BRUEL & KJAER 1617 - Page 48

1000
800
630
500
400
250
WRITING 160
SPEED
mm /sec
50 mm
Writing
Width
100
63
40
25
16
8
4
2
<....... < / /
IX /,
-<.... /
/V" )
/ x /
Y /'X A
\!'&>)-
' <09> ,
" <<x
'V
I I I I I III
I
I I I I
III
2000
1600
1250
1000
800
500
t>~
M ).
IV)SOA
V~/ ,
~<0<; - A
'1'0;
k/A
~ r®>
~®A
315
WRITING
SPEED
200
mm/sec
125
100 mm
Writing
80
Width
50
31.5
16
8
4
I
I I I I
III I I I I I I II
0. 001
0. 01 0.1
1.0
10.0
Effective Aver ag ing Tim e (seco nd s)
173109
Fig.4 .2 . Effective averaging times as a function of the Level Recor-
der writing speed
Averaging Time
0,1
0,3
1 3 10 30 100 300
T A (s)
Paper speed for
DC recording
10 3
1
0,3 0,1
0,03 0, 01 0,003
(mm/s)
770409
Table 4.2. Paper speeds for DC recording
paper speeds for the normal range of averaging times, based on the assumption that the
time required to sweep over a bandwidth is equal to five averaging times. With the Type
1617 , normally the step between filters will be attained in less than 2mm (i .e. 40% of
the bandwidth), and even in the very worst case of a single sinusoidal component lo-
cated between two filters, it will be reached within 75% of the bandwidth.
Where the spectrum is known to be relatively flat with a maximum fall less than 5 dB be-
tween adjacent filter bands, it is possible to increase paper speed by one step (i .e. a fac-
tor of approximately 3). The time taken to sweep over one bandwidth will be approxi-
mately 1,6 times the averaging time, which is about the minimum required for the detec-
tor to respond, even on peaks.
The same basic
considerations apply for AC recording, except that it will be the writing
speed rather than the averaging time wh ich determines how rapidly the recording pen
can respond to a
new level. Table 4.3 gives suitable paper speeds for the range of writ-
43