Re: Analysis of gas mode MM interferometer operation using standard SR formulae.

On May 6, 8:22 pm, Jerry <Cephalobus_alie...@xxxxxxxxxxx> wrote:
On May 6, 3:02am, Surfer <n...@xxxxxxxx> wrote:
On Mon, 05 May 2008 22:08:42 -0500, Tom Roberts

Jerry misspoke slightly -- the result is forced to be NEAR zero at
markers 1 and 16 (it is near zero, not exactly zero, because the 20
values for marker "17" are not quite the same as the 20 values for
marker 1 [19 of them are identical]). But Miller then averaged the two
1/2 turns, and COPIED the first value to the 9th position. THAT makes
the resulting graph be periodic with period 1/2 turn,

Here are some counter examples:

Averaged raw data =
{10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10, 90, 10}
Output =
{10., 90., 10., 90., 10., 90., 10., 90., 10.}

Pathetic. This is simultaneously periodic over 1/2, 1/4,
and 1/8 turn.

So is the raw data. The point is the processing has NOT produced any
EXTRA component periodic in half a turn.

Averaged raw data=
{10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150,
160, 170}
Output=
{10., 10., 10., 10., 10., 10., 10., 10., 10.}

Big swings in raw data values, but they do not cause outputs periodic
in half a turn.

Doubly pathetic.

The point is a linear temperature drift does not produce any EXTRA
component periodic in half a turn.

Here is a simulated linear temperature drift plus signal.

Averaged data=
{19, 20, 39, 40, 59, 60, 79, 80, 99, 100, 119, 120, 139, 140, 159,
160, 179}

Output =
{10.`, 90.`, 10.`, 90.`, 10.`, 90.`, 10.`, 90.`, 10.`}

Again nothing EXTRA periodic in half a turn has been added by the
processing.


and the zero is
constrained to be near the endpoints, but not constrained to be
precisely at an endpoint. And this is EXACTLY what his Fig 8 (my fig 1)
shows.

Here is a counter example:

Averaged raw data =
{30, 20, 10, 0, 10, 20, 30, 40, 30, 20, 10, 0, 10, 10, 20, 30, 30}
Output =
{30., 20., 10., 0., 10., 15., 25., 35., 30.}


Despite your carefully engineered "raw data", note the beautiful
periodicities:

35 35
30--------------30--------------30--
25 25
20 20
15 15
10 10 10 10

00 00

This would simulate a case where the velocity vector happened to have
the same value when measurements were made at Marker 1, Marker 9 and
at Marker 1 again. Unusual but not impossible.

The point is, contrary to Tom Robert's claim, the zero is NOT
constrained to be near the end points.


-- Surfer

.

Relevant Pages
Quantcast