From: Bruce Dunn <bpdunn@home.com>
Newsgroups: sci.space.tech
Subject: Re: liquid hydrogen - how long to boil off
Date: Sat, 13 Jan 2001 23:37:54 GMT

A NASA sponsored study in the early 1980's, the question of storing
hydrogen and oxygen on-orbit for refueling of a reusable orbital
transfer vehicle was examined.  If I remember correctly (I can't find my
copy of the report at the moment), the storage loss for hydrogen in
tanks encased in multilayer insulation was estimated to be on the order
of 1% per month.

David Cogen wrote:
>
> Didn't Zubrin's plan for Mars expeditions include storing liquid hydrogen for the
> entire duration, or at least the several months of outbound transit plus the
> time till the next supply ship arrived?
>
> --DavidC

From: Bruce Dunn <bpdunn@home.com>
Newsgroups: sci.space.tech
Subject: Re: liquid hydrogen - how long to boil off
Date: Tue, 16 Jan 2001 08:27:07 GMT

Jeff Greason wrote:
>
> Bruce,
>   Do you have a citation for this?  Looks like a good study to add to the
> shelf.
>                 -- Jeff Greason
>                     XCOR Aerospace


After much thrashing around and digging into piles of material, I found
the study.  I should have kept looking when I failed to find it the
first time I posted....

Future Orbital Transfer Vehicle Technology Study, Volume II - Technical
Report
NASA Contractor Report 3536
NASA 1982  (about 240 pages, soft cover)

Study was carried out by Boeing, with Eldon E. Davis as Study Manager.

Estimated boiloff rates for hydrogen and oxygen propellant storage
tanks, in low earth orbit, are as follows:

Liquid Hydrogen:  0.127% per day, 3.81 % per month
Liquid Oxygen:    0.016% per day, 0.49 % per month

These values are higher than in my previous post, but are freshly
calculated from the data in the report.

Conditions:  A tank set for on-orbit storage of propellants for a
reusable orbital transfer vehicle contains a total of 59400 kg of liquid
oxygen and hydrogen, at a mixture ratio of 6 to 1 (given data).  This
equates to 8486 kg hydrogen, and 50,914 kg oxygen, before boiloff.  The
study assumed multilayer insulation consisting of 50 layers of double
aluminized kapton (0.15 mils thick), separated by dacron net.  Heat
leakage includes that through the insulation, as well as through
tank/shell struts and fill, feed and vent lines.  The heat leakage
causes a boiloff in this specific tank set of 0.45 kg/hour of liquid
hydrogen, and 0.36 kg/hr of liquid oxygen.  Given the loading of each
propellant, this translates into the percentage losses seen above.  It
should be noted that boiloff is governed by heat leakage, and the rate
per hour does not depend on the amount of propellant in the tanks.  With
partly filled tanks, the percentage loss per day or month would be
higher.

As an additional data point, a graph of boiloff vs. layers of insulation
shows that if only 10 rather than 50 layers of insulation are used, the
hydrogen boiloff rate is about 2.2 kg/hr rather than 0.45, and the
oxygen boiloff rate is about 1.6 kg/hr rather than 0.36.

Newsgroups: sci.space.tech
From: henry@spsystems.net (Henry Spencer)
Subject: Re: Cryogenic boiloff rates question
Date: Tue, 30 Jan 2001 21:00:43 GMT

In article <3A77149B.2655DE44@home.com>, Bruce Dunn  <bpdunn@home.com> wrote:
>...In other words, the boiloff vapor from the hydrogen tank
>can be used to keep the liquid oxygen from boiling off entirely.

Indeed, the thermal conductivity of the S-IVB's common bulkhead was chosen
so that heat input to the LOX tank through its walls was roughly cancelled
by cooling through the bulkhead, to minimize LOX boiloff at the expense of
a bit more LH2 boiloff.
--
When failure is not an option, success  |  Henry Spencer   henry@spsystems.net
can get expensive.   -- Peter Stibrany  |      (aka henry@zoo.toronto.edu)

From: Bruce Dunn <bpdunn@home.com>
Newsgroups: sci.space.tech
Subject: Re: Cryogenic boiloff rates question
Date: Thu, 01 Feb 2001 20:23:39 GMT

Henry Spencer wrote:

> Indeed, the thermal conductivity of the S-IVB's common bulkhead was chosen
> so that heat input to the LOX tank through its walls was roughly cancelled
> by cooling through the bulkhead, to minimize LOX boiloff at the expense of
> a bit more LH2 boiloff.

Although an interesting strategy, this isn't quite what I had in mind.
In the scheme quoted by Henry,  the heat input from the LOX tank is
added to the heat input of the LH2 tank.  As noted, this increases LH2
boiloff.  What I was proposing was that the "waste" H2 vapor produced by
the heat leakage of the LH2 tank alone be used to cool the LOX tank "for
free".

Having made this suggestion, I can see that is is inferior to the scheme
of using the "waste" H2 vapor to cool the shielding in some way.  Thanks
to Steve Willner for relaying information about this approach!  Steve -
can you quote any boiloff rates or heat leakage rates using this clever
scheme, as well as elaborate on how mechanically it is implemented...

Newsgroups: sci.space.tech
From: henry@spsystems.net (Henry Spencer)
Subject: Re: Cryogenic boiloff rates question
Date: Tue, 30 Jan 2001 21:03:01 GMT

In article <3A77149B.2655DE44@home.com>, Bruce Dunn  <bpdunn@home.com> wrote:
>...study assumed multilayer insulation consisting of 50 layers of double
>aluminized kapton (0.15 mils thick), separated by dacron net...

One caution here:  that insulation is going to have to be behind a fairing
or some other outer protective structure for launch from Earth.  It's not
mechanically strong, and in general it has to be treated carefully if it
is to work properly (compressing it, even gently, greatly reduces its
insulating value).
--
When failure is not an option, success  |  Henry Spencer   henry@spsystems.net
can get expensive.   -- Peter Stibrany  |      (aka henry@zoo.toronto.edu)

Newsgroups: sci.space.tech
From: willner@cfa183.harvard.edu (Steve Willner)
Subject: Re: Cryogenic boiloff rates question
Date: 30 Jan 2001 16:12:37 -0500

In article <3A77149B.2655DE44@home.com>, Bruce Dunn <bpdunn@home.com> writes:
>   I also note that the hydrogen tank is
> venting 0.45 kg/hour of very cold vapor, at just above the boiling point
> of liquid hydrogen.  This vapor has a very high heat capacity, 12.24
> kJ/kg*K.  If the hydrogen vapor were conducted (very carefully !)
> through a heat exchanger inside the liquid oxygen tank, it would be
> warmed from approximately 20 K to approximately 90 K (exact temperatures
> depending on the pressure the tanks are maintained at).  This has the
> capability of providing as much as (90-20)*12.24=857 kJ/hour of
> refrigeration.  In other words, the boiloff vapor from the hydrogen tank
> can be used to keep the liquid oxygen from boiling off entirely.

Better yet, use vapor cooled shields instead of MLAR.  Most of the
parasitic heat load is carried away with the vapor instead of being
conducted into the liquid.  This is now a standard trick in all sorts
of cryo applications, including space uses.  The idea and first
practical implementation is usually credited to Frank Low, ca. 1960.

--
Steve Willner            Phone 617-495-7123     swillner@cfa.harvard.edu
Cambridge, MA 02138 USA
(Please email your reply if you want to be sure I see it; include a
valid Reply-To address to receive an acknowledgement.  Commercial
email may be sent to your ISP.)