Helium leaks(?) on Boeing Starliner

[Keith Lofstrom]

Not directly related to "power satellite economics", though
whatever vehicle launches systems for power satellites will
use helium for some subsystems, and may be plagued by
expensive-to-solve problems similar to Starliner.

Most of our SBSP cost estimates assume flaw-free production
and perfect performance. This faulty reasoning causes most
system failures, delays, and cost overruns. Hubris kills.

----

Helium is 5 ppm in the atmosphere; if you measure larger
concentrations than that, something man-made nearby is
probably leaking it. With the proper not-too-expensive
equipment, those excess concentrations are easy to measure.


In college, I used an obsolete second-hand vacuum system
to make superconducting electronic chips; hot filament
vapor deposition of lead. Lead is a nasty contaminant,
and wasn't allowed in the newer cleaner vacuum systems.

The seals were old and leaked frequently. We found leaky
seals by placing a helium detector inside the bell jar,
and a hose dispensing a small stream of helium outside.
We moved the hose around the gaskets and seals and weld
joints. A tiny leak was enough to set off the detector.
We (slowly, many hours) learned which seal to replace,
on a system smaller than your kitchen refrigerator.

We did not have the budget to frequently replace all the
seals, and some required custom machining, because that
ancient vacuum system was no longer produced, and spares
were no longer available.

The same one-of-a-kind ad-hoc testing will be true for a
launch system, because those will not be mass produced
continuously in large quantities.

Even commercial airliners are not mass produced like other
products - for instance, we don't use 0.1% of the finished
aircraft for destructive testing, as we do for other high
reliability but lower-cost systems.

----

Boeing is doing ad-hoc testing of the crafted custom-made
Starliner on the pad at KSC, and measuring anomalous levels
of helium coming from "somewhere". I hope they succeed,
and find an easy and reliable fix ... but if you helium-test
in an ad-hoc manner, you will get unreliable indications
rather than conclusive, irrefutable results.

Worse ... your process will be vulnerable to sabotage.
I don't know if we can rule out subtle sabotage.

I can imagine a hostile adversary placing a "helium leaker"
somewhere on the Starliner in an impossible-to-access
location, emitting small burps of helium at random
intervals, practically impossible to track down without an
enormous 24x7xWeeks whole-vehicle search. Boeing, NASA,
and the United States have too many hostile adversaries.

I hope I am not giving a hostile adversary another idea.

Worse, if the Starliner diagnosis IS proved to be "helium
sabotage", I can imagine draconian regulation of the US
helium supply chain, and hundreds of small companies and
universities no longer able to test their vacuum systems.
Saboteurs will evade those regulations by hiring party
clowns and using their helium balloons instead :-)

Another brick in the wall.

Keith L.

--
Keith Lofstrom kei...@keithl.com

[Keith Henson]

"The same one-of-a-kind ad-hoc testing will be true for a
launch system, because those will not be mass produced
continuously in large quantities."

I don't know if even one power satellite will ever be produced, but if
we build 1000 power satellites over 20 years (1/3rd of current power)
that's 50 a year. Each one takes about 500 flights to haul up the
parts and reaction mass needed to get to GEO. That's 25,000 flights
per year. If the rockets last 100 flights, you need a construction
rate of 250 a year, which is getting close to mass production.

Not sure about this, but I don't think the current SpaceX rockets use
helium. Anyone know?

I worry more about ozone damage from the high rocket traffic. I
talked NOAA into looking at this question with the hydrogen burning
Skylon, but it has not been done as far as I know with methane
burners.

KeithH


Best wishes,

Keith

> --
> You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/20240529233723.GA29663%40gate.kl-ic.com.

[k.a.carroll]

Falcon 9 uses helium (stored as a cryogenic fluid) to pressurize both propellant tanks. (They suffered an explosion when one of those He tanks came loose, over-pressurizing the propellant tank very rapidly). Dunno about Starship and its booster.

To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB6p_24Pjp6TNHRCcjbq_y5Lc%2Bxk3DYAZsdLgBEHSXeWoA%40mail.gmail.com.

[John K. Strickland, Jr.]

Liquid Helium needs to be so cold that the cryogenic propellants would not liquefy it.

Nitrogen may be used for the cold gas thrusters but not to pressurize cryogenic propellants.

 

SPS systems could be built very rapidly if someone decided to actually go ahead and build them.

This is very similar to the problem of reusable rockets.  The switch is either ON or OFF.

 

“Billions of dollars to endlessly study global warming but not a penny to solve the problem in a practical way.”

 

John

To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/6657d012.170a0220.cdc94.b75bSMTPIN_ADDED_BROKEN%40gmr-mx.google.com.

[Roger Arnold]

Helium tanks have been used to pressurize fuel tanks on a lot of rockets. I suppose that makes sense if the rocket is carrying supercooled LOX and LH2 or LME; you wouldn't want the pressurizing gas to be condensing in the tank. But I've never understood why it would be used for a LOX - kerosene / RP6 system. Nitrogen should be fine for that. Maybe it's just that, compared to the cost of everything else, the cost of pressurized helium over nitrogen is so far down in the noise that it's not worth departing from a standard and well-characterized system. 

In any case, I think the issue is moot for SpaceX rockets. I believe they use a bleed of combustion gas to pressurize the tanks. I could be wrong about that however. I haven't paid that much attention.

I was surprised at Keith's L's statement that the atmospheric concentration of He was about 5 ppmv. I checked, and he's right. 5.2 ppmv is the "official" figure. I thought it was quite a bit lower -- down in the parts per billion range. CO2 concentration is only 420 ppmv (up from a pre-industrial average of around 280). Xenon is 18 ppmv, and that's enough to produce it as a side product from large cryogenic air separation units. Helium is only about 3x less, so it seems that it, too, could be produced as a side product from ASUs. The price would be quite high, but if we really needed it, that would make it a renewable resource.

If Starliner is only using He for tank pressurization, I'm a little surprised that anyone would be worrying about small leaks. Helium detectors of the sort that Keith talks about for leak detection in vacuum systems are extremely sensitive. A helium pressure tank could be leaking enough to trigger a detector in the region around the tank, but go on leaking at that rate for a year without losing significant pressure in the tank. A case of "an abundance of caution"?

To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/001a01dab233%2453adb460%24fb091d20%24%40sbcglobal.net.

[Keith Lofstrom]

On Wed, May 29, 2024 at 10:25:38PM -0700, Roger Arnold wrote:
> If Starliner is only using He for tank pressurization, I'm a little

> surprised that anyone would be worrying about *small* leaks. Helium

> detectors of the sort that Keith talks about for leak detection in vacuum
> systems are extremely sensitive. A helium pressure tank could be leaking
> enough to trigger a detector in the region around the tank, but go on
> leaking at that rate for a year without losing significant pressure in the
> tank. A case of "an abundance of caution"?

Too many oughta works is a NOT oughta work - and this is
only the second Starliner service module to reach the pad.
They aren't detecting leaks in vacuum, but in sea level
atmosphere on top of the Atlas V. Last I heard, the leak
was traced to one of 28 control thrusters.

(Why does a control thruster use helium? Why didn't
they find the fault sooner? Is this a cruelty-free
spacecraft ... no testing?)

If the system leaks more helium than the prior unmanned
launch, that demands investigation. "Cover your ass" extra
inspection is better than 2 dead astronauts (or worst case
4 extra dead on a pranged ISS) and Rogers Commission II,
the Sequel.

A late launch is better than "The Late Boeing 1917-2024 RIP"

Boeing recently "helped" Alaska Airlines drop a 737 door
plug in an acquaintance's back yard, and a tray table in
a friend's mom's back yard, both locations two miles from
me, less than 200 meters from where I helped search the
woods between neighborhoods with an amateur posse and a
couple of NTSB investigators.

If Boeing drops a Starliner, battered Boeing will be eaten
by greedy lawyers, Boeing friends will be unemployed, and
top bosses will have new jobs making license plates at
FCI Sheridan. :-(

Keith L.

P.S. Licence plates: old joke. Here's what they make in
federal prisons: https://www.unicor.gov/BusinessCatalogs.aspx
Perhaps wiring harnesses for Starliner:
https://www.unicor.gov/publications/electronics/CableAndWireHarnessCatalog.pdf

--
Keith Lofstrom kei...@keithl.com

[k.a.c...@sympatico.ca]

KeithL wrote:

> (Why does a control thruster use helium? ...

The valve in a thruster needs actuating open and closed. That involves applying a force. One way to produce that force is to let a pressurized gas (e.g., helium) push against a piston. Some thruster valves use a small electrically-operated (via a solenoid) valve to open and close access to a pressurized helium source, with the helium then producing a force on a much more powerful main thruster valve to open it (and possibly close it --- alternately you could use a strong spring to close it, once the helium supply has been removed by closing the first valve and venting the helium that actuated it, via another small solenoid-powered valve).

Here's a bit of satellite failure trivia, from a friend who was AMSAT North America's (AMSAT/NA) main systems engineer for many years. The AMSAT Phase 3D satellite exploded in space due to just such a thruster. Phase 3D (which flew under the name AO-40 https://en.wikipedia.org/wiki/OSCAR_40) was built as a collaboration between AMSAT/NA and AMSAT/DL (AMSAT Germany). AMSAT/DL provided a powerful hydrazine-powered thruster for main orbit corrections (among many other AMSAT/DL parts of the satellite), which they got as a donation from its German manufacturer; it was left over from another program (and had been sitting in storage for some years).

As described above, such thrusters need a helium vent port. On this one, the vent port was covered by a plastic cap, a remove-before-flight item, to keep its interior clean during assembly, integration and test. That cap was inadvertently *not* removed before flight. So in-flight, when they first actuated that thruster, the pressurized helium properly forced the thruster's main valve open. But when it was commanded closed at the end of the planned orbit-change manoeuvre, by commanding closed the main helium supply valve and commanding open the helium vent port, the helium couldn't vent properly from that port. As best as they could reconstruct events, the helium did finally escape from the helium-space in the thruster, but slowly --- apparently the vent-port cap remained in place, but the helium slowly leaked past it. So the main thruster valve didn't close rapidly, it instead closed slowly. Slowly enough that some unburned hydrazine ended up leaking out the thruster's nozzle. It's thought this hydrazine stayed in the vicinity of the satellite --- particularly in it interior --- and that a little time later, when a transmitter was turned on, that ignited it, causing a small explosion inside the satellite. Which blew off one of the satellite's panels, taking some electronic boxes with it --- sort of an Apollo 13 situation. As it happens, the satellite was a behemoth (bay far AMSAT's largest and most expensive satellite), with a huge number of different radios, plus another (arcjet) thruster, and what equipment was left was enough to let the satellite carry out a fair fraction of its planned mission (although in the wrong orbit).

Why was the RBF cap not removed? A conjunction of several reasons. First, it was a plastic cap, which was originally coloured red (like all RBF items are supposed to be). But the engine was old, and the red pigment in the plastic had faded to pale pink, making it far less obviously a RBF item.

This wouldn't have mattered, if the AMSAT engineer in charge of that engine had been present for AI&T. However, he suffered a heart attack shortly before the AI&T campaign, and had to undergo a triple-bypass operation, leaving him unable to travel from Germany to America to participate in the AI&T. As with all AMSAT satellites, the documentation was sparse; there was not detailed written work-instructions document for the thruster installation and test procedure, instead that was mainly in that German engineer's head. Whoever took over from him during satellite AI&T missed getting the instruction to remove the pale-pink helium vent cap.

The satellite was assembled in a bonded warehouse at Orlando Airport, to simplify the importation of satellite/rocket type equipment from Germany. As it happens, I had a family trip to Florida while it was undergoing assembly, and was friends with the AMSATters putting it together (they were partnered with us at Dynacon, on the development of our MOST microsat), so I popped in to see the facility. Somewhere or other I have a video I took there, including of the entire satellite --- inside and out! The satellite was large, IIRC octagonal in planform about 6' across-flats, and about 4' tall, with a ~ 3'-wide hollow space in the centre (to get at interior-wall components, and where the thruster were mounted). I was allowed to crawl under the satellite, and pop up into that hollow space to see its interior. I wonder if my video captured that thruster and its RBF cap; I'll have to dig it up and check (if I can find a VCR that still works). Probably not though --- I suspect the thrusters were installed after my visit.

- Kieran

[k.a.c...@sympatico.ca]

John wrote:

 

> Liquid Helium needs to be so cold that the cryogenic propellants would not liquefy it.

 

Indeed. In the Falcon 9, the liquid helium used to pressurize the tanks is stored in a  dewar, that is actually inside of the tank (surrounded by the propellant in the tank). They must use a heater inside the dewar, to heat up the helium in order to drive out of the dewar and into the propellant tank, as and when needed. (I.e., turn the heater on to full power at engine ignition, because they’ll want to empty that dewar within the net couple of minutes.)

 

- Kieran

[james...@aol.com]

I am not responding directly to Keith's comments.

The primary issue with the Boeing Starliner is the lack of adequate technical and programmatic oversight by NASA.

A secondary issue is trying to cram too much operational capability into the payload capability of the launch system. Again, this is an issue with NASA.

I am not arguing that Boeing is fault free. There are many endemic issues at Boeing that have had a cumulative effort on Boeing's technical performance in several areas. In a way, this is a consequence of lack of true competition across the last four decades—a consequence of failed national space policies.

The decline in NASA's capabilities began in the early 1980s following the success of the Apollo program and Skylab. I believe NASA did a good job of bringing the Space Shuttle—in a configuration that was hazardous from the beginning—into operation in the 1980s. Remember, the Space Shuttle was designed when pocket calculators were just becoming available. However, by the 25th flight, NASA's technical and programmatic competence was failing. After the loss of Challenger, it has forever been in a political CYA mode protected by a Congress focused primarily on the jobs in NASA and its contractors.

Mike Snead, PE

-----Original Message-----
From: power-satell...@googlegroups.com <power-satell...@googlegroups.com> On Behalf Of Keith Lofstrom
Sent: Thursday, May 30, 2024 3:05 AM
To: Roger Arnold <silver...@gmail.com>
Cc: Power Satellite Economics <power-satell...@googlegroups.com>
Subject: Re: Helium leaks(?) on Boeing Starliner

--
You received this message because you are subscribed to the Google Groups "Power Satellite Economics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to power-satellite-ec...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/20240530070438.GB29663%40gate.kl-ic.com.

[Keith Henson]

As I recall (and may remember wrong) the helium in an F9 rocket is
just pressurized inside a spherical carbon fiber pressure vessel. The
sphere is inside the LOX tank and as I remember, one of them in the
early days lit on fire due to friction between layers as the tank was
pressurized.

KeithH

PS, asked an AI. "An early Falcon 9 rocket blew up on the pad as it
was being fueled. I think the reason was traced to a helium tank
inside the LOX tank. Can you point me to the details?

Copilot, 10:04 AM
Certainly! The incident you're referring to involved a Falcon 9 rocket
explosion that occurred on September 1, 2016, at Cape Canaveral. The
preliminary findings pointed to an issue with the cryogenic helium
system within the second stage's liquid oxygen (LOX)
tankhttps://newatlas.com/spacex-falcon-9-explosion-helium/45594/.

The anomaly happened during a routine pre-launch static fire test,
where the rocket's engines are fired while it's held down. The
explosion took place approximately eight minutes before the engines
were scheduled to ignite, during the fueling process. No one was
injured, but the rocket and its payload were destroyed, and the launch
complex was significantly
damagedhttps://newatlas.com/spacex-falcon-9-explosion-helium/45594/.

The investigation suggested that there was a large breach in the
cryogenic helium system of the second stage LOX tank. The helium is
used to pressurize the propellant tanks to maintain structural
integrity as the fuel is
consumedhttps://space.stackexchange.com/questions/18329/problems-compressing-the-helium-tanks-inside-the-f9-2nd-stage-lox-tank-at-the-la.
The helium tanks are pressurized to about 5,500 pounds per square inch
and are critical for the proper functioning of the rocket's Merlin
enginehttps://space.stackexchange.com/questions/18329/problems-compressing-the-helium-tanks-inside-the-f9-2nd-stage-lox-tank-at-the-la.

SpaceX conducted a thorough investigation in collaboration with
several agencies, including the FAA, NASA, and the US Air Force, to
determine the exact cause of the
failurehttps://www.space.com/34187-spacex-falcon-rocket-accident-investigation-helium-system.html.
While the specific details of the helium system failure were not
disclosed, the nature of the explosion indicated a catastrophic
scenario likely related to this
systemhttps://newatlas.com/spacex-falcon-9-explosion-helium/45594/.

For more detailed information, you can refer to the discussions and
analyses on space exploration forums and the official SpaceX updates
released after the
incidenthttps://space.stackexchange.com/questions/18329/problems-compressing-the-helium-tanks-inside-the-f9-2nd-stage-lox-tank-at-the-lahttps://newatlas.com/spacex-falcon-9-explosion-helium/45594/https://www.space.com/34187-spacex-falcon-rocket-accident-investigation-helium-system.htmlhttps://spacenews.com/falcon-accident-investigation-points-to-breach-in-rockets-helium-system/.

> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/1b3101dab2a0%24f5d8f980%24e18aec80%24%40sympatico.ca.

[John K. Strickland, Jr.]

The LOX right next to the liquid helium started to freeze and the crystals created friction of some kind that set off the fire.
Since they probably had never had LOX right against liquid helium before, this was a new type of accident creation.


John

To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/CAPiwVB4uN2KwZ2Z2PbYZtVcqhFJagqUfXvi8K5DQDE50Mv%2B2%3DQ%40mail.gmail.com.

[Keith Henson]

On Thu, May 30, 2024 at 1:04 PM John K. Strickland, Jr.
<jkst...@sbcglobal.net> wrote:
>
> The LOX right next to the liquid helium started to freeze and the crystals created friction of some kind that set off the fire.
> Since they probably had never had LOX right against liquid helium before, this was a new type of accident creation.

It was not liquid He, high pressure, 5500 psi.

Keith

> To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/011001dab2cc%24a0dcfa40%24e296eec0%24%40sbcglobal.net.

[k.a.c...@sympatico.ca]

Yeah, I was mistaken about the liquid helium. (A news report around then claimed liquid helium, my friends at the top of SpaceX got back to me and told me that news reporter was making stuff up, it was actually high-pressure gaseous helium. I had forgotten that in the intervening years...)

> economics/CAPiwVB6p_24Pjp6TNHRCcjbq_y5Lc%2Bxk3DYAZsdLgBEHSXeWoA%
> 40mail.gmail.com.

> > >
> > > --
> > > You received this message because you are subscribed to the Google Groups
> "Power Satellite Economics" group.
> > > To unsubscribe from this group and stop receiving emails from it, send an
> email to power-satellite-ec...@googlegroups.com.
> > > To view this discussion on the web visit
> https://groups.google.com/d/msgid/power-satellite-

> economics/6657d012.170a0220.cdc94.b75bSMTPIN_ADDED_BROKEN%40gmr-
> mx.google.com.

> > >
> > > --
> > > You received this message because you are subscribed to the Google Groups
> "Power Satellite Economics" group.
> > > To unsubscribe from this group and stop receiving emails from it, send an
> email to power-satellite-ec...@googlegroups.com.
> > > To view this discussion on the web visit
> https://groups.google.com/d/msgid/power-satellite-

> economics/1b3101dab2a0%24f5d8f980%24e18aec80%24%40sympatico.ca.

> >
> > --
> > You received this message because you are subscribed to the Google Groups
> "Power Satellite Economics" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> email to power-satellite-ec...@googlegroups.com.
> > To view this discussion on the web visit
> https://groups.google.com/d/msgid/power-satellite-

> economics/CAPiwVB4uN2KwZ2Z2PbYZtVcqhFJagqUfXvi8K5DQDE50Mv%2B2%3
> DQ%40mail.gmail.com.

> >
> > --
> > You received this message because you are subscribed to the Google Groups
> "Power Satellite Economics" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> email to power-satellite-ec...@googlegroups.com.
> > To view this discussion on the web visit
> https://groups.google.com/d/msgid/power-satellite-

> economics/011001dab2cc%24a0dcfa40%24e296eec0%24%40sbcglobal.net.

[John K. Strickland, Jr.]

Well the higher pressure could certainly have contributed to the willingness of the LOX to start freezing if transmitted through the !he metal ? lox tamk.
I heard also that it was liquid helium.

To view this discussion on the web visit https://groups.google.com/d/msgid/power-satellite-economics/1ccb01dab2ea%24542170a0%24fc6451e0%24%40sympatico.ca.

[Keith Lofstrom]

On Thu, May 30, 2024 at 09:32:14AM -0400, k.a.c...@sympatico.ca wrote:
> KeithL wrote:
>
> > (Why does a control thruster use helium? ...
>
> The valve in a thruster needs actuating open and closed. That involves applying a force. One way to produce that force is to let a pressurized gas (e.g., helium) push against a piston. Some thruster valves use a small electrically-operated (via a solenoid) valve to open and close access to a pressurized helium source, with the helium then producing a force on a much more powerful main thruster valve to open it (and possibly close it --- alternately you could use a strong spring to close it, once the helium supply has been removed by closing the first valve and venting the helium that actuated it, via another small solenoid-powered valve).

I'll re-state my question - WHY NOT provide valve force
with (more) electricity, rather than a two stage process
that involves helium gas lines and tanks, and TWO valves
per thruster? That's a lot of heavy and leaky plumbing.

My guess is that thruster designers think gases, and not
electrons, and may be unaware of the spectacular improvements
in power FETS and OPTICAL triggering - a power FET could be
switched on and off with a 9-micrometer-core graded-index
optical fiber (plus cladding, shared with many fibers).

The FET voltage drop in the "on" state can be tiny, with a
centimeter square, 100 micrometer thick, 30 milligram
silicon carbide chip.

I would use 4 FET chips in series-parallel with four control
fibers, guarding against rare opens and shorts. Perhaps also
100 grams of super-capacitor and inductor to provide an extra
electrical kick to the no-doubt-high-inertia valve.

Maneuvering thrusters are probably used in tandem, but not
every thruster simultaneously, nor for long periods of time,
so their total energy requirements won't be huge.

The same (or different) optical fibers could be used to
return status information to the controller.

-----

BTW, modern optical fibers for communication are multiplexed
and demultiplexed onto one master fiber with a wavelength-
division etalon filter, precisely etched patterns behaving
like optical bandpass filters.

My home internet connection fiber (along with dozens of
other fibers connected to neighbors) merge at an etalon
at the end of the street, connecting on one fiber, bundled
with fibers to other streets, running in an armored cable
to the provider's switching center 10 kilometers away.

Note that the next generation of integrated circuits will
probably also use optical interconnect through an add-on
multi-laser chip and etalon, connecting to other circuits
and chips on an "optical fiber" motherboard.

The chips themselves can be EXTREMELY radiation-resistant;
bleeding edge new processes are patterned with 3KeV M-band
X-rays. A trillion transistors on a chip must ALL survive
this brutal manufacturing treatment, in an imaging "stepper"
the size of a single family home. The first chip patterning
tool I used in 1973 was about half a cubic meter.

And for the long term - a "primarily electronic" actuator
lends itself to invention and upgrade, and the spinoffs
that NASA parades to justify their budget. I can imagine
(not in actionable detail) a piezoelectric valve that
expands and contracts some shaped material that replaces
a mechanical pivot, or even a material that radically
changes its gas/liquid permeability when subjected to a
voltage. Imagined vaguely, not specified in detail, but
worthy of an R.F.Q. to dozens of research organizations.

I write about all of this HERE because it is relevant to
future solar power satellite designs. Cheap fast good,
PICK ALL THREE. Perhaps that will inspire clever new
SSPS designs, where "clever" means "cheap, easy, and soon."

-----

That said, there will only be zero, one, or a very few ways
to do this right, and a crapton of ways to do this wrong.
Test it first on different unmanned experimental satellites,
then deploy it on crewed systems a decade later. Such wild
experiments was one reason NASA was born, decades ago.

The lack of those experiments in recent decades is why we
don't have such systems (and their industrial and consumer
spinoffs) today. NASA should explore edgy new technologies,
not just perpetuate ancient self-licking ice cream cones.