Using the nominal forward voltage at the two
test currents in Equations #3.4 and #3.5 would
generate the typical linear forward voltage
model as shown below. The nominal linear
forward voltage model (VO nom and RS nom) is
based on the average forward voltages at two
test currents, IF1 and IF2, for a large number of
SuperFlux LED emitters from the same forward
voltage category.
estimated with two permutations of the linear
model as shown in Figure 3.16:
VF min = VO LL + RS LL IF @ VO min + RS min IF
VF max = VO HH + RS HH IF @ VO max + RS max IF
In order to model the variation in electrical
forward characteristics over temperature,
another term can be added to the linear model
as shown in Equation #3.6. Note that the data
shown in AB20-3B represents the forward
voltage at 25°C with the units measured cold (i.e.
TJ = 25°C). Thus, the thermally stabilized forward
voltage at 25°C will be slightly lower than the
values shown in AB20-3B.
(IF1, VF1 nom), (IF2, VF2 nom) Þ ( VO nom, RS nom
)
Then:
VF nom = VO nom + RS nom IF
The values of VF(IF1) and VF(IF2) vary for different
SuperFlux LED emitters from the same forward
voltage category. Statistical forward voltage
data for SuperFlux LED emitters is given in
AB20-3B. Then, the values of VO and RS can be
calculated using the desired limits (i.e. VF max, VF
min, or VF average ± n s ). Worst-case circuit analysis
is concerned primarily with the highest and
lowest forward voltages over the range of IF1 £ IF
£ IF2. In most cases, the worst-case range of
forward current and forward voltage can be
Where:
TJ
= junction temperature, °C
D VF /D = change in VF due to temperature,
@ –2mV°C
VO, RS = measured at a junction temperature
of 25°C
Figure 3.17 Linear Model (m = 1) for
Luminous Flux versus Forward Current for
HPWA-xHOO LED Emitter Shown in Figure
3.10.
Figure 3.18 Exponential Model (k = -0.0110)
and Exponential Curve Fit (k = -0.0096) for
Luminous Flux versus Temperature for
HPWA-xHOO LED Emitter Shown in
Figure 3.12.
18
LUMILEDS [ LUMILEDS LIGHTING COMPANY ]
