MIL- STD-883F 2004 TEST METHOD STANDARD MICROCIRCUITS - 第143页

MIL-STD-883F METHOD 1021.2 15 November 1991 5 3.4 Test cir cuit preparat ion . The tes t ci rcui t shal l be ass embled inc luding a t est c irc uit boar d, li ne driver s, el ectr ical inst ruments , func tional t est e…

100%1 / 708
MIL-STD-883F
METHOD 1021.2
15 November 1991
4
b. Topological analysis approach. If a photomicrograph of the circuit is available, the number of required states can
be reduced by examining the topology of the internal circuits. This allows one to eliminate the need to test paths
with the same output state which have identical internal geometries. For the counter, this reduces the required
number of states to two. This approach is recommended for more complex circuits where the multiple output logic
approach results in too many required state vectors.
3.2 Transient output upset criteria
. The transients that are permitted at logic outputs depend on the way that the system
application allocates the noise margin of digital devices. Most systems use worst-case design criteria which are not directly
applicable to sample testing because the samples represent typical, not worst-case parts, and have higher noise margins.
For example, although the logic swing of TTL logic devices is typically greater than 2 volts, the worst-case noise margin is
specified at 400 mV. In a typical system, much of this noise margin will be required for aberrations and electrical noise,
leaving only part of it, 100 mV to 200 mV, for radiation-induced transients. Thus, the allowable voltage transient is far lower
than the typical logic signal range. Loading conditions also have a large effect on output transients.
However, transient upset testing is usually done at a fixed temperature under conditions that are more typical than they are
worst-case. Thus, the noise margin during testing is much greater. The recommended default condition if not specified by
the system is a transient voltage exceeding 1 V for CMOS or TTL logic devices with 5 V (nominal) power supply voltage, and
30 percent of the room-temperature logic level swing for other technologies such as ECL, open collector devices, or
applications with other power supply voltages. Default loading conditions are minimum supply voltage and maximum fanout
(maximum loading).
The time duration of transient upset signals is also important. If the duration of the transient voltage change is less than the
minimum value required for other circuits to respond to it, the transient signal shall not be considered an upset. The
minimum time duration shall be one-half the minimum propagation delay time of basic gate circuits from the circuit
technology that is being tested.
Testing criteria may also be established for other parameters, such as the power supply current surge. Output current is
also important for tri-state or uncommitted ("open-collector") circuits. These criteria must be specified by the test plan, and
are normally based on particular system requirements.
3.3 Test plan
. The test plan must include the following:
a. Criteria for transient voltage upset, output current, and power supply current, as applicable.
b. Power supply and operating frequency requirements.
c. Loading conditions at the outputs.
d. Input voltage conditions and source impedance.
e. Functional test approach, including dynamic upset, if applicable.
f. Radiation pulse width(s).
g. Sequence used to adjust the dose rate in order to determine the upset threshold by successive approximation.
h. State vectors used for testing (determined from 3.1).
i. Radiation levels to be used for transient response measurements, if applicable.
j. A recommended radiation level at which to begin the test sequence for transient upset measurements, if
applicable.
k. The temperature of the devices during testing (usually 25°C ±5°C).
MIL-STD-883F
METHOD 1021.2
15 November 1991
5
3.4 Test circuit preparation
. The test circuit shall be assembled including a test circuit board, line drivers, electrical
instruments, functional test equipment, transient measurement equipment, and cables to provide the required input biasing,
output monitoring, and loading.
3.5 Facility preparation
. The radiation source shall be adjusted to operate in the specified mode and provide a radiation
pulse width within the specified width range. The required dosimeters shall be installed as close as practical to the device
under test. If special equipment is needed to control the temperature to the value specified in the test plan, this equipment
must be assembled and adjusted to meet this requirement.
3.6 Safety requirements
. The health and safety requirements established by the local Radiation Safety Officer or Health
Physicist shall be observed.
3.7 Test circuit noise check
. With all circuitry connected, a noise check shall be made. This may be done by inserting a
resistor circuit in place of the test device. Resistor values chosen shall approximate the active resistance of the device
under test. A typical radiation pulse shall be applied while the specified outputs are monitored. If any of the measured
transient voltages are greater than 10 percent of the expected parameter response, the test circuit is unacceptable and shall
not be used without modification to reduce noise.
3.8 Bias and load conditions
. Unless otherwise specified, the power supply shall be at the minimum allowed value. Input
bias levels shall be at worst-case logic levels. Outputs shall be loaded with the maximum load conditions in both logic states
(usually equivalent to maximum circuit fanout).
3.9 Temperature
. The temperature of the devices during test should be measured with an accuracy of ±5°C unless
higher accuracy is required in the test plan.
3.10 Procedure for dose rate upset testing
. The device to be tested shall be placed in the test socket. The required
pulse sequence shall be applied so that the device is in the state specified by the first of the state vectors in 3.1.
a. Set the intensity of the radiation source to the first radiation test level specified in the test plan. Expose the device
to a pulse of radiation, and measure the transient output responses and power supply current transient. For
sequential logic circuits, perform a dynamic functional test to see if changes occurred in internal logic states.
b. Repeat 3.10a for all other state vectors and radiation levels specified in the test plan.
3.11 Radiation exposure and test sequence for upset threshold testing
. The device to be tested shall be placed in the
test socket. The required pulse sequence shall be applied so that the device is in the state specified by the first of the state
vectors determined in 3.1, or is operating with the specified test vector sequence for dynamic upset.
a. Set the intensity of the radiation source to the initial level recommended in the test plan, and expose the device to
a pulse of radiation. Determine whether a stored data upset, logic state upset, or dynamic upset occurs, as
appropriate.
b. If no upset occurred, increase the radiation level according to the sequence specified in the test plan; if an upset
is observed decrease the radiation level. After the radiation source is adjusted to the new intensity, reinitialize the
part to the required state vector, expose it to an additional pulse, and determine whether or not upset occurred.
Continue this sequence until the upset response threshold level is bracketed with the resolution required in the
test plan.
c. The power supply peak transient current shall be monitored and recorded during radiation testing unless it is not
required by the test plan.
d. Repeat test sequences 3.11a through 3.11c for all of the state vectors.
MIL-STD-883F
METHOD 1021.2
15 November 1991
6
3.12 Report
. As a minimum the report shall include the following:
a. Device identification.
b. Test date and test operator.
c. Test facility, radiation source specifications, and radiation pulse width.
d. Bias conditions, output loading, and test circuit.
e. Description of the way in which state vectors for testing were selected.
f. State vectors used for radiation testing and functional test conditions for each state vector.
g. Criteria for transient output upset.
h. Records of the upset threshold and power supply current for each state vector.
i. Equipment list.
j. Results of the noise test.
k. Temperature (see 3.9).
4. SUMMARY
. The following details shall be specified.
a. Device type and quantity to be tested.
b. Test circuit to be used, including output loading impedance.
c. State vectors to be used in testing and device output pins to be monitored.
d. Functional test sequence.
e. Power supply voltage and bias conditions for all pins.
f. Pulse width of the radiation source (see 2.1).
g. The method of selecting steps between successive irradiation levels and the required resolution.
h. Restrictions on ionizing (total) dose if other than that specified in 3.1.
i. Temperature of the devices during testing.
j. Requirement for measuring and recording power supply peak transient current (see 3.11c).
k. Failure criteria for transient output voltage upset.
l. Failure criteria for power supply current and output current, if applicable.