Переохлаждение компонентов.

[Salo]

https://www.reddit.com/r/spacex/comments/3xkk6g/propellant_densification_and_f9_v11_to_v12/

Propellant Densification and F9 V1.1 to V1.2 Evolution (self.spacex)
отправлено 8 часов назад автор 2p718

It appears that LOX densification has a significant payoff. Cooling LOX from its NBP (Natural Boiling Point) of 89.8K down to 66.5K increases its density by 9.7%. That is a big win! These figures are from Liquid Oxygen Propellant Densification ... for the X33 RLV.
The payoff for RP1 is about 2% for cooling it from 20degC to -6.7degC. Cooling RP1 rapidly increases its viscosity, so going even lower might not be possible. These figures are from data for Kerosine, RP1 should be pretty close).
Assuming F9 V1.1 with 300t of propellants and a LOX/RP1 ratio of 2.56, that would be 216t LOX and 84t RP1. Densification with the published temperature figures would raise that to 236t LOX and 85.7t RP1 in the same tank volumes. To retain the LOX/RP1 ratio of 2.56 the tank volumes would of course have to be adjusted.
We already know that the F9 V1.2 has been stretched to accommodate larger tanks and AFAIK it has 30% more thrust, some of which is needed to propell the increased propellant mass.
Looking at the changes from V1.1 to V1.2 I get the impression that this is a rather bold and big step to take and not at all cautious and incremental.
Some of the questions that pop into my mind are:
 

[/li]
• Was the first stage substantially redesigned or strengthened to cope with the greater forces?
• What is the effect of the lower LOX temperature on thermal stresses and metal embrittlement?
• Can the rapid expansion of LOX potentially lead to it freezing? (LOX freezing point is 54.4K).
• A lot of things cannot be tested on the ground, e.g. dynamic loads in flight, thermal behaviors in diminishing ambient pressure, etc... So, how confident can SpaceX really be that the significant changes it made will not cause unexpected problems in flight?
  

//[–]FoxhoundBat  5 часов назад*

We already know that the F9 V1.2 has been stretched to accommodate larger tanks and AFAIK it has 30% more thrust, some of which is needed to propell the increased propellant mass.
 

No, the engines produce 15.6% more thrust (on S1), not 30%. The sum of all the upgrades (15.6% more thrust, stretched S2, higher thrust + Isp on M1D Vac, densification, etc) all in total give 33% greater payload performance. S1 is not stretched, S2 is by 10%. I have heard murmurs about possible very tiny S1 stretch done on v1.1FT but as i haven't seen any proof i think it can be discarded until proven otherwise.
LOX is densified about 9.9% and RP-1 is 2.6%. And if you think changes between v1.1 and v1.1FT are bold, they are much greater between v1.0 and v1.1. FT is an evolution of v1.1, v1.1 was a completely new rocket vs v1.0.

//[–]scarlaton  6 часов назад 
https://twitter.com/elonmusk/status/677666464208416768