beautypg.com

Appendix c. theory and additional details, C.1 swept frequency theory, C.2 additional theory on multiplier and offset – Campbell Scientific AVW1 and AVW4 Interfaces for Vibrating Wire Sensors User Manual

Page 33

background image

C-1

APPENDIX C. THEORY AND ADDITIONAL DETAILS

C.1 SWEPT FREQUENCY THEORY

Example:

f2 = 31 hundred Hz
f1 = 24 hundred Hz

X = (30*f1*f2)/(f2-f1)
= 3189

where f1 and a f2 are the starting and ending
frequencies in hundreds of Hz respectively.
One clock cycle (CC) occurs every 813.8 ns or
at a rate of 1.2288 mHz.

Rule #1: 65535 > X > 256

the 256 constraint is somehow due to an 8 bit
constraint. The 65535 constraint is some limit
where the swept frequency can no longer be
done in exactly 15 ms.

The minimum increment in frequency is 1 clock
cycle.

How many times do we have to change the half
period by 1 clock cycle to cover the frequency
range?

1/(F1 * 2) = 1/(2400 * 2) =
.208333 ms = half period

1/(F2 * 2) = 1/(3100 * 2) =
.161290 ms = half period

Change in half period =
.208333 - .161290 = .047043 ms

# of clock cycle increments to cover the
.047043 ms half period range is:

= (.047043 ms/.0008138 ms) =
57.81 freq. increments

How much time is there between frequency
increments to cover the frequency range in 15
ms?

time = 15 ms/58 frequency increments =
.2586 ms/freq. incr.

C.2 ADDITIONAL THEORY ON

MULTIPLIER AND OFFSET

The result (X) of Instruction #28 is:

X = 1/((t ms)2) =
1,000,000/((t s)2

where t is the period in milliseconds. Since
frequency (f) is the inverse of period, this can
also be expressed as:

X = (f kHz)2 =
[(f Hz)2]/1,000,000