4
dc1717afa
DEMO MANUAL DC1717A
The second summand, t
pVALID(OFF)
, should be taken into
account if the associated LTC4417 input does not have
any bypass capacitor and the rail can be disconnected
from the input instantly.
The third one must be calculated as:
t
GATE _ THRES
=
R
S
•
C
S
–In 1–
V
THRES
V
SINK
⎛
⎝
⎜
⎞
⎠
⎟
⎡
⎣
⎢
⎤
⎦
⎥
(2)
It is possible to determine the minimum capacitive load
required to hold the output up during switchover as a:
C
LOAD(MIN)
≥
I
LOAD(MAX)
•
T
DISCON
V
OUT(DROOPMAX)
(3)
where:
•
I
LOAD(MAX)
is the maximum load current, A
•
V
OUT(DROOPMAX)
is the maximum acceptable voltage
droop, V
Design proceDure for moDification of Dc1717a
R
S
•
C
S
–In 1–
V
THRES
V
SINK
⎛
⎝
⎜
⎞
⎠
⎟
⎡
⎣
⎢
⎤
⎦
⎥ =
10µs
Calculate R
S
with C
L(INIT)
Recalculate C
L
with R
S
Done.
Use C
L(INIT)
and R
S
No
Is recalculated C
L
lower than initial
C
L(INIT)
?
As shown in the equation (3), the use of external slew rate
control will add additional delay to the total switchover
time. Unfortunately, the actual components cannot be
chosen until the load capacitance is known. This circular
issue can only be resolved through an iterative process.
The process starts by calculating the C
LOAD(MIN)
, assuming
that t
GATE_THRES
=10μs. For clarity this value will be labeled
C
LOAD(INIT)
. Using the calculated C
LOAD(INIT)
, calculate R
S
from the expression of the T
DISCON
. To ensure the newly
calculated R
S
based on C
LOAD(INIT)
is sufficient, calculate
C
LOAD
with the calculated R
S
.
If C
LOAD(INIT)
(the initial calculated C
LOAD
) is higher than
the newly calculated C
LOAD
then the process is completed.
If the C
LOAD(INIT)
is lower than the newly calculated C
LOAD
,
calculate R
S
using the higher value and repeat this process.
Figure 2
