Intelsat Signs First Commercial Falcon Heavy Launch Agreement with SpaceX

[JNobles]

It has been previously mentioned that Falcon Heavy will not always use cross feed,  would bet that this launch will only need a tricore design rather than relying on an unproven technology.

I would be surprised to see them launch the first FH without testing some of the crossfeed hardware.  If they don't test that what would they be testing?  Just the ability to tie the cores together, trying to fire 27 engines at once, and the new avionics?  I know that seems like a lot but SpaceX is not particularly timid.  I would indeed be a little surprised if crossfeed was not tested to some extent.

[Joel]

Speaking of the 27 on an actual FH, does anyone know if SX has any plans to produce a larger engine so they need fewer of them? I know it has engine-out capability, I know bigger engines are harder, but still... 27 is a lot ...

I keep reading this over and over again. Being a mathematician I cannot for the world understand why fewer engines would be any safer. Having three engines with the ability to losing one engine (a third of the thrust) should be compared to losing 3 engines for a stage with 9 engines.

Doing the maths is simple if you know a bit of probability. If the probability of engine failure if 5 % (and multiple engine failures can be considered independent events).

Losing more than one engine out of 3 can be easily shown to be:

1-((1-p)^N + binom(N,1)*(1-p)^(N-1)*p)

with p=0.05 and N=3 and where binom(n,k) is the binomial function "n choose k".

Evaluating this will give you a 0.73 % risk of loosing more than a third of the thrust.

Similarly, losing more than 3 engines out of N=9 can be easily shown to be:
1-((1-p)^N + binom(N,1)*(1-p)^(N-1)*p + binom(N,2)*(1-p)^(N-2)*p^2 + binom(N,3)*(1-p)^(N-3)*p^3)
 
Evaluating this will give you a 0.064 % risk of losing more than a third of the thrust, so more than 10 times safer.

Why would having less engines ever be more safe?

[Jim]

 should be compared to losing 3 engines for a stage with 9 engines.

That is false.
a.  during early portions of the flight, losing one engine is fatal.
b. losing two is always fatal

[rklaehn]

Why would having less engines ever be more safe?

The thing is that a certain fraction of engine failures will be catastrophic, so the engine failure will take out the neighboring engines.

What number of engines is optimal from a safety point of view depends on the value of this probability. If you think that the majority of engine failures will be catastrophic, it is not possible to have engine out redundancy at all. If you think that the vast majority of failures will be benign, it makes sense to use multiple engines for redundancy.

There are a lot of people that think that modern rocket engines can usually be shut off before exploding when they develop problems. And there have been many cases (on both saturn and shuttle) where missions were completed with engines shut off. So personally I think that for a well-characterized engine with good instrumentation (chamber pressure, turbine speed etc.) it will be possible to safely shut it down most of the time.

[Joel]

 should be compared to losing 3 engines for a stage with 9 engines.

That is false.
a.  during early portions of the flight, losing one engine is fatal.
b. losing two is always fatal

I meant in general, not for F9 explicitly. Having more engines should make you safer (if there is engine-out capability). I don't see why losing one out 9 (or one out of 27) would have to be fatal. Especially if you have extra propellant available that you would normally use for a propulsive landing. I mean, you can decide to save the payload instead of the launch vehicle.

[Joel]

There are a lot of people that think that modern rocket engines can usually be shut off before exploding when they develop problems. And there have been many cases (on both saturn and shuttle) where missions were completed with engines shut off. So personally I think that for a well-characterized engine with good instrumentation (chamber pressure, turbine speed etc.) it will be possible to safely shut it down most of the time.

So with that logic, given modern instrumentation and control algorithms that shut down engines early and safely when they start to behave abnormally, my reasoning would hold, right?

[JBF]

I meant in general, not for F9 explicitly. Having more engines should make you safer (if there is engine-out capability). I don't see why losing one out 9 (or one out of 27) would have to be fatal. Especially if you have extra propellant available that you would normally use for a propulsive landing. I mean, you can decide to save the payload instead of the launch vehicle.

Now this is an interesting idea. The question becomes would it be possible to make this sort of decision on the fly or would it have to be programed into the flight computer.

[MP99]

Agreed, interesting.

Would suggest re-posting to http://forum.nasaspaceflight.com/index.php?topic=27748.0 for further discussion.

cheers, Martin

[Joel]

Agreed, interesting.

Would suggest re-posting to http://forum.nasaspaceflight.com/index.php?topic=27748.0 for further discussion.

cheers, Martin

Ok, done.

[Jim]

Especially if you have extra propellant available that you would normally use for a propulsive landing

Not a near term capability.

[fatjohn1408]

In the launch manifest it recently slipped from 2015 to 2017. Any reasons for that?

[smoliarm]

In the launch manifest it recently slipped from 2015 to 2017. Any reasons for that?

This 2017 date for FH Intelsat has been there for a month at least, may be two. As I recall it occurred after SpaceX finished acceptance testing for the first F9 v1.1 (I do not imply any link, just a time-mark)