On Feb 18, 6:06 pm, M100C <
clbe...@gmail.com> wrote:
> y = -($B$3*(1/COS(RADIANS($B$2))^2)/(2*$B$1^2)*A7^2+A7*TAN(RADIANS($B
> $2)))
>
> Can anyone look this over and see the error in my formula?
Your problem is simple, easy to see and easy to solve. Not everyone
knows Excel here, so your formula is virtually unreadable. But more to
the point: it's unreadable *to you*. The proof is that it's not doing
what you expect, therefore what you think you're reading is unrelated
to what it's actually saying. QED. The second proof is ... well just
look at it!
So, the solution, quite simply, is to translate the formula back into
normal ordinary math and see what you get. Then translate it back into
Excel and see if you get the same thing.
(Ideally, there should be either (a) two separate people involved in
each part of the Double-Translation Protocol, or (b) the same person
but on separate occasions.)
It's the same general method I use to reverse-engineer specifications,
code, binary, programs (which the way most people write programs is
just as bad as reading binary, so I sometimes just skip the code and
read the binary). Add to that: mathematical texts and ordinary
language text.
In flight-critical or other safety-critical systems ... in the US at
least ... one has the DO-178B standard which basically prescribes just
that: strict object-code validation. Your problem with the Excel code
is a perfect reflection of what goes on there. People write unreadable
code, they get tripped off, and pretty soon the vehicles aren't going
on the trajectories they're supposed to be following because the code
isn't doing what you expected it to ... because it's manifestly
unreadable.
If the same person is involved in the Double-Translation Protocol then
item (b) above applies. But if there's only time for "one separate
occasion" then add (c) the same person on one occasion in two separate
ways. So, you should first *separately* formulate just what the
solution is that you're trying to implement.
For typical text-book ballistics, one has 3 formulae:
a = constant, v = v_0 + a t, r = r_0 + v_0 t + 1/2 a t^2
for the vectors a, v and r representing acceleration, velocity and
position respectively. Just plug in the appropriate values from the
boundary conditions
r_0 = 0
v_0 = 700 yd/sec (i (cos 2 deg) + j (sin 2 deg))
a = (-10.728 yd/sec^2) j
i = unit vector pointing horizontally
j = unit vector pointing up
deg = abbreviation for the number pi/180.
The Wikipedia articles on ballistics are far more detailed (and much
better-referenced and far broader) than any textbook reference and are
not required for toy problems in high school or college Physics.
And just to add to this: they're far-better researched than any
textbook, with real-time error tracking/correction, unlike textbooks
and monographs ... even from Cambridge University Press which are
FILLED WITH ERRORS because their publishers are too worried about
making deadlines to actually REVIEW what they publish before
publishing it.
Textbooks are just as bad, if not worse. I've found numerous errors in
my old high school and college texts, for instance. Americal
Mathematical Society, just as bad. North Holland ... put out
monographs that didn't even try to get its math or proofs right.
You're not going to find in a typical text the details on bullet and
gun construction (and how the particular constructions used are
carried out so as to create bullets with stable trajectories, just to
give you an example. Those are all linked to from the Wikipedia, along
with references to the professional research literature.
So, don't let anybody tell you that the Wikipedia is somehow "wacky"
or "substandard" without taking to task their ancient, 20th century
obsolete Baby Boomer laden mentality ("media is something we're fed as
consumers by Old Media content providers, which we sit on our fat
aging asses and passively watch") by demanding of them: "inaccurate
compared to what? Cambridge University Press? AMS' monographs? They
don't correct or log their errors in real-time in cyberspace under
simultaneous review by thousands of knowledgeable ."
I shouldn't have to be constantly calling on them to *review their
books* before they publish them so that I don't have to keep sending
them error lists. I don't have time for that and I'm certainly not
going to be doing their job for them for free!