Everyone, thank you for your replies so far. It's a very informative
discussion.
@Bill I appreciate your perspective and your experience, but with respect
this project is not about pride. Also, I strongly believe there is value in
having ambitious (or perhaps in some views, unrealistic) goals. They are a
source of inspiration, they force us to attack problems creatively, and
they usually bring out the best in the strong competitors amongst us. A [n
ambitious] "goal will serve to organize and measure the best of our
energies and skills" as a more skilled orator once put it.
@Bob Thanks for the information, as always. I think our LOX main valve
weighs more than 8 lb just by itself! I'm still very curious on how the
data is spread on feed system mass vs p_ch and thrust for small rockets in
general. My intuition is that there will be a trend.
@David W. Thanks for the advice. This is certainly not a 1-year project (or
likely even a 5-year project). PSAS's project is a long term one (it's not
a typical university rocket club). In case you're curious, you can find out
about some of our more recent projects here: http://psas.pdx.edu/projects/. It
is my hope that there will be many versions of the LF rocket, and that
eventually one of them will make the goal. In the past our group has flown
high powered rockets with COTS solids (usually N's) and has mostly focused
on developing avionics. More recently we've started working on propulsion
projects, including a 2 kN 3D printed engine, which we designed and built
this past spring, and which we are planning to static test in late
September or early October.
@Ignacio I strongly agree with your main point, but I should stress that
the purpose of this little script is neither a faithful estimation of cost,
nor is it strictly speaking really a design optimization; that would
certainly be premature. As we're all well aware, by nature, the aerospace
design process is highly iterative. This python script is a glorified excel
trajectory program, of the type many interested rocketeers have
experimented with at some point in their lives; hardly a grand programmable
schema as Bill so colorfully put it. It merely provides a starting point in
the iterative process, and while I admittedly start from a simple premise
(that minimizing GLOW is a Good Thing) I maintain no illusions as to its
predictiveness as a model. For a general iterative process a wild guess, or
personal intuition for an initial design can sometimes suffice, but I
notice that many projects ultimately suffer from cost or schedule issues
due to poor choices that were committed to early in the design process. And
as you mention, students may work for free, but they do not have unlimited
time. There are still no free lunches, and opportunity costs are
associated with all of our decisions. We have limited resources and need to
guess at the best way to allocate them. My goal with this question is to
help in figuring out requirements for future projects. I imagine that some
working engineers here have at some point experienced frustrations during
design projects with inadequately specified/ambiguous/poorly-understood
reqs., but this problem is probably even more acute with engineering
students (due to lack of experience among other things). While they are not
impossible to build, we (I) understand very little about 100km LF sounding
rockets. As I see it this is part of the process of learning more and
outlining the problem.
On Sun, Aug 21, 2016 at 5:49 PM ignacio belieres <
ignacio_belieres@xxxxxxxxxxx> wrote:
IMO the only way to faithfully estimate cost is through many bottom upAd astra
cost estimates. Some may say otherwise, but parametric models are
constrained to similar vehicles and design philosophies, your project does
not fall into this category. Your cost drivers are entirely different to
everyone else's.
My advice is that you read more on rocket design, do a first draft on each
subsystem and manually iterate until you are comfortable with the
performance. Optimization is useless without the proper models, and those
come only after a loooot of work and reading. You should also understand
that, while for you time may be free, its still limited.
Anyways, good luck!
IB
------------------------------
De: Troy Prideaux <troy@xxxxxxxxxxxxxxxxxxxxx>
Enviado el: 21/08/2016 20:14
Para: arocket@xxxxxxxxxxxxx
Asunto: [AR] Re: LF Feed-Sys Mass Survey for a LV Design Optimization
Project
From my experience with solids, cost does scale with size or total impulse,
but that's solids and the same probably wouldn't hold with the hybrids I've
made. So, I'd be very reluctant to attribute any particular rule or guide
to
it as it probably depends a fair bit on the construction methods and lots
of
other stuff, but I can certainly see the theory behind the argument.
Troy
-----Original Message-----
From: arocket-bounce@xxxxxxxxxxxxx [mailto:arocket-bounce@xxxxxxxxxxxxx
<arocket-bounce@xxxxxxxxxxxxx>] On
Behalf Of Zachary Martinez
Sent: Monday, 22 August 2016 9:02 AM
To: arocket@xxxxxxxxxxxxx
Subject: [AR] Re: LF Feed-Sys Mass Survey for a LV Design Optimization
Project
For student groups the engineering and development cost is essentially
nothing.
Zach Martinez
On Aug 21, 2016, at 5:04 PM, David Weinshenker <daze39@xxxxxxxxxxxxx>wrote:
mass - in fact I would use dry mass not glow but perhaps in narrow cases it
On 08/20/2016 10:27 PM, David Gregory wrote:
More stuff means more money, period. Cost scales almost linearly with
doesn't matter which.
That assumes that the sheer quantity of consumed raw material is the
principal cost driver, rather than the engineering and development
costs.
This is arguably not the case - see e.g., Truax, "The Future of Earth-
to-Orbit Propulsion".
-dave w
--
