We need a better system for quick disconnects for hydrogen fueling.

[Exoscientist]

 I was quite startled to read from space reporter Eric Berger that the SLS rocket can only do 2 more rollbacks before it is decertified(!) Huh? A $2 billion spacecraft just gets thrown away, unused!?!

Eric Berger  
@SciGuySpace  
Also, per a source, NASA has certified the SLS rocket for just two more rollouts from the VAB. So if they were to roll back to VAB this month and then back to the pad, they would have just one roundtrip left. So that's kind of a serious constraint on this hardware.  
5:14 PM · Sep 7, 2022  
120 Retweets 28 Quote Tweets 1,526 Likes  
https://twitter.com/sciguyspace/status/1567622377185415173?s=21&t=5LtvwPXyKM1uFyyqGcs9WA

As this article shows hydrogen leaks during fueling have been a recurrent problem going back 40 years with the Shuttle:  

Years after shuttle, NASA rediscovers the perils of liquid hydrogen  
"Every time we saw a leak, it pretty quickly exceeded our flammability limits."  
ERIC BERGER - 9/3/2022, 6:38 PM  
https://arstechnica.com/science/2022/09/years-after-shuttle-nasa-rediscovers-the-perils-of-liquid-hydrogen/  

  These quick disconnect fittings that are the source of the problem are quite low tech:  

Valve Disconnect  
A closeup of the 7-inch quick disconnect that will be replaced on the hydrogen vent line to the Ground Umbilical Carrier Plate of space shuttle Discovery's external fuel tank. The replacement will be made on Launch Pad 39A at NASA's Kennedy Space Center in Florida. A leak of hydrogen at the site during fueling caused the STS-119 mission to be scrubbed at 2:36 p.m. March 11. 
Photo Credit: United Space Alliance
March 11, 2009

 Surely someone can up with a more advanced tech solution that will stop the hydrogen  leaks in a quick disconnect system???

  Robert Clark

Edited September 8, 2022 by Exoscientist

[JoeSchmuckatelli]

32 minutes ago, Exoscientist said:

Surely someone can up with a more advanced tech solution that will stop the hydrogen  leaks in a quick disconnect system???

Surely.

I mean, it's not like hydrogen is difficult to work with, or anything; you'd think professional rocket scientists, chemists and engineers would have solved this years ago.

rite?

[Terwin]

1 hour ago, Exoscientist said:

 Surely someone can up with a more advanced tech solution that will stop the hydrogen  leaks in a quick disconnect system???

Complexity is the enemy of reliability.

Also, we are dealing with liquid hydrogen.  I expect lots of potential moving parts start having problems when they need to regularly cycle between Florida Summers(~100f/~310K) and liquid hydrogen(~33 Kelvin).  I also expect that icing may be a problem when you are operating at ~240 degrees below freezing in a swamp and within sight of the ocean.

I find it pretty impressive that they can even get the accumulated ice to let go after pumping several hundred thousand gallons of liquid hydrogen into the rocket with humidity that is probably not often below 90%.

Edited September 8, 2022 by Terwin

[Gargamel]

As somebody who has personally machined high tolerance quick disconnects for hydraulic fluid, these hydrogen dc's would not be easy to make.  My product was fairly simple, if not precise.  Make part to high tolerance at 20c.   Thermal expansion that happens from operating temps (0c-40c?) is handled by O-rings and choice of alloy.  But this... you machine the device at 20c, give or take, and then have it operating in temperatures from >30c (It's Florida) to whatever the god awful temperature of liquid hydrogen is.... of course it's going to leak.  

The problem might not be with the coupling.   That might be at it's engineering limits.   The problem might lie with the protocols and procedures surrounding hydrogen leaks.  They might be too strict. 

58 minutes ago, Exoscientist said:

These quick disconnect fittings that are the source of the problem are quite low tech:

Far from it.   Just looking at the picture I can see from the mirrored finish of the male probe the level of machining that went into this.  That's a an RA of <10.   That's not a low tech piece.

[magnemoe]

15 minutes ago, Gargamel said:

As somebody who has personally machined high tolerance quick disconnects for hydraulic fluid, these hydrogen dc's would not be easy to make.  My product was fairly simple, if not precise.  Make part to high tolerance at 20c.   Thermal expansion that happens from operating temps (0c-40c?) is handled by O-rings and choice of alloy.  But this... you machine the device at 20c, give or take, and then have it operating in temperatures from >30c (It's Florida) to whatever the god awful temperature of liquid hydrogen is.... of course it's going to leak.  

The problem might not be with the coupling.   That might be at it's engineering limits.   The problem might lie with the protocols and procedures surrounding hydrogen leaks.  They might be too strict. 

Far from it.   Just looking at the picture I can see from the mirrored finish of the male probe the level of machining that went into this.  That's a an RA of <10.   That's not a low tech piece.

Yes, you must be able to connect then warm and then disconnect both then its very cold and warm.

Now how about just suck air around it away, think vacuum cleaner integrated into saws and stuff and simply disperse it, the problem as I see it is that it build up and is an fire hazard not that is leaking a lot. 

[farmerben]

We could blow shielding gas like nitrogen or carbon dioxide around the fittings.  I assume these fittings connect the craft to the launch tower, in which case the launch tower is not mass constrained.  

The shielding gas could exhaust several inches away from the fittings because flammability at 10% oxygen is not a factor.  https://en.wikipedia.org/wiki/Flammability_diagram

 

 

[magnemoe]

29 minutes ago, farmerben said:

We could blow shielding gas like nitrogen or carbon dioxide around the fittings.  I assume these fittings connect the craft to the launch tower, in which case the launch tower is not mass constrained.  

The shielding gas could exhaust several inches away from the fittings because flammability at 10% oxygen is not a factor.  https://en.wikipedia.org/wiki/Flammability_diagram

 

An even better solution. 

[mikegarrison]

Hydrogen is notoriously hard to contain. It is the smallest possible atom, and it loves to escape through pretty much any gap. It also actually permeates right through many other materials.

Now let's add in the fun of it being about 20 Kelvin....

[Nuke]

we certainly have come a long way since we had to use cow intestines to store the stuff. 

in ww1 the germans actually banned sausage production so that their air ships had enough gas volume. if you know anything about german cuisine, you know that's a big ask. 

Edited September 13, 2022 by Nuke

[wumpus]

Anyone know how this effects the proverbial "hydrogen car"?  I'm guessing that the "gas pump tank" would be more or less at ground temperature, and the hydrogen would rapidly drop in pressure/temperature down the pipe as it encountered a relatively empty tank.  Temperature fluxuations would also be a thing, but certainly not hundreds of kelvin.

[CBase]

The problem is not leaked hydrogen, but increased concentration in the air around the rocket and associated fire risk.

Wouldn't a constant airflow around the valve prevent the fire risk ? So just putting something like this near the leakage ?

[tater]

Hydrogen is incredibly difficult to deal with in general, and this use case—a quick disconnect has got to be even more tricky.

A friend of mine has a pretty expensive dilution refrigerator in his lab. A single fitting was insufficiently tightened (think they had moved it) and they quickly lost > several thousand bucks worth of Helium (³He/⁴He I think). He is easier to deal with than H, and this was in a setting where the thing is set up and rarely messed with—not something sitting outside that has top pop off in a moment at launch.

Edited September 14, 2022 by tater

[magnemoe]

On 9/14/2022 at 8:07 PM, CBase said:

The problem is not leaked hydrogen, but increased concentration in the air around the rocket and associated fire risk.

Wouldn't a constant airflow around the valve prevent the fire risk ? So just putting something like this near the leakage ?

Not an big fan of a big fan like this at it will likely take damage during launch.

Better to suck air or blow air or nitrogen trough pipes as its simple to make the pipes solid enough to don't care about the launch. 

[CBase]

18 hours ago, magnemoe said:

it will likely take damage during launch.

They are currently on the ridge of wasting millions because the leaked hydrogen kills several certified limits. And you are afraid of loosing a fan ? I mean with 2 billion per launch and only a handful of launches I am not sure that building pipes would ever pay off.

This is the very difference between SpaceX and the rest: Good enough let's you focus on the next challenge. Optimise later.

[tater]

On 9/8/2022 at 8:59 AM, Exoscientist said:

 I was quite startled to read from space reporter Eric Berger that the SLS rocket can only do 2 more rollbacks before it is decertified(!) Huh? A $2 billion spacecraft just gets thrown away, unused!?!

This is the bigger issue, IMO.

It's combined with other poor choices. Clearly they knew that there was possible damage done via roll out/back. It's very tall, so small variations on the ground could possibly damage connections via minute flexing, etc. That said, it's likely actually fine, but they have to be conservative.

Still, knowing this, why design the FTS system such that it can only be checked/updated (battery replaced) in the VAB? Why not design the mobile lancher with swing arms to accommodate whatever routine work would otherwise require a rollback?

My guess is that this goes back to another awful decision, and this is simply the result of that. The awful decision was the very idea of the ICPS.

ICPS was supposed to be flown ONCE. The first, unmanned test flight. You know, to speed development along! That's why SLS flew on time in 2016! Oh, wait.

With EUS, SLS can fly pretty much any day. The TLI burn just happens at a different time of day. They can pick a launch window they like, park the thing in LEO, check it out, then go for TLI on orbit N. With ICPS, SLS puts the ICPS/CSM stage into an elliptical parking orbit, which means the apogee needs to be in the right place at the time of launch. Block 1 is forced to continuously adjust the launch trajectory throughout the launch window to achieve this, and the geometery results in the launch windows posted in the SLS thread. A few days, then some no go days, then a few more days, then a week or 2 of no windows, rinse, repeat.

This massively increases the chance for roll backs to the VAB. Miss a window, sit on the pad for a couple weeks—then the batteries need replacing. Alternately, miss a window, and yikes, here comes a hurricane—back to the barn!

So they went with ICPS which was always stupid, and this is the result. The fact that at least 1 more flight will be with ICPS means we have to deal with this twice... unless EUS is delayed, but that;s impossible, right?

Edited September 19, 2022 by tater

[magnemoe]

2 hours ago, CBase said:

They are currently on the ridge of wasting millions because the leaked hydrogen kills several certified limits. And you are afraid of loosing a fan ? I mean with 2 billion per launch and only a handful of launches I am not sure that building pipes would ever pay off.

This is the very difference between SpaceX and the rest: Good enough let's you focus on the next challenge. Optimise later.

Now this is an valid point the fan would be standard large fans so cheap and very quick to install, much cheaper and quicker than installing piping and it does not matter much if they get destroyed on launch. That is even if you had to use fans who could be used in environments with explosion risks who is an valid point. 

[Exoscientist]

On 9/17/2022 at 12:04 PM, magnemoe said:

Now this is an valid point the fan would be standard large fans so cheap and very quick to install, much cheaper and quicker than installing piping and it does not matter much if they get destroyed on launch. That is even if you had to use fans who could be used in environments with explosion risks who is an valid point. 

 

 That would be embarrassing if the answer were that simple!

 

   Robert Clark

[Gargamel]

On 9/14/2022 at 1:41 PM, wumpus said:

Anyone know how this effects the proverbial "hydrogen car"?  I'm guessing that the "gas pump tank" would be more or less at ground temperature, and the hydrogen would rapidly drop in pressure/temperature down the pipe as it encountered a relatively empty tank.  Temperature fluxuations would also be a thing, but certainly not hundreds of kelvin.

Shouldn’t affect it at all.   If you’re standing there, and can manually release the clamp, it’ll come off.    This has to drop away 100% of the time automatically, and is several orders of magnitude larger than a car sized hose would be, which leads to its own set of issues. 

[Terwin]

On 9/17/2022 at 8:43 AM, CBase said:

They are currently on the ridge of wasting millions because the leaked hydrogen kills several certified limits. And you are afraid of loosing a fan ? I mean with 2 billion per launch and only a handful of launches I am not sure that building pipes would ever pay off.

This is the very difference between SpaceX and the rest: Good enough let's you focus on the next challenge. Optimise later.

How confident are you that an off-the-shelf industrial-strength fan would not have any catastrophic failure modes that could endanger the rocket?  

Steel fan-blades shooting through pipes or poking a hole in the fuel tank would be catastrophic, and I doubt many fans are rated to run while exposed to both evaporating liquid hydrogen and a hydrolox exhaust plume without any risk of throwing a blade.

Putting a fan in a safety-bunker with an exhaust vent that has enough turns to ensure a RUD of the fan could not harm the rocket may be a little more than they are prepared to handle at the moment.

 

[Exoscientist]

On 9/14/2022 at 2:07 PM, CBase said:

The problem is not leaked hydrogen, but increased concentration in the air around the rocket and associated fire risk.

Wouldn't a constant airflow around the valve prevent the fire risk ? So just putting something like this near the leakage ?

 

 With the hurricane threatening to force another rollback, leaving you with only one rollback left before the $2 billion vehicle has to be scrapped for parts,  this becomes of increasing importance to find a solution.  That NASA is dithering on the decision makes it clear the limit on rollback is a serious consideration.

 Even this fan solution if it works as a stop gap measure might be acceptable.  One possible problem though is it might work too well. It might obscure the fact that a serious leak is occurring.

  Robert Clark

Edited September 26, 2022 by Exoscientist

[SunlitZelkova]

2 hours ago, Exoscientist said:

 

 With the hurricane threatening to force another rollback, leaving you with only one rollback left before the $2 billion vehicle has to be scrapped for parts,  this becomes of increasing importance to find a solution.  That NASA is dithering on the decision makes it clear the limit on rollback is a serious consideration.

 Even this fan solution if it works as a stop gap measure might be acceptable.  One possible problem though is it might work too well. It might obscure the fact that a serious leak is occurring.

  Robert Clark

They appear to have fixed the leak, but for all we know the stress from another rollback and rollout will cause another leak.

[Exoscientist]

8 minutes ago, SunlitZelkova said:

They appear to have fixed the leak, but for all we know the stress from another rollback and rollout will cause another leak.

 Good point.

 

  Robert Clark

[tater]

35 minutes ago, SunlitZelkova said:

They appear to have fixed the leak, but for all we know the stress from another rollback and rollout will cause another leak.

Assuming they manage to fly some time soon, they will get to do this all over again in some number of years. Such cadence. Much wow.