SpaceX Falcon 9 FT - ORBCOMM-2 - Dec. 21, 2015 (Return To Flight) DISCUSSION

[guckyfan]

One question I have about all this expectation of tear-downs and forensic examinations is how much it will cost? The manpower and time for what may prove a fruitless search might be a good reason not to do it.

Manpower/time that is only spent once/twice, and MIGHT prove to be fruitless, or MIGHT find something very important. Fruitless is still fruitful if it proves everything works.

I wonder if they cannot get the needed info from nondestructive means. Ultrasound or magnetic tests before and after. How about repeating the structural tests they did for acceptance in the test stand at McGregor? Maybe destructive tests on a very few locations simulations show to experience the biggest stress.

[ratman]

There is no such thing as "venting" from the bottom of a tank.  And still if you remove the LOX, the remaining gas is a mixture of O2 and He.  To recover the He, the gas mixture will be need to be sucked/displaced out of the prop tank.  It would need to be compressed for transportability.  The gas mixture will need to be take to a gas liquidification plant to distill out the He.

The removal ops would not fit in with Spacex turnaround times,  since it would be time consuming.
Just remove the RP-1 and open the all the vents.

Spacex is just going vent the He.  Quicker and cheaper.

Solution:
1. vent it (from the top) into a large helium balloon.
2. drag the balloon to the nearest party store
3. fill great many party balloons and give them away to the local kids

[meekGee]

Funny.  But.

It belongs in a lab, in pieces, to expedite searching for any close-calls or low-margin situations with the rocket.  Destructive testing as well.
…

One question I have about all this expectation of tear-downs and forensic examinations is how much it will cost? The manpower and time for what may prove a fruitless search might be a good reason not to do it.

The better plan could be 'test as you fly' i.e. after checking for gross damage they do a static fire. It is a procedure they do a lot, does not require special expertise, tests engines and structure, and errors will be easily apparent.

Keeping the first stage returned for posterity is pretty sentimental but if that is the price for having a visionary CEO who has got the company this far and could take them much further then we should be willing to bare it.

There is no such thing as a fruitless search in this case, because nobody has ever achieved this before and so nobody knows fully what will happen. We have models of course, but there will always be inaccuracies and the data they get will refine their models.

Exactly.  It's like saying "why did you participate in the race if you didn't win?"

Just like trying several approaches, fully knowing that some might be dead ends and some just plain inferior - that's not "wasting money" - that's research.

[chipguy]

At the time this was explained as needing to have a constant thermal emissivity so that IR imaging could obtain correct temperatures for every surface point.  Otherwise you have to calibrate for the emissivity of every different surface separately.

I don't think it's related to soot.

With the right paint (higher IR emissivity), you can effectively cool via radiation. One reason why electronic heatsinks are black anodized rather than polished aluminum.

If your semiconductor heatsink is cooling appreciably via radiation compared to convection then you
are having a very bad day.

Most heavy duty heatsinks I see are strip or extruded aluminum, albeit matte rather than polished.

In this forum I am obviously talking about near-space/vacuum conditions.

The semiconductors and their heatsinks aren't looking at free space to radiate heat to.

For this type of application the ICs and other components typically contact custom machined/stamped
conduction plates bonded to the PWB. The heat is conducted through the heat plate to the edge of the
boards where it conducts to the assembly case through metalic wedge locks or similar arrangement.
The case itself is cooled by conduction to a cold plate which is actively or passively cooled by a vehicle
level solution.

[Elmar Moelzer]

I wonder if they cannot get the needed info from nondestructive means. Ultrasound or magnetic tests before and after. How about repeating the structural tests they did for acceptance in the test stand at McGregor? Maybe destructive tests on a very few locations simulations show to experience the biggest stress.

There are a lot of ways to do non destructive testing and I have been wondering which ones they are using at SpaceX.
There are micro/nano- CT- scanners that can scan pretty big parts, these days. Then there is Xrays, Ultrasound tactile and laser sensors, etc, etc. They have a lot of options to test parts without destroying them. For the infamous bolts, I have been wondering whether things like the grain structure could have been inspected with a nano- CT scan without destroying the bolt.

[JasonAW3]

I wonder if they cannot get the needed info from nondestructive means. Ultrasound or magnetic tests before and after. How about repeating the structural tests they did for acceptance in the test stand at McGregor? Maybe destructive tests on a very few locations simulations show to experience the biggest stress.

There are a lot of ways to do non destructive testing and I have been wondering which ones they are using at SpaceX.
There are micro/nano- CT- scanners that can scan pretty big parts, these days. Then there is Xrays, Ultrasound tactile and laser sensors, etc, etc. They have a lot of options to test parts without destroying them. For the infamous bolts, I have been wondering whether things like the grain structure could have been inspected with a nano- CT scan without destroying the bolt.

For strut testing, prior to installing in the fuel tank, I would think a "tone" test could work.  Essentially, make a few copies of the struts, as close to perfect as possible, send an audio signal pulse through the struts from one end to teh other and set up a baseline average of what signal SHOULD be found.  When testing, the struts that fail the tone test, are rejected for further testing.

     Mind you, this is based upon the concept that the same materials, forged in slightly different fashions, will pass sound through them at different speeds and will change the frequency of the sound being passed through them.  Microfractures would likewise affect changes in both speed and frequency of sound passing through the strut.

      While this will not give as detailed an assesment of the struts, it should be a fairly quick and accurate initial pass/fail test for any struts that are brought in from outside manufacturers.  And it has the advantage of being nondestructive.

[inonepiece]

There are micro/nano- CT- scanners that can scan pretty big parts, these days. Then there is Xrays, Ultrasound tactile and laser sensors, etc, etc. They have a lot of options to test parts without destroying them. For the infamous bolts, I have been wondering whether things like the grain structure could have been inspected with a nano- CT scan without destroying the bolt.

I don't think exposing steel to soft X-rays will damage the grain boundary structure (therefore won't damage structural properties).  I used to do calculations on this sort of thing (X-ray extinction, not damage) -- that was a long time ago, but I think at least relatively high energy soft X-rays (about 10keV) have a extinction depth in steel long enough to exceed typical grain size for good quality steel, hence have a chance to "see" a lot of potential problems.

[Elmar Moelzer]

I don't think exposing steel to soft X-rays will damage the grain boundary structure (therefore won't damage structural properties).  I used to do calculations on this sort of thing (X-ray extinction, not damage) -- that was a long time ago, but I think at least relatively high energy soft X-rays (about 10keV) have a extinction depth in steel long enough to exceed typical grain size for good quality steel, hence have a chance to "see" a lot of potential problems.

Sorry, I think you misunderstood my post. I am well aware that a CT- Scan of a bolt won't damage it. I was simply comparing it to other, destructive methods of testing that finally revealed bolts with a "broken" grain structure. I agree on the second sentence.

[Robotbeat]

I wonder if they cannot get the needed info from nondestructive means. Ultrasound or magnetic tests before and after. How about repeating the structural tests they did for acceptance in the test stand at McGregor? Maybe destructive tests on a very few locations simulations show to experience the biggest stress.

There are a lot of ways to do non destructive testing and I have been wondering which ones they are using at SpaceX.
There are micro/nano- CT- scanners that can scan pretty big parts, these days. Then there is Xrays, Ultrasound tactile and laser sensors, etc, etc. They have a lot of options to test parts without destroying them. For the infamous bolts, I have been wondering whether things like the grain structure could have been inspected with a nano- CT scan without destroying the bolt.

For strut testing, prior to installing in the fuel tank, I would think a "tone" test could work.  Essentially, make a few copies of the struts, as close to perfect as possible, send an audio signal pulse through the struts from one end to teh other and set up a baseline average of what signal SHOULD be found.  When testing, the struts that fail the tone test, are rejected for further testing.

     Mind you, this is based upon the concept that the same materials, forged in slightly different fashions, will pass sound through them at different speeds and will change the frequency of the sound being passed through them.  Microfractures would likewise affect changes in both speed and frequency of sound passing through the strut.

      While this will not give as detailed an assesment of the struts, it should be a fairly quick and accurate initial pass/fail test for any struts that are brought in from outside manufacturers.  And it has the advantage of being nondestructive.

Or ultrasound. Stick it in an ultrasound scanner tank.

[deruch]

Instead of somewhat uselessly going back and forth over whether and how SpaceX might capture and reuse the He, ect. from the returned stages, why don't we look at what they've said they are going to do. 

SpaceX laid out their plans for operating the landing pad as part of the Environmental Assessment done for building the pad at LC-13:

Operations of LC-13 would support preparations for, and the landing of the Falcon stage.  It would also support the post-flight  landing  and  safing.    Safing  activities would  begin  upon  completion  of  all  landing  activities  and  engine shutdown.  The LOX oxidizer system would be purged, and any excess fuel would be drained into a suitable truck mounted container or tanker.  Any remaining pressurants (i.e., helium or nitrogen) would be vented, and any FTS explosives would also be rendered “inert” prior to declaring the vehicle safe. (emphasis added)

Sorry I can't find a link to the final, but the above is from their draft found at [.PDF]: http://www.patrick.af.mil/shared/media/document/AFD-141107-004.pdf

It's towards the end of section 2.2

[dorkmo]

would they need to have the crane lifting from the top before they depressurize?

[rower2000]

Perhaps this isn't apples to apples, but as a mixed gas technical diver, one reason people are moving to rebreathers is because of helium cost. For a single moderately deep open circuit dive, say 2 x 100 cu ft at 50% He, and a 40 cu ft at 30% He - the rest of the tanks containing O2 / N2 only blends - so 112 cu ft of He, costs in the range of $300, which leads to $2.68 / cu ft. So extending this out to the the above, you get 18,500 x 4 x 2.68 = $198,214. Point being, I don't think your numbers are in the right ballpark.

It's all about quantity. For commerical bulk users, the average crude He price in 2014 was 3.43 USD/m^3 or 95 USD/1000 cu ft. A diving store, just quickly googled, charges much more, 0.03 GBP/litre for helium which corresponds to 1230 USD/1000 cu ft. The industrial bulk market has different laws than the private user hobby market .

[Robotbeat]

Instead of somewhat uselessly going back and forth over whether and how SpaceX might capture and reuse the He, ect. from the returned stages, why don't we look at what they've said they are going to do. 

SpaceX laid out their plans for operating the landing pad as part of the Environmental Assessment done for building the pad at LC-13:

Operations of LC-13 would support preparations for, and the landing of the Falcon stage.  It would also support the post-flight  landing  and  safing.    Safing  activities would  begin  upon  completion  of  all  landing  activities  and  engine shutdown.  The LOX oxidizer system would be purged, and any excess fuel would be drained into a suitable truck mounted container or tanker.  Any remaining pressurants (i.e., helium or nitrogen) would be vented, and any FTS explosives would also be rendered “inert” prior to declaring the vehicle safe. (emphasis added)

Sorry I can't find a link to the final, but the above is from their draft found at [.PDF]: http://www.patrick.af.mil/shared/media/document/AFD-141107-004.pdf

It's towards the end of section 2.2

Pretty much goes without saying that SpaceX isn't /currently/ recovering helium, nor are they likely to do so any time soon. So this is irrelevant.

[Digitalchromakey]

Perhaps this isn't apples to apples, but as a mixed gas technical diver, one reason people are moving to rebreathers is because of helium cost. For a single moderately deep open circuit dive, say 2 x 100 cu ft at 50% He, and a 40 cu ft at 30% He - the rest of the tanks containing O2 / N2 only blends - so 112 cu ft of He, costs in the range of $300, which leads to $2.68 / cu ft. So extending this out to the the above, you get 18,500 x 4 x 2.68 = $198,214. Point being, I don't think your numbers are in the right ballpark.

It's all about quantity. For commerical bulk users, the average crude He price in 2014 was 3.43 USD/m^3 or 95 USD/1000 cu ft. A diving store, just quickly googled, charges much more, 0.03 GBP/litre for helium which corresponds to 1230 USD/1000 cu ft. The industrial bulk market has different laws than the private user hobby market .

He for diving requires a much higher grade of purity than is the case for 'industrial' He which has a significantly lower % purity (and thus cost).

[Comga]

Here's an enhanced image of the engine closeup. (snip) 
The white stuff in the nozzles appears to be some liquid that has dried, perhaps some leaking TEA-TEB fluid, or an additive in the RP-1. (snip)

Wow
Those patterns inside the two outer (what I assume are the numbers 1 and 5) engines are nearly mirror images of each other.
That can't be just a coincidence.
It must be deterministic.

Would there be a reason to inject the TEA-TEB into all three engines on all three restarts, even though the valves are only open for the center (#9?) engine for the entry and landing burns? 
Would that make the system simpler or more complex?

I agree the whiter areas inside two engines opposite one another hint that those engines were the ones restarted. I assume the center engine has similar white area, but the "clocking" makes it less visible.

Something like ignition fluid sure makes sense, but I'm not sure: is there a way we can prune from the causal tree the possibility that these white areas are associated with second/third/fourth engine shutdown transients?

I don't think that the center engine has the same white areas.
My guess is that the landing burn burnt them off.
I don't think it's "leaking" TEA/TEB as Steven suggested.  I think it is full flow for engines that were not restarted for the last burn or last two burns.
Asked above: Would it make the system simpler or more complex to inject TEA/TEB into all three each time or to inject them separately only when needed?

[cscott]

One central TEA/TEB tank, one actuator, three tubes from there to engines?

I think we got the hint that all TEA/TEB is supplied from the ground for launch, so having a single manifold might follow?

Taking the other side, if there is an valve actuator on each engine, then the only difference is really trivial amounts of wiring and electronics.  And there's a big disadvantage, I'd think, in not being able to test the engines separately.

[Kabloona]

Asked above: Would it make the system simpler or more complex to inject TEA/TEB into all three each time or to inject them separately only when needed?

Simpler to inject into all three engines at every restart (inculding landing burn) because then you need only two valves (one each for the TEA and TEB) with 3-way manifolds downstream, instead of six valves (two per engine).

In that scheme, TEA/TEB would also be injected into the two non-starting engines during start of landing burn (believed to be #1 and #5, the ones with the white residue) and would explain the presence of that white residue, since it doesn't get burned off the nozzle subsequently by kerolox as happens in the center engine.

I don't have experience with TEA/TEB, but the combustion byproducts aluminum oxide and boron oxide are both white, so it seems a good hypothesis that the white residue is from TEA/TEB injected during landing burn restart.

[MechE31]

Asked above: Would it make the system simpler or more complex to inject TEA/TEB into all three each time or to inject them separately only when needed?

Simpler to inject into all three engines at every restart (inculding landing burn) because then you need only two valves (one each for the TEA and TEB) with 3-way manifolds downstream, instead of six valves (two per engine).

In that scheme, TEA/TEB would also be injected into the two non-starting engines during start of landing burn (believed to be #1 and #5, the ones with the white residue) and would explain the presence of that white residue, since it doesn't get burned off the nozzle subsequently by kerolox as happens in the center engine.

I don't have experience with TEA/TEB, but the combustion byproducts aluminum oxide and boron oxide are both white, so it seems a good hypothesis that the white residue is from TEA/TEB injected during landing burn restart.

I can confirm that TEA/TEB residue is indeed white and that is most probably what is seen here.

[CuddlyRocket]

For strut testing, prior to installing in the fuel tank, I would think a "tone" test could work.  Essentially, make a few copies of the struts, as close to perfect as possible, send an audio signal pulse through the struts from one end to teh other and set up a baseline average of what signal SHOULD be found.  When testing, the struts that fail the tone test, are rejected for further testing.

     Mind you, this is based upon the concept that the same materials, forged in slightly different fashions, will pass sound through them at different speeds and will change the frequency of the sound being passed through them.  Microfractures would likewise affect changes in both speed and frequency of sound passing through the strut.

      While this will not give as detailed an assesment of the struts, it should be a fairly quick and accurate initial pass/fail test for any struts that are brought in from outside manufacturers.  And it has the advantage of being nondestructive.

Or ultrasound. Stick it in an ultrasound scanner tank.

Given the struts were (supposedly!) engineered to withstand five times the maximum load they are expected to experience in flight, it would seem the obvious test would be to load them to three or four times that load for three to four times flight time. Simple and direct.

[rower2000]

Perhaps this isn't apples to apples, but as a mixed gas technical diver, one reason people are moving to rebreathers is because of helium cost. For a single moderately deep open circuit dive, say 2 x 100 cu ft at 50% He, and a 40 cu ft at 30% He - the rest of the tanks containing O2 / N2 only blends - so 112 cu ft of He, costs in the range of $300, which leads to $2.68 / cu ft. So extending this out to the the above, you get 18,500 x 4 x 2.68 = $198,214. Point being, I don't think your numbers are in the right ballpark.

It's all about quantity. For commerical bulk users, the average crude He price in 2014 was 3.43 USD/m^3 or 95 USD/1000 cu ft. A diving store, just quickly googled, charges much more, 0.03 GBP/litre for helium which corresponds to 1230 USD/1000 cu ft. The industrial bulk market has different laws than the private user hobby market .

He for diving requires a much higher grade of purity than is the case for 'industrial' He which has a significantly lower % purity (and thus cost).

Well, grade A He with N4.6 is not too much more expensive, about a factor of two, according to the USGS source. Grade A is sufficient for diving applications. But I fear we are getting off topic here.