| PDF
Test Flows for Space
PEMs
Harry
Shaw, Jeannette Plante, and Brandon G. Lee
Goddard Space Flight
Center/NASA
In the past
PEMs were not considered to be appropriate for space mission applications
because of their commercial connotation. However, the use of PEMs
has been gradually increasing and is expected to continue to increase
for several reasons. PEMs have been gaining attention as alternatives
to military standard parts, which do not keep up with technological
changes. Their advantages in size, weight, cost and performance
also encourage their use. Customers demand "faster, cheaper,
better and more reliable products" in order to meet their needs
for state-of-the-art technology.
The risks associated
with PEMs are evolving over time as the technology improves and
established manufacturers learn how to control their processes for
high yield. Manufacturers with less experience may produce parts
with well-understood failure mechanisms: moisture induced failures
and ion driven internal corrosion. PEMs are almost always sold as
commercial-off-the-shelf (COTS) components, so small volume users
cannot dictate screening testing or that the manufacturer perform
reliability testing. It is extremely difficult, if not economically
impossible, for small volume users to procure single packaging lot
date codes and to have any traceability to wafer lot (and certainly
not single wafer lot date code). These drawbacks point to the greatest
underlying concerns associated with PEMs:
- Are the
parts reliable on some level?
- How do we
establish what the Level is?
- How do we
upscreen to achieve a level appropriate for use in NASA spacecraft?
- How do we
establish accelerated test conditions without complete knowledge
of the encapsulation materials?
- Lastly,
where QML PEMs are available, should NASA treat them in the same
manner as QML hermetic, full temperature range devices?
An examination
of PEMs manufacturer’s websites indicates that there is a wide variation
in the reliability validation done by them. Some do "one-time"
qualification testing while others do regular life, moisture and
burn-in testing.
Goddard Space
Flight Center (GSFC) projects are using PEMs and an analysis has
been done on three manufacturer’s PEMs as part of the PEMs assessment
activity. The following process was used:
- Understand
the testing that space users are applying for upscreening.
- Understand
the testing the manufacturers are doing including, the frequency
of the test and sample criteria.
- Determine
a test flow based on the particular requirements of the program.
- Eliminate
tests done regularly by the manufacturer on the candidate parts
or parts sufficiently similar to the project candidates.
- Establish
a test flow for the project and address particular electrical
performance parameters that need to be screened or monitored in
qualification testing.
The three manufacturers
being considered for a GSFC instrument are Maxim, On Semiconductor,
and Analog Devices.
PEMs screening
and qualification tests have been developed and recommended by those
involved in high reliability space applications including NASA/GSFC,
the Jet Propulsion Laboratory (JPL) and the Johns Hopkins Applied
Physics Laboratory (APL). Though they have similar goals for achieving
space level reliability their approaches to screening and qualification
testing are slightly different and vary based on different spaceflight
missions. Those approaches have been reviewed to establish the most
suitable testing flow for the GSFC project. Some of these flows
are still drafts and unpublished so they are shown for illustration
only.
The testing
flows consider mission life and in some cases mission environmental
conditions. They generally consider tests that fall into three categories:
moisture resistance, operating characteristics over the entire temperature
range and device life under high temperature. Table 1 and 2 show
the comparison of the upscreening flows found. The following are
the major tests used in the upscreening flows and by the manufacturers
of PEMs:
· Visual
Inspection:
Visual inspection
is conducted in the beginning and end of both screening and qualification
to examine any defects present such as broken wire bonds, cracked
die, misaligned leads, etc.
· Burn-In:
Used to remove
infant mortals from the lot, a 10% PDA is generally applied.
· Acoustic
Microscopic Imaging (AMI):
Acoustic Micro-Imaging
(AMI) is a popular Non-Destructive Analysis (NDA) method to identify
material boundaries and voids that can induce moisture absorption
and CTE mismatch stress.
· Radiographic
Examination (X-ray):
Radiographic
Examination, or X-ray shows the part’s internal defects such as
broken wires and lead connections, cracked leads and extraneous
material. Real-time X-ray is often used to obtain high-resolution
images in various planes by rotating the device inside the chamber.
Recent findings indicate that X-ray may be destructive because it
deposits ionizing doses into the semiconductor, so care must be
taken when using this tool.
· Temperature
Cycling:
Temperature
Cycling is conducted to determine the resistance of devices to alternate
exposure at mission temperature extremes. Physical damage caused
by thermal expansion and contraction during the testing can lead
to defects including cracking and delamination of packages and internal
structures. It can also lead to changes in the electrical characteristics
resulting from mechanical damage.
· Life
Testing:
Life testing
is performed under biased conditions to identify latent failure
mechanisms and to quantify part reliability. It can detect design,
metal integrity, silicon contamination, manufacturing, and assembly-related
defects. High Temperature Operating Life (HTOL) is a life test.
· DPA:
Along with
X-ray inspection, the purpose of Destructive Physical Analysis is
used to gain knowledge of device construction technology. DPA involves
destructive cross-sectioning and chemical etching to inspect internal
features such as wire bonds and die attach.
· Moisture
Testing:
Moisture testing
drives moisture into the package that can lead to moisture-induced
defects such delaminations and cracks. Pressure Pot, 85˚C/85
RH, and HAST tests are moisture tests. Highly Accelerated Stress
and Temperature (HAST) testing is quickly replacing 85/85 testing.
It serves the same basic function as 85/85 in typically 10% of the
situations, making HAST tests useful for immediate feedback and
corrective action at the production level. A biased HAST test can
provide a Burn-in condition while doing moisture testing.
· Electrostatic
Discharge (ESD):
ESD Testing
is used to establish how susceptible the parts are to ESD damage.
Mil-Std-883 Method 3015 is used. The purpose of this test is to
measure the relative ability of the part to withstand accidental
electrostatic discharge. Both the Human Body Model and the Machine
Model are used in testing. Devices are divided into groups and tested
at specified increments across the ESD test range.
· Radiation
Hardness Assurance (RHA):
Just as military
products must be assured for operation in the ionizing radiation
environment of space, commercial PEM parts do too. A collective
knowledge about the affect of the plastic material on radiation
performance has not been achieved, but investigations are starting
and continuing. Radiation tolerance is believed to be a function
of the silicon’s design and materials and less a packaging issue.
Maxim’s, On
Semiconductor’s and Analog Devices’ test flows were reviewed. Maxim
performs electrical testing, burn-in and post burn-in electrical
test on a weekly basis; while temperature cycling, high temperature
with bias and HAST testing with bias are conducted quarterly. On
Semiconductor conducts qualification testing of interest to NASA,
but not the screening required. They are performing life testing,
Temperature Cycling and HAST yearly. Analog Devices performs very
little long-term bias or temperature testing.
Maxim selects
test samples on a package and semiconductor technology basis. ON
samples on this basis as well. Analog Devices samples in this way
and also by location. Research is still being done to find out if
Maxim and ON do testing by location as well. Table 3 shows the results
of the research done to understand the flows and methods used by
these manufacturers.
Further research
and study should be continued to gain more knowledge about PEM manufacturer
flows and data. Also to better understand how user test flows can
be reduced without negative impact to the reliability of the parts
used by the projects.
Table
1. Space PEMs Screening Tests: Five References Compared
|
Projects
Screen
Test
|
Ref
1
|
Ref
2
|
Ref
3
|
Ref
4
|
Ref
5
|
|
<
1yr
|
1-5yrs
|
10-15
yrs
|
|
Pre-Encapsulation
Visual Inspection
|
No
|
MIL-STD-883
TM 2010 CONDITION B
|
No
|
No
|
No
|
No
|
Yes,
the nearest applicable standard.
|
|
Initial
Electrical Measurement
|
No
|
FUNCTIONAL
OR DYNAMIC DC ANDAC TESTS 25C PER DEVICE SPECIFICATION
|
Mil-Std-883
TM 5005
|
+25C;
-55C
|
-55C;
-15C;+55C;+70C
|
Yes,
Not specified
|
Yes,
mission temp profile extreme.
|
|
Life
Test
|
No
|
No
|
No
|
No
|
2000hrs
at +70 C
|
Not
specified.
|
No
|
|
AMI
|
In
accordance with IPC JEDEC J-STD-020
Level
1 and Level 2:
100%
Level
3:
MIL-STD-1916, VL IV, Normal Attribute Sampling |
MIL-STD-883
TM2030
|
YES
|
Top,
bottom, thru
|
Top,
bottom, thru
|
Top,
bottom, thru
|
No
|
|
Interim(pre-burn-in)
electrical parameter
|
In
accordance with device specification 4/
|
No
|
No
|
No
|
No
|
No
|
No
|
|
Post
AMI
Electrical
Measurement
|
No
|
NO
|
No
|
+25C;
-55C
|
-55C;-15C;+55C;+70C
|
Yes,
Not specified
|
No
|
|
Serialization
|
Yes
|
No
|
No
|
No
|
No
|
No
|
No
|
|
High
Temperature Storage
|
No
|
MIL-STD883
TM1001.8, JEDEC 22-A103
|
No
|
No
|
No
|
No
|
No
|
|
Burn-In
|
Static
Burn-in
Level
1:
·
MIL-STD-883, TM 1015, Condition A or B,
-48
hrs minimum for Note 1
-240
hrs minimum for Note 2
Level
2 and Level 3:
·
MIL-STD-883, TM 1015, Condition A or B
- 48
hrs minimum for Note 1
- 160
hrs minimum for Note 2
Dynamic
Burn-In
Level
1:
·
MIL-STD-883, TM 1015, Condition D, 240 hrs for Note
3
Level
2 and Level 3
·
MIL-STD-883, TM 1015, Condition D, 160 hrs for Note
4
|
MIL-STD-883
TM1015.7 CONDTION D, Dynamic Burn-In 160HRS @ 125 °
C
|
72/160hrs.
Max
|
Dynamic,
5V for 72hrs at +55C
|
Dynamic,
5V for 72hrs at +55C
|
Dynamic,
5V for 72hrs at +55C
|
No
|
|
Post
Burn-In Electrical Measurements
|
Final
electrical Test
In
accordance with device specification, except parameters
specified in 311-INST-001 shall be tested as a minimum.
a.
Static tests@ 25° C, Max, Min rated operating temp.
b.
Dynamic and functional tests@ 25° C, Max, min operating
temp.
c.
Switching test @ 25° C, Max, Min operating temperature.
|
DC
PARAMETRIC FUNCTIONAL AND AC PARAMETRIC TESTS AT AMBIENT
TEMPERATURE
OF –55,25 AND 125.
|
Mil-Std-883
TM 5005
|
+25C;
-55C
|
-55C;-15C;+55C;+70C
|
Yes,
Not specified
|
No
|
|
Temperature
Cycling
|
MIL-STD-883
TM1010 Condition C(or the manufacture’s specified storage
condition, whichever is less), 20 cycles minimum
|
Mil-Std-883
TM 1010 Cond B 40 CYCLES AT 125C
|
Mil-883
TM 1010
10cycle
|
-60C
to +25(10cy)
|
-60
to +60C(10cy)
|
Yes,
Not Specified
|
No
|
|
External
Visual
|
In
accordance with GSFC PEM Guidelines, paragraph 7.1
|
MIL-STD-883
TM2009,JEDEC-STD-22 METHOD B100
|
Yes
|
No
|
No
|
No
|
No
|
|
Radiography
(X-ray)
|
No
|
MIL-STD-883
2012
|
No
|
No
|
MI-STD-883
TM 2012
|
MI-STD-883
TM 2012
|
Yes,
MIL-STD-883. TM 2012
|
|
DPA
(sample)
|
No
|
No
|
yes
|
No
|
No
|
No
|
No
|
|
Calculating
PDA(<10%)
|
Level
1:
5%
Leve2
and Level 3
10%
|
No
|
yes
|
No
|
No
|
No
|
|
|
Packing
and Shipping
|
IPC/
JEDEC J-STD-033 5/
|
No
|
No
|
No
|
No
|
No
|
No
|
|
Note:
1/:
Applicable to Digital (CMOS, PMOS, NMOS, BiMOS); Linear
(MOS Line Driver/Receiver); Linear (MOS Analog Multiplexer/Switch);
Mixed Signal, MOS A/D and D/A Converters, and similar technologies.
2/:
Applicable to Linear (Voltage Regulator, Voltage References);
Linear (Pulse with modulators); Linear (Timers), and similar
technologies.
3/:
Applicable to Digital ( TTL, ECL, DTL); Linear(Op-Amp, Instrument
Amp, Sample/Hold, Comparators); Linear( Active filters);
Mixed Signal, Non-MOS A/D and D/A Converters; Linear Non-MOS
Line Driver/Receiver; Linear Non-MOS Analog Multiplexer/Switch,
and similar technologies.
4/:
Optional test point. If delta parameter are not required.
5/:
48 hours @ 125°C, or the manufacturer maximum storage temperature,
whichever is less. Reference IPC/ JEDEC J-STD-033 for further
guidance. Vapor barrier bags (MIL-B-8 1075, Type I), packed
with approved dust-free desiccant (MIL-D-3464, Type II),
and moisture indicator card(MIL-I-8835, MS51015) shall be
used.
|
Table
2. Space PEMs Qualification Tests: Five References Compared
|
Projects
Qual
Test Condition
|
Ref
1
|
Ref
2
|
Ref
3
|
Ref
4
|
Ref
5
|
|
<
1yr
|
1-5yrs
|
10-15
yrs
|
|
Pre-Conditioning
|
Moisture
intake
JESD
22-A113, Paragraph 3.1.5.
(168
hrs, +85%RH)
Reflow
simulation with flux, clean, and dry
JESD
22-A113, Paragraphs 3.1.6, 3.1.7, 3.1.8, and 3.1.9(Use vapor
phase profile +219 ° C, no preheat)
|
No
|
(5
pieces max.) SMD to be machine soldered
|
No
|
No
|
No
|
No
|
|
Visual
Inspection
|
|
No
|
yes
|
No
|
No
|
No
|
No
|
|
Life
Test ( or HTOL)
|
MIL-STD-883
TM1005 Condition D
1000
hours at max operating junction temperature
|
Mil-Std-883,JQA108
|
No
|
JEDEC-JESD22-A108(5V
for 72 hour at +55)
|
JEDEC-JESD22-A108(200hrs
at +70C)
|
JEDEC-JESD22-A108(not
specified)
|
JEDEC-JESD22-A108,
following by mission temp extreme and ambient
|
|
Electrical
Testing
|
Per
device specification
|
No
|
No
|
+25C,-55C(after
Life Test)
|
-55C;-15C;+55C;70C(after
Life Test)
|
Yes.
Not specified(after Life Test)
|
Following
by electrical test at mission temp extremes and ambient
|
|
Temp
Cycle
|
No
|
883
TM1010 Cond C, JEDEC 22-A-104
|
20
cycles
|
No
|
No
|
883
TM 1010, 1000cyc, -50C to +150
|
JESD-22-A
104. Following by electrical test at mission temp extremes
and ambient
|
|
HAST
with Bias
|
JESD
22-A110, with continuous bias(250hours, +130° C, 85%RH)
|
JEDEC
22-A-110
|
No
|
No
|
No
|
JESD22-A110(96hrs
@ +130/85%RH)
|
No
|
|
Temp
cycle, humidity and bias
|
No
|
JEDEC
22-A-100A
|
No
|
No
|
No
|
No
|
No
|
|
Power
and Temp cycling
|
No
|
JEDEC
22-A-1O5A
|
No
|
No
|
No
|
Power
cycle only
(1000cycle
at max rated power)
|
No
|
|
Thermal
Shock
|
No
|
883
1011, JEDEC 22-A-106
|
No
|
No
|
No
|
No
|
No
|
|
Solder
heat Resistance
|
No
|
JEDEC
22-B-106A
|
No
|
No
|
No
|
No
|
No
|
|
Solvent
Resistance
|
|
883
TM2015
|
No
|
No
|
No
|
No
|
No
|
|
ESD
Sensitivity Test
|
No
|
883TM3015,
JQA2
|
No
|
No
|
No
|
EIA/JESD22-A114-A
|
No
|
|
Radiation
Hardness Test(RHA)
|
No
|
883
TM1019, ASTM F1192-88
|
No
|
No
|
No
|
No
|
MIL-STD-883,
TM 5005. Group E or equivalent
|
|
Solderability
|
|
883-TM2003&2022
|
No
|
No
|
No
|
No
|
No
|
|
Lead
Integrity
|
|
883
TM2004
|
No
|
No
|
No
|
No
|
|
|
85C/85%RH
|
No
|
|
yes
|
No
|
No
|
No
|
JESD-22-A101.
Following by electrical test at mission temp extremes and
ambient
|
|
High
Temp Storage
|
No
|
883
TM 1008, JQA 103
|
No
|
No
|
No
|
No
|
No
|
|
Bake
out
|
No
|
|
16
hrs
|
No
|
No
|
No
|
No
|
|
Salt
Atmosphere
|
No
|
JEDEC
22-A 107A
|
No
|
No
|
No
|
No
|
No
|
|
Outgassing
|
No
|
ASTM
E595
|
No
|
ASTM
E 595-93
|
No
|
ASTM
E 595-93
|
No
|
|
Flammability
|
No
|
UL-94-V
OR V1
|
No
|
No
|
No
|
|
No
|
|
Traceability
(Date
C.)
|
No
|
No
|
No
|
No
|
No
|
Wafer
Lot preferred
|
No
|
|
Traceability
(QML)
|
No
|
No
|
No
|
No
|
No
|
QML
Vendor Preferred
|
No
|
|
Current
Density Cal
|
No
|
No
|
No
|
No
|
No
|
Worst
Case
|
No
|
|
Data
Retention
|
No
|
No
|
No
|
No
|
No
|
1000
hrs @ +150C
|
No
|
|
SAM
|
No
|
No
|
yes
|
No
|
No
|
No
|
No
|
|
DPA
|
SEM(883TM2018);
Bond Pull(883TM2011); Decap Internal Visual (883,2010);
Die shear (883 TM2019)
|
No
|
yes
|
No
|
No
|
No
|
MIL-STD-1580
|
|
Reflow
Phase
|
No
|
No
|
220C,
1 pass
|
No
|
No
|
No
|
No
|
|
Internal
Water Vapor Content
|
883TM1018
|
No
|
No
|
No
|
No
|
No
|
No
|
|
Final
Visual Examination
|
No
|
No
|
yes
|
No
|
No
|
No
|
No
|
Table
3. Tests Performed by the Manufacturers
|
MAXIM
Commercial
|
MAXIM
Military
|
ON
Semiconductor
(all
testing is reported in a yearly summary by test hours)
|
Analog
Devices
(Performed
quarterly. Number of lots tested is based on percent of
total production)
|
|
Life
Testing - quarterly
|
Rapid-Response
Reliability Monitors
Operating
Life Test
Weekly
|
Autoclave
+ Moisture Level
Preconditioning
|
Autoclave
+ Moisture Level
|
|
Burn
In - weekly
|
Pressure
Pot - Weekly
|
Bond
Pull Strength
|
Temperature
Cycle
|
|
Pressure
Pot Test – one time qual
|
X-Ray
- Daily
|
Bond
Shear
|
Thermal
Shock
|
|
HAST
or 85/85 Test – quarterly
|
Solderability
- Monthly
|
Destructive
Physical Analysis
|
Highly
Accelerated Stress Test (HAST)
|
|
Temperature
Cycling Test - Quarterly
|
Mark
Permanency
Monthly
|
Electrostatic
Discharge
|
High
Temperature Storage Test (called life test by mfr)
|
|
High-Temperature
Storage Life Test - Quarterly
|
Solder
Thickness
Monthly
|
High
Temperature Bake
|
|
|
E.S.D.
and Latch-Up Testing – Qualification Only
|
Open
Short Test
Quarterly
|
High
Temperature Operating Life
|
|
| |
Long-Term
Reliability Monitor Program
(Quarterly)
Operating
Life Test
(Op
Life) 1000 Hours
|
HAST
with Bias
|
|
| |
Biased
Moisture Life
Test
(85/85) 1000 Hours
OR
Highly
Accelerated
Stress
Test (HAST)
100
Hours
|
Moisture
Level Preconditioning
|
|
|
|
Temperature
Cycle
1000
Cycles
|
Solderability
|
|
| |
High-Temperature
Storage
1000
Hours
|
Temperature
Cycling + Moisture Level Preconditioning
|
|
|
|
Autoclave
(Pressure
Pot
w/o Bias) (PPT)
168
hours
|
|
|
(back
to the top)
|
Test Flows for Space PEMs