Can't rule out negative acceleration by accident, but I suspect it would be deadly. As soon as LOX is in motion in the head space there will be ullage collapse to some degree in the main tank. Reapplication of acceleration would slam the bulk of the LOX back against the bulk head with resulting splash back especially if any turning moment was applied. Any significant ullage collapse would lead to pressure reduction and cavitation. And even if the header tank somehow avoided this fate the pressure loss in the main tank would have robbed all of the advantages of hot staging as vast quantities of cold compressed (heavy) gas would be required to repressurise it.I've been noodling the issue of the three engines staying lit while purportedly experiencing negative acceleration. You know a lot more about this than I do. Could you (and everybody else for that matter) critique an idea?And opposing the idea is that the centre engines remained lit throughout hot-staging, and all but one of the centre 13 relit after hot-staging had completed and the flip started. And that the entire reason hot-staging was implemented was to keep booster prop settled through the staging sequence without needing to re-pressurise the tanks or resettle prop.It's not a certisnty, but there's certainly supporting evidence, both in the vented gasses and in the slosh simulation, and also force analysis supports the idea.This means the SH fuselage will experience a brief slam of negative acceleration relative to its own prop.
I was disabused of this notion back about 2,000 pages.
Without more telemetry data, or a much more serious analysis, all we can say is that the acceleration of SH was REDUCED from the pre-hot staging acceleration. We really don't have enough information to say that it we reduced far enough to go negative. As long as the acceleration remains at least slightly positive (say, went from +3G to +0.01G), the propellants will remain settled at the bottom of the tank. This is the simplified view; one would also need to take into account the rotational acceleration to truly model the behavior of the propellants, but that kind of work makes my head hurt.
Whatever happened that caused the last engine in the boostback reignition sequence not the light, and the lit engines to start shutting down, occurred well into the post-staging flip.
For the sake of argument assume there really was a negative acceleration and let's only look at the earliest moments before the flip starts. And let's further assume the methane header holds enough that, at least initially, methane starvation is not an issue. We'll concentrate on the LOX.
The simplest of model has the LOX moving towards the common dome as a simple cylinder of fluid. But there is a force moving a portion of the lox towards the engine inlets. Inertia. It wouldn't be applied to the entire fluid cylinder. It would mostly be in the center around the downcomer and above the three center engine inlets. It would also probably be in the form of three vortexes which (I think) would help keep the inertia effects strongly localized.
The simple cylinder of fluid would most likely transform into a rough cone, apex to the inlet area with a fluid cylinder above that. This would be a strictly transient effect with a full detatch inevitable if the negative acceleration continues. Once the flip starts the LOX would go chaotic.
If I've noodled this correctly, the three center engines could keep running for a short time while the LOX was otherwise reacting to the negative acceleration. As long as we saw them running? I do not know.
If my surmise is realistic we can't rule out negative acceleration.