about the 10x, thanks, I have corrected it. I read it off the calculator wrong! – Organic Marble May 28 '20 at 22:12
• Understood re countdown clock... interesting to hold for an hour at T-3! Thanks for confirming the x10 error. – Attie May 28 '20 at 22:17
• @Attie thanks for checking my work! – Organic Marble May 28 '20 at 22:19
• @Attie I don't seem to find a description of the LO2 loading on the site here but there is one for LH2 here - different in details but similar in big picture space.stackexchange.com/a/34977/6944 Venting of LO2 happened through the ET vent valve - that was open for the slow fill & closed for the fast fill (most loading happened in fast fill), reopened for topping & replenish. I picked replenish because that was a long stage, and the only one I could quantify the boiloff for. Venting would be happening any time the vent valve was open. – Organic Marble May 29 '20 at 11:43