13
LT1737
1737fa
APPLICATIO S I FOR ATIO
WUUU
which reduces the size of the primary-referred flyback
pulse used for feedback. This will increase the output
voltage target by a similar percentage. Note that unlike
leakage
spike
behavior, this phenomena is load indepen-
dent. To the extent that the secondary leakage inductance
is a constant percentage of mutual inductance (over
manufacturing variations), this can be accommodated by
adjusting the feedback resistor divider ratio.
Winding Resistance Effects
Resistance in either the primary or secondary will act to
reduce overall efficiency (P
OUT
/P
IN
). Resistance in the
secondary increases effective output impedance which
degrades load regulation, (at least before load compensa-
tion is employed).
Bifilar Winding
A bifilar or similar winding technique is a good way to
minimize troublesome leakage inductances. However, re-
member that this will increase primary-to-secondary ca-
pacitance and limit the primary-to-secondary breakdown
voltage, so bifilar winding is not always practical.
Finally, the LTC Applications group is available to assist
in the choice and/or design of the transformer. Happy
Winding!
SELECTING FEEDBACK RESISTOR DIVIDER VALUES
The expression for V
OUT
developed in the Operation sec-
tion can be rearranged to yield the following expression for
the R1/R2 ratio:
RR
R
V V I ESR
V
N
OUT F SEC
BG
ST
12
2
+
()
=
++
()
•
where:
V
OUT
= desired output voltage
V
F
= switching diode forward voltage
I
SEC
• ESR = secondary resistive losses
V
BG
= data sheet reference voltage value
N
ST
= effective secondary-to-third winding turns ratio
The above equation defines only the ratio of R1 to R2, not
their individual values. However, a “second equation for
two unknowns” is obtained from noting that the Thevenin
impedance of the resistor divider should be roughly 3k for
bias current cancellation and other reasons.
SELECTING R
OCMP
RESISTOR VALUE
The Operation section previously derived the following
expressions for R
OUT
, i.e., effective output impedance and
R
OCMP
, the external resistor value required for its nominal
compensation:
R ESR
DC
RK
R
R
RR
OUT
OCMP
SENSE
OUT
=
⎛
⎝
⎜
⎞
⎠
⎟
=
⎛
⎝
⎜
⎞
⎠
⎟
()
1
1
112
–
||
While the value for R
OCMP
may therefore be theoretically
determined, it is usually better in practice to employ
empirical methods. This is because several of the required
input variables are difficult to estimate precisely. For
instance, the ESR term above includes that of the trans-
former secondary, but its effective ESR value depends on
high frequency behavior, not simply DC winding resis-
tance. Similarly, K1 appears to be a simple ratio of V
IN
to
V
OUT
times (differential) efficiency, but theoretically esti-
mating efficiency is not a simple calculation. The sug-
gested empirical method is as follows:
Build a prototype of the desired supply using the eventual
secondary components. Temporarily ground the R
CMPC
pin to disable the load compensation function. Operate the
supply over the expected range of output current loading
while measuring the output voltage deviation. Approxi-
mate this variation as a single value of R
OUT
(straight line
approximation). Calculate a value for the K1 constant
based on V
IN
, V
OUT
and the measured (differential) effi-
ciency. These are then combined with R
SENSE
as indicated
to yield a value for R
OCMP
.
Verify this result by connecting a resistor of roughly this
value from the R
OCMP
pin to ground. (Disconnect the
ground short to R
CMPC
and connect the requisite 0.1µF
filter capacitor to ground.) Measure the output impedance
with the new compensation in place. Modify the original
R
OCMP
value if necessary to increase or decrease the
effective compensation.
