Averaging
Time
Control
This
ability,
allows
the
Type
1617
to vary
the
averaging
time
of
the
Measuring
Amplifier
Type
2607
or
the
Narrow
Band
Analyzer
Type
2120
during
the
course of
the
fre-
quency
scan. The advantage is
that
the
averaging
time
can be kept as
short
as
possible
to
maintain
an ac-
ceptable
analysis
time
,
while
being
long enough at
lower
frequencies
to
achieve
optimum
measuring
accu-
racy, i.e.
to
maintain
a good
confi-
dence level. The best analysis
condi-
tions
are obtained
when
the
product
of
bandwidth
B (in
Hz)
and averag-
ing
time
T (in
s)
is held
constant
throughout
the
scan.
Therefore
the
Type
1617
is equipped
with
averag-
ing
time
programs
which
can be se-
lected according
to
test
conditions
to
maintain
the
BT
product
as
near
constant
as possible and give a con-
stant
confidence level.
At
higher
frequencies,
if
a
con-
stant
BT
product
is
to
be
main-
tained
,
the
averaging
time
becomes
so
short
that
the
writing
speed
of
the
Recorder
will
be
the
practical li-
mitation
on
the
speed of
the
analy-
sis.
Therefore
the
shortest
averag-
ing
time
that
can be selected by
the
programs in
the
Type
1617
is
0,1 s.
However
, some
high
fre-
quency
signals
like
the
noise or vi-
bration
from
a
slowly
rotating
ma-
chine
may have a
low
repetition
fre
-
quency
that
causes
low
frequency
modulation
as
shown
in
Fig.7,
where
0, 1 s averaging
time
is
too
short
. To
allow
satisfactory
averag-
ing
times
to
be selected
for
this
type of signal, a choice
of
three
min
-
imum
times
(0
,1 s, 1,0 s, and
10
s)
is provided on
the
Type
1617
. Fig .8
is a
chart
indicating
the
averaging
times
and
change-over
frequencies
in
the
nine
programs
available,
to-
gether
with
the
minimum
and
maxi-
mum
averaging
times.
When
operating
with
variable av-
eraging
times
, selection
of
the
re-
quired
program
is made
with
the
two
Averaging
Control
switches
on
the
front
panel, and
the
Recorder
Averaging
Time
switch
on
the
rear
panel
must
also be set
to
"Variable
Averaging
Time".
Table 3
shows
how
the
programs
can be selected
by suitable
combination
of
the
Pro-
gram
and
Minimum
Time
switch
po-
sitions. The
programs
are
obtained
6
A
ve
raging
Time
(S)
100
C,F.J Slow
Random
30
10
S
G,H,J
10
3
0,3
0,1
L-----~-
___
D
__
-,~,-----------·.------------'B=~--=~F~~~-,~~~~~~~~--C--D~;~:~.:
A,B,C
, ! ,
! , , I ! ,
2 5
10 20
50
100 200 500
lk
2k
5 k 10 k 20 k
Sw
itchover Frequency (Hz)
Fig
.S.
Averaging
times
and
changeover
frequencies
Minimum
Av
.
Ti
me(s) 0 ,1 1,0
10
Slow
Rand
om
C
F J
Fa
st Ra
nd
om
B
E H
Sin
e A 0
(G)
Table
3.
Selection
of
the
required
averaging
time
program
via
the
15
lead AVERAGING TIME
CONTROL
socket on
the
rear panel.
With
a Level Recorder,
automatic
averaging
time
control
functions
as
follows
. The scan
starts
with
the
Re-
corder being
held
for
a period
of
five
times
the
programmed
averag-
ing
time
for
the
band,
during
which
time
the
pen comes up
to
the
cor-
rectly averaged level.
Then
a long
averaging
time
is selected
that
keeps
the
output
constant
while
the
Recorder
draws
the
trace
by advan-
cing
the
paper at
high
speed
to
the
next
frequency
band.
The
Filter
steps
to
the
next band in
the
se-
quence
, and
the
programmed
aver-
aging
time
is selected,
after
which
the
Recorder hold period begins
again.
Using an X-V Recorder,
where
the
X-axis is
controlled
by
the
DC
ramp
output
from
the
Type 1 61 7 ,
the
Recorder pauses
for
five
times
the
averaging
time,
and
then
takes
eight
steps
through
each
third-
octave band.
Figs.9
,
10
, and
11
show
analysis
of
the
same
input
signal recorded
using
three
different
methods
to il-
lustrate
their
relative
merits
.
Built
-
in
IEC
Interface
The
Type
1617
is
fully
program
-
mable
via a
built
-
in
interface
that
sat-
isfies
the
requirements
of
the
pro-
posed
IEC
Standard
Digital
Interface
for
Programmable
Instrumentation
.
This
permits
the
filter
frequency
bandwidths,
and averaging
time
to
be varied in an
arbitary
sequence,
and recording
initiated
to
suit
any
particular
measuring
program. It
also
permits
on-
line
changes to be
made to accomodate
new
events as
they
occur
.
-
Br
uel & Kj
ae
r
Br
uet &
Kjm
r
Sru
et &
Kje
r
DO
OOOOO
DOaOOQOoaOOOOODOOOOOODDOOOOODOODDOOOODDODD
~
-
ruel
&
Kjmr
P
otentiomet
er
R
a
n
ge
:
~
dB
R
ec
t
i
f
ier
:
~
l
ower
lim
.
Freq
.:
~
Hz
Wr.
Speed
:~
mm
/
5ec.
Pap
e r
~
e
d
:-L
mm
/ .e c
C
open
hag
en
50
2
1075
dBdS
40
2050
~
eas
u'ing
Ob
j.:
Fi
xed
---
~~;~ag;ng
-
3 r
s
40
1/3
octave-
Time
used
250 s
20
f,
0 30
~~
~.'
B
~
~~
~~f"
---
1 20
~
Date
-==
5
ign
.:
__
_
10
20
'
'tt
z
50
1
00
200
OP
1124
Multipl
v Fre
qu
ency
Scal
e
by
Ze
ro
l
eve
l:
1612
/ 2
112
Fig.9
.
Frequency
analysis
with
Level
Recorder
and
fixed
averaging
time
d dB
60
3C
15
ABC
76
1042
To Next Page
Loading...
