
Preferred 100 kW Transient Voltage
Suppressor for AIRCRAFT POWER
BUS PROTECTION
WWW
.Microse
m
i
.CO
M
S C O T TS DALE DIVISION
RT100KP33A thru RT100KP400CA, e3
RT100KP33
–
400CA,
e3
GRAPHS
NOTE: This PPP versus
time graph allows the
designer to use these parts
over a broad power
spectrum using the
guidelines illustrated in
App Note 104 on
Microsemi’s website.
Aircraft transients are
described with exponential
decaying waveforms. For
suppression of square-
wave impulses, derate
power and current to 66%
of that for exponential
decay shown in Figure 1.
P
PP
Peak
Pulse
Power
vs.
Pulse
Time
–
kW
Non-Re
p
etitive
Pulse
Peak
Pulse
Power
(P
PP
)or
cont
inuous
Power
in
%
o
f25
o C
ratin
g
tp – Pulse Time – sec.
TL Lead Temperature oC
FIGURE 1
FIGURE 2
Peak Pulse Power vs. Pulse Time
POWER DERATING
To 50% of Exponentially Decaying Pulse
Correct
Microsemi
Scottsdale Division
Page 3
Copyright
2007
8-22-2007
Wrong
FIGURE 3
FIGURE 5
FIGURE 4
FIGURE 6
INSTALLATION
TVS devices used across power lines are
subject to relatively high magnitude surge
currents and are more prone to adverse
parasitic inductance effects in the mounting
leads. Minimizing the shunt path of the lead
inductance and their V= -Ldi/dt effects will
optimize the TVS effectiveness. Examples
of optimum installation and poor installation
are illustrated in figures 3 through figure 6.
Figure
3
illustrates
minimal
parasitic
inductance with attachment at end of device.
Inductive voltage drop is across input leads.
Virtually no “overshoot” voltage results as
illustrated with figure 4.
The loss of
effectiveness
in
protection
caused
by
excessive parasitic inductance is illustrated
in figures 5 and 6. Also see MicroNote 111
for further information on “Parasitic Lead
Inductance in TVS”.
8700 E. Thomas Rd. PO Box 1390, Scottsdale, AZ 85252 USA, (480) 941-6300, Fax: (480) 947-1503