iiiihiiiiihililiililiiihiiiiiiiiihiiii · 2014-10-01 · telescope mount charger battery regulator...
TRANSCRIPT
IIIIHIIIIIHIlilIIlilIIIHIIIIIIIIIHIIIIN7510468 I_ I_ ma'5wiMn.
INVESTIGATIONOF TANTALUM WET SLUGCAPACITOR FAILURES IN THE APOLLO TELESCOPEMOUNT CHARGER BATTERY REGULATOR MODULES
INTERNATIONAL BUSINESS MACHINES CORP.,
HUNTSVILLE, ALA. ELECTRONICS SYSTEMSCENTER
31 JAN 1973
U.S. DEPARTMENT OF COMMERCENational Technical Information Service
https://ntrs.nasa.gov/search.jsp?R=19750002396 2020-03-22T16:53:10+00:00Z
Tailored to Your Needs!
SelectedResearch InMicroficheSRIM ® is a tailored information service that
delivers complete microfiche copies of
government publications based on your
needs, automatically, within a few weeks of
announcement by NTIS.
SRIM ® Saves You Time, Money, and Space!
Automatically, every two weeks, your SRIM ® profile
received by NTIS and the publications microfiched
is run against all new publications
for your order• Instead of paying
approximately $1 5-30 for each publication, you pay only $2.50 for the microfiche version.
Corporate and special libraries love the. space-saving convenience of microfiche.
NTIS offers two options for SRIM ® selection criteria:
Standard SRIM®-Choose from among 350 pre-chosen subject topics•
Custom SRIM®-For a one-time additional fee, an NTIS analyst can help you develop a
keyword strategy to design your Custom SRIM ® requirements. Custom SRIM ® allows your
SRIM ® selection to be based upon specific subject keywords, not just broad subject topics.
Call an NTIS subject specialist at (703) 605-6655 to help you create a profile that will retrieve
only those technical reports of interest to you.
SRIM ® requires an NTIS Deposit Account. The NTIS employee you speak to will help you set
up this account if you don't already have one.
For additional information, call the NTIS Subscriptions Department at 1-800-363-2068 or
(703) 605-6060. Or visit the NTIS Web site at http://www.ntis.gov and select SRIM ® from
the pull-down menu.
• __ _'_._ I U.S. DEPARTMENT OF COMMERCE
_J ',_' _;_ I Technology Administration
• _ • JNational Technical Information Service
s,,.,f,.,,,v,e,,_s o* I http:l/ww_'.ntis.gov
INVESTIGATION of TANTALUM WET SLUGCAPACITOR FAILURES in the APOLLOTELESCOPE MOUNT CHARGER BATTERYREGULATOR MODULES
(NASA-C_-120q97) INVESTIGATION OF
TANTALUM UET SLUG CAPACITOR FAILURES IN
THE APOLLO TELESCOPE MOUBT CU&RGER
BATTE_¥ REGULATOR (International BusiLess
Machines Corp.} 112 p HC $5.25 CSCL 09A
N75-10461
Unclas
G3/38 52213
i
J
INVESTIGATION of TANTALUM WET SLUGCAPACITOR FAILURES in the APOLLOTELESCOPE MOUNT CHARGER BATTERYREGULATOR MODULES
JANUARY 31, 1973 PREPARED UNDER
CON T R ACT N AS8-20899
SKYLAB PROGRAM OFFICE
DI RECTIVE 0370
IBM NO. 73W-00050
PREPARED BY
TECHNICAL APPROVAL
SKYLAB PROGRAM OFFICE APPROVAL
J. F. WILLIAMS
ELECTRICAL SYSTEMS DESIGN
D. H. WIEDEMANRELIABILITY and PARTS
ENGINEERING
V. H. JOHNSON, MANAGER
E LECTR ICAL SYST EMS DESIGN
A. P. DAVIS, MANAGERRELIABILITY and PARTS
ENGINEERING
M. R. ROBINSON, MANAGER
SKYLAB PROGRAM OFFICE
_'__ Federal Systems Division, Electronics Systems Center/Huntsville, Alabama
ACKNOWLEDGEMENTS
Appreciation is expressed to all members of the MSFC CBRM team for their assis-
tance in providing much background information and technical expertise as a base
for the activities discussed in this report.
Personnel other than CBRM team members also provided assistance in various
areas of the CBRM capacitor investigation. Special thanks are due to them as
representatives of the foilo_ing organizations:
Boeing Company
General Electric Company
Federal Electric Corporation
Martin Marietta Corporation
McDonnell Douglas Corporation
NASA-Kennedy Space Center
NASA-Marshall Space Flight Center
Sperry Rand Corporation
TABLE OF CONTENTS
Section
1.0 INTRODUCTION ...............................................
I. 1 CBRM FAILURE BACKGROUND ..........................
I. 2 CAPACITOR DESCRIPTION ...............................
2.0 SUMMARY .................................................. .
2.1 CBRM LEVEL .......... . ...............................
2.2 CAPACITOR LEVEL ..... . ...............................
3.0 CONCLUSIONS ................ ................................
4.0 RECOMMENDATIONS .........................................
4 I CBRM-RELATED• _ o oo ql o eQ apmlm4, o apo e e o 6 6o o o oe oo ooo @o@@oo
4.2 GENERAL ..............................................
5.0 CAPACITOR CONSTRUCTION AND OPERATING PRINCIPLES ......
5.1 INTERNAL CAPACITOR CONSTRUCTION .................
5. 2 CAPACITOR OPERATION ...............................
6.0 INVESTIGATION AND TEST DETAILS ...........................
6.1 FAILURE HISTORY REVIEW .............................
6.1. i Summary ......................................
6 1 2 Conclusions• • • se eooee ee ee • ee • • e eeqpeee oe@ ee oeemeo
6 2 PROCEDURE REVIEW• eeeeeom le • aoeegooeo e ee eeoeeeeoeeeeee
6 2 1 CBRM Acceptance Tests• • oe oeoo oloqpoode_e o Io_oooo
6.2.2 Power Up/Down Sequence ..................... ....
Page
1
1
1
6
6
6
7
9
9
9
10
10
10
15
15
15
15
17
17
17
p zCZD G pAOZSUm'
Section
6.3
6.4
6,5
6.7 '
6.8
6.9
6.10
6.11
6.12
Table of Contents (Concluded)
SNEAK CIRCUIT ANALYSIS ..............................
6.3.1 Summary ......................:.:i...i:i: .i.i...il. :i:..:ii:.i
6.3.3 Details ........................................
COMPUTER-AIDED TRANSIENT ANALYSIS ...............
OVER-VOLTAGE TESTS ................................
6.5.1 Summary .......................................
6, 5. 2 Conclusions ....................................
6.5.3 Details ............ ............................
REVERSE BIAS TESTING/EVALUATION ..................
6.6.1 Summary .......................................
6.6.2 Conclusions ....................................
6.6.3 Details ........................................
CURRENT RIPPLE TEST ...............................
RINGING TESTS ........................................
6.8.1 Summary ......................................
6.8.2 Conclusions ....................................
6.8.3 Details ........................................
MSFC CBRM LABORATORY TESTS .......................
KSC TESTS ............................................
PROTOTYPE CBRM TESTS .............................
6.11.1 IBM Laboratory Tests ...........................6.11.2 ATM Breadboard Tests .........................
CHARACTERIZATION OF CAPACITORS REMOVED
FROM FLIGHT CBRMs ................................
6.12.1 Summary .......................................
6.12.2 Conclusions .....................................
6.12.3 Details ........................................
Page
19
19
19
19
29
45/46
45/46
45/46
45/46
49
49
49
49
71
73
73
73
73
77/78
79/80
81
81
81
105
105
105
105
vi
LIST OF ILLUSTRATIONS
Figure
1, 1-1
1.2-1
1.2-2
2.2-1
5.2-1
6.1.1-1
6.3.3-I
6.3.3-2
6.3.3-3
6.4-I
6.4-2
6.4-3
6.4-4
6.4-5
6.4-6
Schematic of CBRM Input and Output Filters ...............
Photo 1.2-1. CBRM capacitor showing internal assemblyand element construction ...................
Electrical Configuration of Capacitor Prior to Mid-1971 .....
Electrical Configuration of Capacitor After Mid-1971 .......
Capacitor Anode Silver Plating Criteria ....................
Photo 5.2-1 ...........................................
Photo 5.2-2 .... ........................................Photo 5 2-3
• om'o" ,g "eoooeiBoeeil, J,ooaeoeooGoeoomll e_l .
Autonetics Reverse Bias Test Data .......................
Flight CBRM Failure/Test History .........................
Simplified CBRM/ESE Schematic (Sheets 1 through 6) .......
Schematic of CBRM Trickle Charger Circuit ...............
Schematic of CBR M Input and Output Filters Showing
Decoupling Capacitors .................................
Equivalent Circuit for Electrical Conditions ImmediatelyFollowing ESE Turn-Off ................................
Plot of Output Capacitor Voltage vs Time ..................
Plot of L1 Current vs Time ..............................
Plot of lnput Capacitor Voltage vs Time ...................
Plot of L1 Voltage vs Time ..............................
Equivalent Circuit for the Electrical Conditions During
Trickle Charger Turn-On ...............................
Page
2
3
5
5
8
12
12
12
13
16
21
27
28
30
31
32
33
34
35
vii
Figure
6.4-7
6.4-8
6.4-9
6.4-10
6.4-11
6.4-12
6.4-13
6.4-14
6.4-15
• 6.6.3.3.2-1
6.6.3.3.2-2
6.6.3.3.2-3
6.6.3.4-1
6.6.3.5-1
6.6.3.5-2
6.6.3.5-3
6.6.3.6-1
6.6.3.6-2
6.6.3.6-3
6.6.3.6-4
List of Illustrations(Continued)
Page
Plot of Input Capacitor Voltage vs Time ..................... 36
Plot of Output Capacitor Voltage vs Time ................... 37
Plot of L2 Current vs Time ................................ 38
Plot of RC Current vs Time ............................... 39
Plot of L1 Current vs Time .............................. 40
Equivalent Circuit for Electrical Conditions During Trickle
Charger Turn-Off ....................................... 41
Plot of Input Capacitor Voltage vs Time .................... 42
Plot of Output Capacitor Voltage vs Time ................... 43
Plot of L2 Current vs Time ............................... 44
Sample No. 1 Data for Duty Cycle Test No. 2 (Sheet 1 and 2)... 55
Sample No. 2 Data for Duty Cycle Test No. 2 ............... 57
Sample No. 3 Data for Duty Cycle Test No. 2 (Sheet I and 2)... 58
Reverse Bias Diode Effect ................................ 61
Normal Capacitor AC Waveforms .......................... 62
Test Schematic for AC Test on Capacitors .................. 62
AC Voltage and Current Waveforms Showing Reverse
Leakage Current .........................................
AC on DC Test Summary .................................
Raw Data Sample AC on DC Test ..........................
Detail Plots of AC on DC Test Data ............... :: .......
AC on DC Test Schematic
64
65
66
67
................................. 69/70
viii
Figure
6.7-1
6.8.3-1
6.11.1-1
6.11.1-2
6.11.1-3
6.11.2.3-i
List of Illustrations (Continued)
AC Ripple Test Configuration .............................
Schematic of Lab Model of CBRM Input Filter ................
Photo 6 8 3-1• • eoe ma _$eemee oeee Jee oe $o leeleoe $eooooeeeo eol
Photo 6 8 3-2• _ •eloea BOB••CO0040 e_o• •O•eooOoOeoeomaOm_le • J
IBM Laboratory Test Configuration for PrototypeCBRM Tests ...........................................
Photo 6, 1 I. I-I
Photo 6. II, 1-2
Photo 6.1 I. i-3
Photo 6.1 i. 1-4
Photo 6.11. I-5
Photo 6.11.1-6
Photo 6, 11.1-7
Photo 6.11.1-8
Photo 6.11.1-90
Photo 6, 11.1-10
Photo 6.11.1-11
Photo 6.11.1-12
Photo 6.11.1-13
Photo 6.11.1-14
Photo 6.11.1-15
Photo 6.11. I-] 6
Photo 6.11.1-17
Photo 6, 11.1-18
Photo 6.11.1-19
Photo 6, 11.1-20
CBRM Po_ver Line Transient Test Schematic ...............
Trickle Charger Transient Test Schematic .................
Schematic of ATM Breadboard Test Setup ..................
Photo 6. II. 2.3-1 ........................................
Photo 6.11.2.3-2 ........................................
Photo 6.11.2.3-3 ........................................
Photo 6.11.2.3-4 ........................................
Photo 6.11.2.3-5 ........................................
Page
72
74
75/76
82
83
83
83
84
84
84
85
85
85
86
86
86
87
87
87
88
88
89
89
89
9O
91
94
97
97
97
98
98
ix
Figure
6. 12.1-1
List of Illustrations (Concluded)
Photo 6.11.2.3-6
Photo 6.11.2.3-7
Photo 6.11.2.3-8
Photo 6.11.2.3-9
Photo 6.11.2.3-10
Photo 6. II. 2.3-11
Photo 6.11.2.3-12
Photo 6. ii. 2.3-13
Photo 6.11.2.3-14
Photo 6.11.2.3-15
Photo 6.11.2.3-16
Photo 6.11.2.3-17
Photo 6.11.2.3-18
Characterization of FLight CBRM Output Fitter Capacitors ..
Page
98
99
99
99
100
100
100
101
101
102
102
103
103
106
X
LIST OF TABLES
Table
6.6.3.1-1
6.6.3.2-1
6.6.3.3.1-1
6.11.2.1-1
6.11.2.3-1
6.11.2.3-2
Page
Reverse Voltage Evaluation of Capacitor Assemblies ........ 49
Steady State Reverse Bias Voltage Evaluation Data ........... 52
Reverse Bias Duty Cycle Test No. I Data .................. 54
CBRM Transient Data .................................... 93
Trickle Charger Transient ............................... 95
Transient at SAS Power On and Off ....................... 96
x|
INVESTIGATION OF TANTALUM WET SLUGCAPACITOR FAILURES IN THE APOLLOTELESCOPE MOUNT CHARGER BATTERYREGULATOR MODULES
1.0 INTRODUCTION
1.1 CBRM FAILURE BACKGROUND
The Apollo Telescope Mount (ATM) contains eighteen Charger Battery Regulator
Modules (CBRMs). Each CBRM has an input filter and an output filter containing
tantalum wet slug capacitors. (See Figure 1.1-1). During ATM system tests at
MSFC, MSC and KSC, seven CBRM's experienced capacitor failures. One
failure occurred in the input side of the input filter, and seven in the output side of
the input filter. No failures have been experienced in the output filter.
Early failures were thought to have happened because of age and/or abuse since the
failed capacitors _ere 1967 date coded. The decision was made to replace all 1967
date coded capacitors with 1972 date coded capacitors. Prior to the retrofit cycle, a
failure of one of the 1972 date coded capacitors occurred. Subsequently, a team _as
organized to investigate the capacitor failures and recommend a solution to the
problem. IBM was asked to participate in the investigation. This report describes
IBM_s activity in association with other contractors and NASA toward a solution
of the problem of CBRM capacitor failures.
The purpose of the investigation described in this report was to determine the
mechanism of the capacitor failures and to identify the cause of the failure
mechanism.
Tantalum wet slug capacitors are historically susceptible to degradation via reverse
bias. Caution is ah_ays in order to prevent the application of voltage to the capacitor
in the reverse direction (reverse bias). A concentrated effort, therefore, was persued
during this investigation to locate a source of reverse bias as a possible cause of
capacitor failure.
1.2 CAPACITOR DESCRIPTION
The capacitors failing in the CBRM were wet tantalum electrolytic devices manu-
factured by the General Electric Company. Each capacitor is, in reality, an assembly
containing sixteen individual capacitor elements potted in a case. (See Photograph
1.2-I).
Prior to mid-1971, the assembly contained six'teen capacitors _ired in parallel.
Each capacitor element is rated at 27 microfarads, 100 volts dc. Capacitor
(3IJJZ
[I •
III
I
t -IL_
00 a. uu_
n
-)!
) .-._.-.__
,/
7--,
P
"I-
_m0
o
W
aD
I( °I("i,_°If uI_ _.
if.uIV
r_
I( _I( u
- j
d
, 0
o _o _. ,-
-Iu..
0
tI
'-I
I]
Figure h 1-h Schematic of CBRM Input and Output Filters.
CAPACITOR
" %'_'g I?5"C
J. YL'_ 85"G
DE "POTTED '
I!
,)
!_!_ i "ELEMENT!_+, -7i i*:
' SLUG
CASE SPACER
'i
................•
Photo I.2-1. CBIRM capacitor showing internal assembly and element construction.
REPRODUchli _1'1 OF l'/ia_
ORIGINAL PAGE 18 POOR
assemblies manufactured after mid-1971 have an internal series/parallel arrange-
ment consisting of two series groups of eight capacitors in parallel. These capacitor
elements are rated at 110 microfarads, 50 volts de. (See Figures 1.2-1 and 1.2-2).
Details of capacitor construction and operation are given in Section 5.0.
Data relative to the capacitor assembly are as noted:
Military Part No. SCL55CN441MP3
General Electric Part No. 69Fl164
Ratings: 440 microfarads
100 volts de @ 85°C
70 volts de @ 125°C
116 volts de surge @' 85°C
81 volts dc surge @ 1250C
4.8 amperes (ripple) 60 Hz @ 25oc *
22 volts (ripple) 60 Hz @ 25°C *
22 micro amperes max leakage @ 100 volts de
• Correction factors required for increased temperatures and/or frequencies.
The following terminology is used throughout this report when discussing the CBRM
capacitors:
• capacitor element - One of sixteen capacitors contained in each CBRM
capacitor.
• capacitor or capacitor assembly - Capacitor as installed in the CBRhI.
• reverse bias - Reverse polarity voltage, transient or steady state,
applied to CBRM capacitors.
+
rIIII!IiL_u_
CAPACITOR ,,'_
"-I
I _.
i %%IIII
__ _ ...J__--CAPACITOR E LEMENT
• Figure 1.2-1. Electrical Configuration of Capacitor Prior to Mid-1971.
F =w I •
IIiIII
I __CAPACITOR ,'j
-"'1
!IIiiII
L I
APACITOR ELEMENT
Figure 1.2-2. Electrical Configuration of Capacitor After Mid-1971.
2.0 SUMMARY
2,1 CBRM LEVEL
The CBRM capacitor investigation included a review of the history of the CBRMs
having capacitor failures in an attempt to establish trends or patterns of failures.
All failures were found to have occurred after the CBRM had been installed in a
system configuration and mated with Electrical Support Equipment (ESE). Also,
failures occurred only in the CBRM input filter.
A review of acceptance test procedures (ATPs) and power up/down sequences revealed
areas of potential transient problems. These areas of concern suggested further
analysis and tests which were pursued.
Both sneak circuit analysis and computer-aided transient analysis revealed sources
of reverse bias that could be applied to the capacitors under investigation.
CBRM tests at IBM, at the MSFC ATM Breadboard and at KSC verified sources of
reverse bias that could be applied to the input filter capacitors in the CBRM. Also
verified were procedural changes that eliminated reverse bias from the capacitors.
2.2 CAPACITOR LEVEL
Various tests were run on CBRM capacitors in an attempt to duplicate the failures
that occurred during ATM system test. The only method by _vhich the failures were
duplicated was the application of a reverse bias voltage to the capacitor followed by
the application of rated voltage in the normal forward direction. Swelling of
capacitor cases was observed under these conditions. The capacitors tested were
not confined; if confined, the cases probably would have exploded as did some of the
flight units that failed.
Subsequent tests were run on capacitors and capacitor elements in an attempt to
define failure characteristics with regard to over-voltage of correct polarity, mag-
nitude of steady state reverse voltage and current, magnitude and frequency of
pulsating reverse voltage and current ripple as experienced with a CBRM _ith the
capacitor operating at 90% rated voltage.
It was found that both over-voltage and reverse voltage are damaging to the
capacitors, and the damage is directly proportional to the magnitude of each.
Each capacitor, as _ell as each capacitor element, _as found to have.its own
characteristics, similar to others but sufficiently different as to prevent
predictability.
Normal CBRM current ripple (12 amperes peak) at normal CBRM operating voltage
(90 vdc) apparently caused neither physical damage nor degraded capacitor operations.
The anode of the capacitor elements is tantalum, and the cathode is silver. The
suspected cause of failure was plating of silver from the cathode onto the anode.
Figure 2.2-1 shows the criteria established for defining the relative quantity of silver
plated on the anode. Capacitors tested and opened were visually characterized for
degree of silver plating. It was impossible to correlate failure with quantity of
silver. Also, some off-the-shelf capacitors were opened and were found to have
slight silver plating. Hence, although failure analysis with regard to quantity of
plated silver was inconclusive, the silver plating with subsequent dielectric penetra-
tion is the apparent failure mechanism.
Replacement of all tantalum wet slug capacitors in .the CBRM with tantalum foil
capacitors was considered. Although less capacitance per unit volume is available
from tantalum foil capacitors, they have a greater tolerance to reverse bias. How-
ever, there were insufficient foil assets available to replace both input and output
filter capacitors.
It was decided to replace input filter capacitors with tantalum foil capacitors, since
all capacitor failures had occured in the input filter and the reverse bias possibility
was shown to be only on the input filter. Foils have a history of satisfactory operation
in similar applications; and it was determined that the CBRM would function properly
with the reduced capacitance in the input filter.
Although no failures had occured in the output filter capacitors, they were removed
from the CBRMs, and a teardown inspection was performed on them.
No anomalies were observed during teardown; so the decision was made to replace all
output filter capacitors with 1972 date coded tantalum _et slug capacitors. The series-
parallel arrangement of elements of these capacitors give a greater reliability to the
output filter.. If one element fails (short), the filter will continue to function since
the output filter operates at a voltage less than the rated value of a single element.
Also, no reverse bias was ever identified on the output filter capacitors. In addition,
foil capacitors would have changed the output impedance of the CBRM possibly causing
problems with other equipment.
3.0 CONCLUSIONS
The failure mechanism of the CBRM tantalum wet slug capacitors is a short through
the Ta20_ dielectric caused by either the penetration of silver or working voltagepuncture after dielectric degradation.
Both the penetration of silver migrating from the cathode and dielectric degradation
are caused by reverse bias voltage having been applied to the capacitors.
Figure 2.2.1.
8
"+__"t'+:+J"
--" _.+ ,--_+.
O_
Iii
_,:
.... -................ - ............ l _'
co _-.,+,_Capacitor Anode Silver Plating Criteria. _<3 ._
O_p-
ILl
.P..+,y_u-..?Aii._v-"-,-,-+ 1
+t,.., .j, ..+.-+,.-,'_ ".:, :@L_ "_._.;,'-'.'.,."; ,'PF'; -,+ 3
+
The cause of catastrophic failure in the flight CBRMs is reverse bias followed by the
application of greater than 50_ rated voltage. Reverse bias causes degradation or
penetration of the dielectric resulting in reduced working voltage capability. Applica-tion of forward voltage then causes dielectric breakdox_n. Catastrophic faiLure in the
CBRMs occurred because the forward voltage applied to the capacitors was supplied
from a high current source. Upon dielectric breakdown the capacitor presented a
short circuit to the voltage source and sufficient energy was available to cause explo-
sion of the capacitors.
Reverse bias on the CBRM input filter capacitors is possible under the following
c onditions:
Instant turn on/off of Solar Array Simulator (SAS) t5 volts to the CBRM
Instant turn on/off of CBRM battery trickle charger _hen SAS voltage
is at zero.
Reverse bias caused by the above actions can be prevented by eliminating the sudden
turn on/off of SAS 15 volts and by holding SAS 15 volts on while turning on/off the
trickle charger.
4.0 RECOMMENDATIONS
4.1 CBRM-RELATED
Modify the CBRM ground support procedures to require the follm_ ing:
• Ramp SAS voltage up from 0 volts and down to 0 volts.
• Hold SAS 15 volts at the CBRM while turning trickle charger on and off.
• Ramp trickle charger voltage up from 0 volts and down to 0 volts when
using portable trickle charger.
4.2 GENERAL
When using tantalum wet slug capacitors, insure that no reverse bias can be applied
to the capacitors. If the application involves interface with other components or
systems, e.g., ESE, test equipment, etc., insure that no sneak circuits exist where-
by a reverse bias condition can occur. Also, review procedures for possible se-
quences which could generate negative transients producing reverse b.ias on capacitors.
9
5.0 CAPACITOR CONSTRUCTION AND OPERATING PRINCIPLES
5.1 INTERNAL CAPACITOR CONSTRUCTION
The elements utilizedwithin the capacitor assembly are individual tantalum _et slug
capacitors. Each element has an anode made from powdered tantalum which is formed
in a cylindrical shape utilizingtemperature and pressure.
The capacitor dielectric is tantalum pentoxide, Ta20 _. The pentoxide is formed on
the anode or slug by subjecting the formed anode to an electrical potential (forming
voltage) in the presence of an oxidizing agent such as phosphoric acid. The thickness
of the pentoxide film is primarily dependent upon the potentialapplied and the
duration of the forming voltage. The finished film thickness contributes directly to
the unitVsvoltage rating and the capacitance value.
The electrolyteis sulphuric acid (H2SO4) = 40% concentration. A filler(CAB-O-
SILL @) has been added to the electrolyte producing a gel.
The case is coined silver, which has had the internal surface etched and coated with
platinum black. Normal electrical connection to the unitis plus to the anode or slug
and ground or negative to the case.
5.2 CAPACITOR OPERATION
The general theory of operation is simplified by considering the total unit capacitance,
i. e., tantalum anode one plate, the Ta205 film the dielectric and the case the cathode
or opposing plate and the electrolyte as a conductor or extension of the case/cathode.
When a positive voltage is applied to the anode, electrons leave the anode and travel
thru the external circuit to the case. The electrons enter the electrolyte _here
negative ions are formed. These negative ion s collect at the sulphuric acid side of
the pentoxide film as the film is an ionic barrier. Current (electron movement) in
the external circuit continues until the charge across the pentoxide film equals the
applied potential. Since the Ta205 dielectric contains minute flaws, the capacitor will
exhibit a forward leakage current (usually in micro-amperes or nano-amperes). Thetantalum pentoxide film has a unique characteristic that can be detrimental. This
characteristic is the fact that the Ta205 film is an ion barrier and not an electron
barrier. When the capacitor is reverse biased (ground or minus to the slug and plus
to the case), electrons leave the slug thru the film and are converted to negative ions
at the electrolyte. The reaction at the case is for silver to go into solution and in turn
plate out on the slug. Since the Ta205 flit _ allows electron flow in the reverse mode,
a reverse leakage current is developed which is magnitudes larger than normal forward
leakage. This reverse leakage current will continue as long as the reverse voltage is
applied. Catastrophic failure of the capacitor can be attributed to heat generated by
10
the reverse current and/or excessive liberation of gas from the electrolyte. Even if
catastrophic failure does not occur while in the reverse bias mode, the capacitor has
usually been degraded. Degradation occurs at the pentoxide film covering the tantalum
slug. This degradation is caused by silver plating out on the film and/or excessive
gas generation (bubbles} at the anode.
Since the pentoxide film is relatively weak physically, bubbles can destroy the film
integrity exposing the tantalum anode to the electrolyte. When the unit is again
forward biased there is no ionic barrier at the tantalum/electrolyte interface.
Depending on available voltage and current in the circuit, catastrophic failure can
occur due to high forward current creating heat and excessive gas generation.
Note: If the circuit is current limited, the tantalum has the potential capability
of healing by forming new tantalum pentoxide film at the tantalum/
electrolyte interface. The formation of new film over a fault area
appears to make the unit operational. However, there is no known test
data available that can attest to the reliability of a unit that has repairedor healed itself.
Silver deposited on the pentoxide film with time, can migrate thru the film (pentoxide
layer) creating a short from anode to cathode. A typical silver deposit is shown in
Photos 5.2-1, 5.2-2, and 5.2-3.
Another area of concern is silver deposition on flaw sites. It was previously mentioned
that the forward leakage current is due to fla_s in the pentoxide film. Under normal
forward conditions oxygen is present at these flaw sites. The normal reaction of
oxygen at these sites is the formation of pentoxide film. Once a site has been covered
by silver, new film cannot be formed. The net result is, as more sites are covered
with silver, the forward leakage current can increase. When the forward leakage
becomes large enough, catastrophic failure can occur primarily due to excessive
heat and gas generation.
During the investigation of the capacitor failures in the CBRM, a concentrated effort
was expended to determine the fragility level of typical capacitors under various
degrees of reverse bias conditions. (Reference paragraph 6.6}. During this
investigation, available data from other sources were also researched.
Figure 5.2-1 presents results of tests performed by North American Rockwell,
Autonetics Division. Tantalum wet slug capacitors x_ere subjected to reverse
voltages of 100 to 500 millivolts for 160 hours and then 50% of rated forward voltage
was applied. As expected, the failure rate increased _ith increasing reverse voltage.
It is concluded that the slope of the plot can be raised or lowered by varying reversebias application time and/or amplitude of the forward voltage applied after the reverse
time period has been completed. It is also suspected that the degree of current
11
Photo 5.2-1. Mag. 63X
• .1
#
-i
Photo 5.2-2. Mag. 350X
°
Photographs of tantalum slug showing
silver crystal deposits on the pentoxide
film-capacitor S/N 1098, Element No. 6
Date Code 1967.
Photo 5.2-3. Mag. 600X
Oa_G_aL- _rr oF rH_PAGE, IS POOR
12
©l=
\\
_ai,.u u.i.Jl-.
_wu,.I-doZZ
i 6
XXxlx
141
Dw_01.9
'" __._
X _0
\\\
_- z_-_D 0
_1o
\\
\
! I I ! 8 !
O O O O O O
S3_IR-IIV-i -IO 'ON
\\
\
_l=0\
\\
\
!
0¢'N
!
0
\\
-8
0143
0
o3
8 o_1
Z0 --
CO
I,M(/3
>u.It,r
0
0
Figure 5.2-1. Autonetics Reverse Bias Test Data.
13
limiting utilizedduring the applicatior,of forward voltage would be a contributing factor
to the failure rate.
The reverse bias testing contained in this report, the Autonetics test previously
mentioned and numerous other reports generally conclude that reverse bias on
tantalum wet slug capactiors can be detrimental and may lead to catastrophic
failure of the capacitor.
The primary differences between various evaluations of reverse bias are in the
following areas:
Amplitude of reverse voltage/current
@Steady state versus pulse
• Frequency (ac)
Duration of applied reverse bias
Amplitude of forward or normal working voltage applied afterreverse stress
Variations and combinations of the conditions noted above can produce varying failure
rates for given applications. However, for reliable operation it is concluded that noreverse bias should be tolerated.
14
6.0 INVESTIGATION AND TEST DETAILS
6.1 FAILURE HISTORY REVIEW
6.1.1 Summary
Records of CBRM capacitor failures were studied in conjunction with CBRM log books.
An attempt was made to correlate failures with age, location, number of turn on/off
cycles, rework, number of test cycles and type of test, i.e., component of system,
during which the failures occurred.
The review of all available records revealed that all known capacitor failures occurred
after the CBRMs had been mated with ESE in a pre-launch configuration. No other
trends or correlations were apparent. (See Figure 6.1.1-1.)
6.1.2 Conclusions
The cause of the CBRM input filter capacitor failures exists only in the pre-launch ESE
configuration and is probably the result of a sneak circuit present only in the ESE
configuration.
15
>.
zz
z_-:_ _-
I I I I I I a
WW_- 0
MJ
i.=,¢
U
eL
_..'-",,¢,¢
,_.
z
i=-q,.
_w
I-
I-,¢
oU_,_ _ _.._ (._
©
F--F-
I--
I
OUU_-UU
©
U
I-
F.,1{
P
I-,(
I-
O(3_)u. U
®>I-
¢J
P
I,-
o(JU_U
m_v
®L
I--I--
I-
Ira-.,i(
OUU_
o_v
©
I'-"l--Ct(
¢[
h-¢[
I--
o_U_ U
I-
>h-
©U
I-
oU oU
i-,(
U
I-
h-
I-_t
I-
I-cE
oU
m_
Dm,.
_U
(N
('4¢
P
(4¢
W_
o_
Figure 6.1.1.1. Flight CBRM Fai/ure/7"est H/story.
16
6.2 PROCEDUREREVIEW
6.2,1 CBRM Acceptance Tests
The MSFC ATP 40M26710A was reviewed in detail by IBM and MSFC personnel.
Instrumentation was used in an effort to determine if reverse bias was present during
an actual acceptance test. This test was conducted with the Flexible Automatic
Circuit Tester (FACT) and included functional tests as follows:
Voltage (Cells and Battery) ESE Trickle Charge
Charge Retention Battery Charger
Insulation Resistance Power Sharing
Continuity Command Signal
OutputVoltage Telemetry Status
Also, the MSFC ATP 40hL26705, utilized by Brown Engineering, was reviewed by
MSFC personnel.
It was concluded that the ATPs were adequate and that no tests or sequences resulted
in reverse bias being applied to the input or output filter capacitors.
6.2.2 Power Up/Down Sequence
The overall ATM power up/down sequence was reviewed with MSFC personnel at
Huntsville and KSC. The purpose of this review was to determine configuration
differences and possible reverse bias sources at the system level since previous
failures had occurred during system test.
The review revealed procedure and configuration differences for the trickle charger
power input to the CBRI_I at the component level, at the ATM Breadboard and at the
ATM System level at KSC.
The following differences were noted:
• Trickle charger connections to the CBRM do not involve the CBRM
input filter at the component level. At the system level different
connections are used at the CBRM whereby the CBRM input ftLter
circuit becomes part of the trickle charge circuit.
17
Procedures allow trickle charger power on/off with no regard to the
status of CBRM input voltage. In the system configuration, trickle
charger on/off _ith CBRM input voltage at zero will cause ringing in
the input filter and, consequently, possible reverse bias on the input
filter capacitors.
The method of turning trickle charger on/off differs between the ATM
Breadboard and KSC.
At the ATM Breadboard, the ac input power switch to the trickle
charger supply is used to turn trickle charger voltage on/off. A
ramping action occurs such that the CBRM input filter is not subjected
to an instant turn-on/off of current in the filter.
At KSC, a relay (dead face) is used to turn trickle charger voltage
on/off. An instant current turn-on/off occurs in the CBRM input
filter. Ringing will occur in the input filter possibly resulting in
reverse bias being applied to the input filter capacitors.
There is no transient suppression in the trickle charger circuit as is
present in uther ESE power circuits. Lack of suppression is a
potential contributor to a reverse bias problem.
The above findings identified areas of concern _'hich were investigated by means of
sneak circuit analysis, computer aided transient analysis and CBRM tests at theMSFC ATM Breadboard and at KSC.
18
6.3 SNEAK CIRCUIT ANALYSIS
6.3.1 Summary
Sneak circuit analysis was performed both on all CBRM internal circuits and
CBBM/ESE circuits. A sneak path was found whereby reverse bias could be applied
to the CBBM input filter capacitors upon trickle charger turn-on/off.
6.3.2 Conclusions
The only sneak path, whereby reverse bias can be applied to the tvet slug tantalum
capacitors in the CBRM° is in the CBRM/trickie charger circuit. Only the input
filter capacitors are involved with the trickle charger. No sneak paths were found
relative to the output filter capacitors.
6.3.3 Details
The following documents were used to perform an analysis of both internal CBRM
circuits and circuits connecting the CBRM with ESE:
40M26258 CBRM Pin. Function List
40M26259 Electrical System Schematic CBRM
40M26201 Wiring Diagram Electronics CBRM
40M33662 ATM Electrical Schematic
40M68374 ATM ESE LC-39 Advanced Electrical
System Schematic
40M70829 Power Interconnection Diagram ATM ESE
40M68507 Cable Interconnection Diagram ATM ESECheckout
40M35659-2B ICD Definition of ATM/AM Electrical
Interface
PRO-TC P-P-40004 ATM/Prototype Test and Checkout
Procedure ..
40M26995 A TM CBRM Engineering and Development
Report
19
1953805 (Bendix) ATM Control and Display ConsoleElectrical Schematic
61J760071 (MI)AC) Electrical Schematic Diagram ATM Carrier
Instruction Manual Rowan Power Supply
Model 120-20
Instruction Manual Kepco Power Supply
Model JQE55-2 (M)
An initial documentation check was performed for two reasons. The first was to look
for wiring errors that could be the source of capacitor failures. Secondly, it was
necessary to verify the accuracy and compatibility of the documentation to be used
during the failure investigation. The documentation check included a complete wire-
by-wire comparison of the CBRM electrical schematic against the CBRM wiring
diagram.
Figure 6.3.3-1 shows a simplified schematic generated from the above documentation
for use during the CBRM capacitor investigation.
Sneak circuit analysis was performed on the CBRM/ESE circuitry to determine
if a sneak path existed whereby reverse bias could be applied to wet slug tantalum
capacitors in the CBRM. Such a path _as found to exist in the trickle charge circuit.
When the trickle charger is turned on/off, reverse bias can be applied to the input
filter capacitors due to resonant ringing of input filter components. (See Figure
6.3.3-2. ) Subsequent tests verified that reverse bias transients can occur b hen
switching the trickle charger on and off.
Decoupling capacitors were suspected as providing a sneak path for reverse bias.
Sneak circuit analysis revealed such a sneak path (see Figure 6.3.3-3). However,
laboratory measurements proved that the reverse bias thus generated was too small
to cancel or reverse the normal static forward voltage present on the wet slug
tantalum capacitors.
2O
)z_
I
!
II
/I
I
_1 I I
'_VO_z l_N,V..a 9N_ UVlO$
P.]_EOD_CB_T_ O_ORIGINAL PAGE IS POOR
I
Figure 6.3.3.1. Simplified CBRM/EsE Schematic. (Sheet ! of 6J
I I..,-_ I_.t_ _
1
II
IIIII
,l
I_ _ i _ i_
n_ 'iz3 '6,
T l"_' =l
| ,_..aa,.! 'ii(;, °_ g#-_ l>
-_, _11 - - 4H_,-,HHI,-
!_-_l" "t_l ' t_'[
I-
Zi
v
$
Figure 6.3.3.1. Simp/ified CBRM/ESE Schematic. (Sheet 2 of 6)
22
I
p
!
LI1
e/,,j
t_ v "3
r" _ '9
t,,.._Jr',,ff'l
Figure 6.3.3-1. Simplified CBRM/ESE Schematic. (Sheet 3 of 6)
N
9.3
Figure 6.3.3.1. Simplified CBRM/ESE Schematic. (Sheet 4 of 6)
24
0
I
Figure 6.3.3-1. Simplified CBRM/ESE Schematic. (Sheet 5 of 6)
25
,,, =-I-
Figure 6.3.3-1. Simpfified CBRM/ESE Schematic. (Sheet 6 of 6)
26
01
5.
cO
o
_. _,} Zi,,
---I (-I,.0
I
=EC¢:n*t
'J w
-:1-p
0
I'"
REPRODUCIBILITY OF
ORIGINAL PAGE IS POOR
).. wrv"
Figure 6.3.3-2. Schematic of CBRM Trickle Charger Circuit.
IIJ
_"I_ I='"
2'/
rI
I
IIII
II
I=.,I _-
0
I_J
Figure 6.3.3-3. Schematic of CBRM Input and Output Filters Showing Decoupling Capacitors.
28
6.4 COMPUTER-AIDED TRANSIENT ANALYSIS
A computer-generated analysis of the CBRM input filter was performed under various
configurations to determine if any conditions exist in which a reverse bias is present
on the input capacitors.
Transient analysis was performed with the use of SCEPTRE (Systems for CircuitEvaluation and Prediction of Transient Radiation Effects), an IBM System/350 Com-
puter Program which automatically computes the transient response of large electrical
networks. The program automatically writes the set no differential equations describ-
ing the circuit configuration and subsequently solves the equations by numerical inte-
gration routines.
The equivalent circuit for the electrical conditions immediately following ESE turn-off
is shown in Figure 6.4-1. An initial worst case condition of 20.0 amperes was placed
on the two inductors, and the capacitors were assumed to be charged to 26:0 volts.
The load generator, ILOAD, was made voltage dependent such that after the output
voltage drops below 9.0 volts, the load current is removed. This simulates the turn-
off of the CBRM electronics after the input voltage drops below a preset level. The
electrical transient resulting fromthese initial conditions is shown in Figures 6.4-2
through 6.4-5. The voltage across the input capacitors can be seen in Figure 6.4-4
to reverse polarity to a maximum of 1.4 volts, approximately 4.7 milliseconds after
ESE turn-off.
The equivalent circuit for the electrical conditions during battery trickle charger turn-
on is shown in Figure 6.4-6. All initial conditions were assumed to be zero. Small
negative voltages were predicted on both the input and output capacitors in the input
filter during turn-on. The input capacitors were reverse biased to a maximum of
0. 022 volts and the output to a maximum of 0. 055 volts. The transient waveforms
are shown in Figures 6.4-7 through 6.4.11.
The equivalent circuit for the electrical conditions during battery trickle charger turn-
off is shown in Figure 6.4-12. An initial condition of 1.5 amperes was assumed
through inductor L2. Again reverse bias conditions were predicted across both the
input and output capacitors in the input filter during the transient. The input capaci-tors saw a maximum of 0.93 volts reverse bias, and the output capacitors experienced
a 0.19 volt reverse bias. The transient waveforms are shown in Figures 6.4-13
through 6.4-15.
29
zC_
I--Z
I,.I.• I..1.1
¢;3
I,,I,,I
he'
Z
I--
,.-I
=Em
I.MI.-I,.I.
t.a'3,...I
(Z:I.--ZC)¢.3
Z
I--
.=,10
I.--ZM.I
t.,OZ
nr-I--
Figure 6.4-1. Equivalent Circuit for Electrical Conditions Immediately Following ESE Turn-Off.
3O
REPRODUCIBILITY OF THE
ORIGINAL PAGE IS POOR
(St"fOAl Gr^
Figure 6.4-2. Plot of Output Capacitor Voltage vs Time.
31
ttttIll!Ill,It
• i i! i w,,i _1N l,
!
I'HI:1I" ,'i
I",1
IH
E
/" ,'1
/l
li3!
t
o
==
==
8
g==
=,0
Figure 6.4-3. Plot of L 1 Current vs Time.
4
4
=l
I
J -
i i ! _,,eI-
II
!
o Z,.0L !
I!I
_4
!II+, ,'4
,!
Figure 6.4-4.
,4,I
LII
t
ir
It
i
LSI-IO^) kd
Plot of Input Capacitor Voltage v$ Time.
!i "'_ :.
I!
!_""
33
ol
L !
!llJ
II
OO
,j..!!
I
• 4 46 4
I!
odo _g
OO
4m i_
D
D
II, g
_e N
)
)
PP)PJbJDD
DDb
bJDPDDD
||
4 J4
ID|
q4
D4 •44 •
B44 •q
J,
q
1
O41'
O,
,I
OI
Otl •
e'll
41 •
I!
w,4 _ ,U •
O
S±10^1 C1^
mw 41 _
tJ
• 41
O
• Q LI
4
t
I
Idm ao _o_ 4o
I"I
II
• •
D
O0
!
I
o.Ti
! oI !
III!!II!
I!
=* I
i
nd
CC
t
0
w00mn
qm
II
I
e_
c
•, N
M
;
Figure 6.4-5. Plot of L 1 Voltage vs Time.
34 ¸
II I! II II I| II
Figure 6.4-6. Equivalent Circuit for the Electrical Conditions During Trickle Charger Turn-On.
35
I
IIIII-II
II
I.Ii
IIII-.!
1I
!i-I
1I
II-,II!
II
I
i"fImm m,I
I!
!I
I
l-II
IIII •
I °II
I!!II
_ w,Q e,,q
¢!
f
t
!!
0
I
!
0
!I
• tl
I!
PC!
tl
i,ii J _1 m
I0
tl,
I
r,I
e
_m mareIII
I
r,!
O
II
I
_ o,e I-m
gP
80
e
(S.I.IOA
!t
,IF
I
!
It
ld
tl
eee. I
!
I
0,Ib
p,q !,_
I!
i •
:O
0
I"m• •
Ii
0
!
IIIlII
-IIIt!I!I
-I!I
i!II
.I
ii
II
-II!!
I
I|
iI
III!
iI
!• I
!
• II
IIII!
-IIII
II
1
O
II
¢¢
00
000
m
¢C
00
_J000
G¢
G
GG
0
_r
c
c
oo
u,
0
d
...J
Figure 6.4-7. Plot of Input Capacitor Voltage vs Time.
36
uJ
I'-
Figure 6.4.8. Plot of Output Capacitor Voltage vs Time.
37
=..* ...d =,4 ,.-t
=d
=,4
I
I ,,4
411
dl J=_
!!
!_-e N .-* ,,_
00
N
N
!!
!
III
G
8G
II
I!
,_1_, I.eaa
00
!.4
elul
S
O1al
S
g
Ie
"41
m ,I ! °O
e_
O I
Z
I
I. tl
! !
O OO O
_.1 ml
,[$31d.qdWV} _"_I
mlla iii I NI _
el
" 0
OO
I) •
I
!
• el
0O0
t_
!I
ee
i"
r,!
i •
I
.,I!
ItII
-t!
I
1
i!|1!
II
" II!
I!!
III
" II
!
|1III
I
I
0iI'
I!I
Cl
e.
¢
D
D
D
0
C_
Z
0CJUJ
..J
.J
UJ
Figure 6.4-9. Plot of L2 Current vs Time.
38
Figure 6. 4-10. Plot of RC Current vs Time.
39
_l==llm =1_ ,lla!!
I
i-II!
i-llil!.!!
i.g,
i!
1!
I
!!II,,i
IIJ!
I!
I"!!!!!
!
I!II!|
I..,I!!!!
J 0
O_I_Q_wqi..G° i,...t
C
!!
!I
dm
I!
41
!!
_7
i
i
4141,41
!I .
,!
I,IQ
I,t
uI.
|
0
0
tJ
O
LJO
t
b=
• '_iO 41
p
io
I'
ID
+ ....,:...
r_!
C
[S :I I:I'+IdlNV_, Ill
411,ii,
4.."'" ' I
i.i K lqo I
I i
,, I
I
.,,, ..,_.I,
,,+.-
I
I
iI
I
....IIItI
I1 °
II
1t ,,I!
gl01
.,II
lI "
gl|1g|
,,1l|||!
I
II
|1
II
ii|1II
!
!
Jo_.q
O!
11
0
ILl
J
W
Figure 6.4-11. Plot of L 1 Current vs Time.
40
I'- ¢x.
=-- ,= -,-
_, <_<_,.u__ •
_. r,_ _ r._ r,_ =,,.11 il II I!
g e
N
v_..-- z-- N,,, z z _ c:) z z _-
_:) -" i.u uJF- z :E z z x x z
C3 _ C3 c:3 ,,J _j r,jP _ P P z z _
_" _ Z z (:_ C_ I.u
Figure 6.4-12. Equivalent Circuit for Electrical Conditions During Trickle Charger Turn-Off.
41
REFRODUCIBILITY OF TIlE
ORIGINAL PAGE IS POOR
i
mq ._ .,q
g,4
i.d
N
p_
p_ m.J N o.o _
OO
,.d
Figure 6. 4-13.
Qt
I
Oo
o0
II
• •
I
O
a4 o,i 4,_ j
Q
• o,I •• If I
ID
r.
0O0
I
g
04141'41'11
D • I
tl gg
!
O0
Q
O00
00
"OOQIQII
QO00000
gO_
_00
OO00
OOO_
g_
|I0
,D •
tl, •
• 'O
• !
t,
• g
io.N°I
CSJ.'IOA) Icl
Plot of Input Capaci'tor Voltage vs Time.
42
I
I,
_,,=•
|!
!I
gt
0
O
• •1
01
I!
!
!!
!I_=¢.= .=i,._I
$
?
IIII
I _!
II!I
!
I
II
!
I!I
I!
,i
0
I
II
gI
0
r,
0,00
r,
0
Z
0
L}1.1,.I
..I
0 ul
_ q...
OO
O
O
f
),,,a
II!
IIlI *
4
,0
..I
i.i
J.i
n. *,_
o,.
ZlI
iuOOO
N
!I
!
• 4
im ip
g
Il
• , w,i
!
OOO
,.4
• 4
O_
I
II
III
• , •
I°
Io
• Ot
6
p •Ill
IIII •
I •IIl,II
II
• I
• I
e_ i,,,4
))
)g
) •i
IQ
41I
IpII
Ill
II
• 4
I
I) 4
4 4,1
• 441, 4
I.t 1,4
41 4
• 4
,ll •
4)_ O,
0
I,..4, w.0
0
0
ISJ.'i OA) _d
ii,i!
I I
, l
• 4
Ima, _
i
0
,?
I
1,4 i_.1
i,,,i q l.e,
,!!f,a _1
000
IN1
4_
II|!I!I
.II!
I!!!I
I
I
N I
_ 4
i ,
4)
M1
¢
b-
Figure 6.4-14. Plot of Output Capacitor Voltage vs Time.
43
,IO ;
,,.a
"i
I
0" I!
!
"I I,
Ii
i ,
i gt
, OIql
I I
I!"e
i $
I
q_,
i
.I..
...a
"T ....T"" T ""
i.
• 1_0 e! Q •
_ F ee
! , ! _,40 4J_ 0 0 g
: _ J*o..O.oo.,'o. o.o-.i i
, _ I n
000 t I! !" " "i" " " " "l" "
0 O
, Ta I
($3 I::3:Id&NV) _"I|
"T"
I
-q- _"
I i.
ii÷|
"'"T
IIII
-i
I
11I
-IIII
I
II
"I
.lI
4=3
I
8
D
i
r,
8N
0
0
Figure 6.4-15. Plot of L2 Current vs Time.
44
6.5 OVER-VOLTAGE TESTS
6.5.1 Summary
Over-voltage tests were run on capacitors to establish the over-voltage range required
to cause degradation or catastrophic capacitor failure.
6.5.2 Conclusions
The capacitor rating of 100 volts dc has an apparent margin of safety; i. e., the
lowest over-voltage failure occurred at 130 volts. Failure of the capacitors in the
CBRM is probably not caused by over-voltage.
6.5.3 Details
Three new 1967 date coded capacitors were subjected to various over-voltage conditionsas noted:
Sample No. 1 - DC voltage increased from 100 to 170 volts thru a 100 kohm
limiting resistor. Voltage increased in ten volt increments at 5 to 10 minutes per
increment. Forward leakage current was continuously monitored.
Leakage current was stable up to 140 volts. At 150 volts leakage current increased
from 13 to 30 micro amperes. As supply voltage was increased above 150 volts,
current increased to 180 micro amperes. Voltage applied to the capacitor did not
exceed 150 volts. The capacitor was unstable (voltage varying) at 150 volts and
failed (shorted) after 40 minutes at 150 volts.
Sample No. 2 - Repeated the voltage step stress performed on sample no. 1
except current limiting resistor was reduced from 100 kohms to 1 kohms. Currents
in the high voltage range, (150 volts and higher), were similar to those experienced
with the 100 kohm limiting resistor. After step stress _vas completed, the capacitor
was subjected to a hard turn on at 130 volts dc, current limited to five amperes.
The capacitor failed in a shorted-mode.
.Sample No. 3 - DC voltage was applied instantaneously and was removed when
the capacitor was fully charged. The capacitor was subjected tO dc voltages of 130,
140 and 150 volts current limited to 5 amperes. The capacitor failed at 150 volts in ashorted mode.
45/46
6.6 REVERSE BIAS TESTING/EVALUATION
6.6.1 Summary
Capacitors and capacitor elements were subjected to various levels of reverse voltages
and currents. The testing was primarily oriented toward short time reverse effects;
minutes and hours as opposed to days or months.
Degradation in the form of increased forward leakage currents was experienced after
exposure to 10 milliamperes reverse current for 21 hours. Catastrophic failure was
experienced by applying rated working voltage after exposure of 1 ampere reverse
current for a period of two minutes.
Other tests conducted indicate that the effects of reverse bias are cumulative, i.e.,
repeated half-hour exposures to the same amplitude of reverse voltage increased
both the forward and reverse leakage currents. On each successive application of
reverse voltage itwas also noted that the reverse current increases exponentially as
reverse voltage is increased linearly.
Low frequency ac testing was utilized to demonstrate the presence of an abnormal
current component that occurs during the negative half cycle of the voltage waveform.
6.6.2 Conclusions
Reverse voltage/current is detrimental to the reliable performance of tantalum wet
slug capacitors. Predictions as to capacitor failure rate with different levels and
durations of applied reverse voltage appear to be unreliable due to the many variations
of individual capacitor characteristics.
6.6.3 Details
Reverse bias tests were performed at two different part levels:
At the capacitor level as procured from General Electric: Internal
visual inspection of elements (16 per capacitor) after capacitor test
was limited to four elements per capacitor because of shortage of
time and availability of personnel.
At the element level after removal from new capacitors:' Element
testing and subsequent teardo_n inspection provided better correlation
between a given test condition and internal visual observations.
47
The basic categories of reverse bias conditions evaluated were:
• Reverse current amplitudes versus time. (Paragraph 6.6.3.1).
• Steady state reverse voltage. (Paragraph 6.6.3.2).
• Cumulative effects of reverse bias. (Paragraph 6.6.3.3).
• Reverse bias diode affect. (Paragraph 6.6.3.4).
• Application of ac without dc bias. (Paragraph 6.6.3.5).
• Application of ac superimposed on a dc reverse bias. (Paragraph
6.6.3.6).
6.6.3.1 Reverse Current Amplitudes vs Time
Results of tests performed on capacitors are given in Table 6.6.3.1-1. Capacitor
degradation and catastrophic failures were created by the following worst-case
conditions:
Degradation - 10 milliamperes reverse for 21 hours
Catastrophic Failures - 200 milliamperes reverse for 24 minutes
- 1 ampere reverse for 2 minutes
Catastrophic is defined as shorting in the reverse mode or shorting as full rated
voltage was applied in the forward direction after the capacitor was subjected to areverse bias environment.
6.6.3.2 Steady State Reverse Voltage
Four groups of capacitor elements were selected for this evaluation. Each group
contained 8 elements; four elements of each group rated @ 50 volts and four elements
of each group rated @ 100 volts. Each group of elements was subjected to a different
reverse voltage (bias) level ranging from 50 to 500 millivolts. Reverse application
time varied from 17 to 26 hours. Reverse current was monitored periodically, and
forward leakage current was checked randomly during the test period. Forward
leakage measurements were made at rated voltage utilizing a current limiting
resistor. At the conclusion of the test period the units were opened and subjected to
internal visual examination. Table 6.6.3.2-1 presents the results of this evaluation.Several observations can be made from these data:
48
Table 6.6.3. 1-I. Reverse Voltage Evaluation of Capacitor Assemblies. (Sheet 1 of 3)
0
°w.=
r_
U3
O
;>
,.Q
rJ_
;>
.._ (_
I I,,..= _
r./3 r.rJ _J
o
°"_
,o _ "_I _ I-4 _-,
N "_g'_
,-_
_ _-_
@O
I
o==E
@ "" 0
u_
'1:1
E[.-,
O3EC
O
=o"
L)
_.--- =="o -_ mc_
•
(D O .,_I:m _' m
EE
o
_D
o
0
_°°_._
!I
_'_m o ,._
_3 oo ,Q.
2-',_
°_°_.
. "+ r'_
++ .S_N
i
I
e _
_3
O
IA
___.,.. 130 I-, O
t_ _ m _ N
_-.-,
m.,.., m_-+°,=4
e_
E°_,,_
° ,,,,,i
E
E
o
49
Table 6.6.3. 1-1. Reverse Voltage Evaluation of Capacitor Assemblies. (Sheet 2 of 3)
0
G_
.,=,
co
03
Eo,.._
0
°,..d
== ¢,
°0
I
o
_ o
' _'r.
• Q
r_
°
r_ u3
I_0 .
0 C_ k,.c:
e_
E
e_
E
-=
I
r_
,_¢ 0
0
0
<|
E
c_
8=
I I..,
,_
o
..0
0
r-
_-o_;=.,,=, .,_ _ {_
0
4_=...i
I I,_
"6_ '_'_
_ •
0
0
o
E
tl
I
E
0
_ o
•,=. i"L
i'
1=,
E
't_ o o o (_
_ _ I I I I
0 0 0 0
5O
Table 6.6.3. 1-1. Reverse Voltage Evaluation of Capacitor Assemblies. (Sheet 3 of 3)
0.w4
u_
f./}
G)¢;
a)
°r,,4
I-i
=1
Q,
fJ""
¢,,1 o
...3
.._ ¢1
I--4
I l-i g
_ -"
0
I3
_E
*pC
u_
I
t-iQ
0
I
0
Q •
_._
_111 1
:>
¢,4
I.i.2°o_1o
0
g
"r.
Q
0
:> ,J:= _)
I
<..i
b_
,9®o 1.4> j:::
"0 u_
%
51
Table 6.6.3.2-1. Steady Statu Reverse BiDs Voltage Evaluation Data.
25O
500
NOTES
t. CYCLE IS THE NO. OF
TIMES FORWARD
LEAKAGE WAS CHECKED.
2. CURRENTS ARE PRESENTED
ASMAX ANDMIN RECORDED.
3. GLITCH CURRENT
SURGED!VOLTAGE UNSTABLE
AS FORWARD LEAKAGE
WAS CHECKED,
onn_]CIBIL,l'I_O_ '_
52
The degree of silvering found on the slug or anode is not directly
related with the reverse bias condition; i.e., five elements in the
250 millivolt reverse group exhibited light to gross silver; _xhereas
two elements in the 500 millivolt reverse group were fotmd to have
light silvering.
Reverse current levels are tmpredictable at the voltage levels used
for this test. Current in the 50 and 100 millivolt reverse groups hada tendency to decrease during the test whereas current in the 250 and
500 millivolt groups remained constant or increased in the same timeframe.
Temporary breakdown (glitches) and subsequent healing were noted on
several units in the 250 and 500 millivolt groups.
Note: If a glitch occurred, the voltage x_as decreased and then again
ramped to,yard the voltage rating.
6.6.3.3 Cumulative Effects of Reverse Bias
6.6.3.3.1 Duty Cycle Test No. 1 -
. T_enty-three capacitor elements were selected and exposed to file follm_ing testcycle:
100 millivolts reverse for one minute followed by 25, 50 or 100,_ of forward
rated voltage for one minute.
Groups of elements _ere removed from test at the end of 1, 5, 10 and 25 cycles. The
test samples _x:ere opened and subjected to internal visual examination. (See Table
6.6.3.3.1-1). No direct results were obtained from this evaluation; hoverer, it is
assessed that if a longer test time _ere selected, meaningful _'esults could have beenobtained.
Note: The intent of this particular evaluation was to demonstrate that the action
of plating of silver on the slug due to reverse current is greater than
the reaction of normal for_ard leakage current tending to move the silver
back to the case; i.e., when reverse and forx_ard currents are alternatelyequal in amplitude, degradation can occur.
6.6.3.3.2 Duty Cycle Test No. 2 -° .
Three capacitors were selected for this evahmtion. The first sample was subjectedto 1.0 volt reverse bias for thirty minutes; then forward leakage was checked at
rated voltage. This procedure was repeated u:_til failure occurred. The second and
53
To.hie b'.6.3.3. 1-1. Reverse Bias Duty Cycle Test No. 1 Data.
i MINUTE I MthIUTE
REVERSE FORWARD
31AS IMVI BIAS (V)
-tOO 50
25
5O
5O
5O
25
5O
5O
25
5a
25
loo
50
25
[ so
IOa
5O
25
-100 ,50
NUMBER
OF ELE. NO. CAP NO.
CYCLES
1 7 1
1 14 145
1 13 145..d
1 15 621
1 14 621, , L ,
5 11 1
5 3 14S
5 10 145
5 7 621
5 4 621
4 1 1
10 14 1
10 16 1
10 7 145
10 16 145
10 11 621
10 3 621
25 12 1
25 9 621
25 16 621
25 1 145
25 8 136
25 9 1
VOLT
RATING
50
50
50
100
100
50
50
5O
10O
100
50
50
50
50
50
tOO
100
50
100
100
50
50
50
CURRENT (AMP|
LEAK, REVERSE
.7Gx10-6 .28x10-5
.55,d0-6" .02x10-6
.24_10"6 05,10-6
.7x10 _ 92x10"G
.9x10 "6 .36x10-6
.55x10"6 01_10'6
.8x106 3xi06
.18xI0-6 .005_10-6
25>:106 1 x10"6
.£2w10 -6 14=106
"5x10"6 I 30=10"6t08KlO-6
"3×106 I 15 _106• i
.56_10 "6 .28_10 6
7_=iG'6 53,10"6
.E3,i0 6 .08_10-G
.G_,10"6 3Q_06
.06.10 6
35"10 .6 18=10.6
_2x10-6 005_106
.04xi0"6 .02x10-_
.09xi06
-21x10 6 12x10. 6
.50x106 12,10 -G
.66,10-6 5xI06
o7.i06•25_106 25x10 6
.4_10. 6 G5_i0. 6 I
28_10"6 18z10-6
1Ox%O 6 14x_0 6
22,.10 6 17,10 6"I
.4OxlO 6 .14=10.6
.C2_10"6 I 34_10G
.30x10'6 01x10 E
.7=10 "6 21=10 .6 t OK
;9x10"6 I01,106 [
/ 14,1o .6 i oK26_10"6
"'1 15x106
9x10"6 1.41_10. 6,61x,_,06
INTERNAL VISUAL
SLUG H2SO 4
Ag SLIGHT
BROWN BROWN
DEPOSIT
OK OK
OK OK
OK OK
A_ SLIG HT
OK OK
B ROWN
DEPOSIT
OK OK
OK OK
•
OK OK
OK OK
Ag SLIGHT4
Ag SLIGHT
OK OK
DENDRITE
OK
.... i
OK OK
CAS E
OK
OK
Ao
OK
OK
A9
OK
OK
OK
OK
Ag
Ag
Ag
Ag
OK
OK
OK
5._
A
<£=E
E3v
h-zLUet-rf"
LIJ03et"LU
UJet-
100
9O
8O
7O
60
5O
4O
3O
20
. THESE PLOTS REPRESENTBACK BIAS LEAKAGE CURRENT WITH-1.0 VOLT DC APPLIED
ALL PLOTS WERE MADE WITH THESAME CAPACITOR
ATTHE COMPLETION OF EACH30 MINUTE RUN THE FORWARD BIASLEAKAGE AT +100 VOLTS WASCHECKED AND MARKED AT THE ENDOF EACH PLOT
RUN NO. 8 WAS ALLOWED TO OPERATEFOR 2.0 HRS. THIS RUN IS SHOWN FORTHE FULL 2.0 HRS. ON SHEET 2 OF THISFIGURE
27 pa
RUN NO. 4
10
I i I 1 | |
0 5 10 15 20 25 30
TIME (MINUTES)
I
35
Figure 6.6.3.3.2-1. Sample No. 1 Data for Duty Cycle Test No. 2. (1 of 2)
55
100 *SEENOTE4 ON SHEET1 OF THIS FIGURE
I 1 1 1 1 I 1 I
0 15 30 45 60 75 90 105
TIME (MINUTES)
120
Figure 6.6.3.3,2-1. Sample No. 1 Data for Duty Cycle Test No. 2, (2 of 2)
5G
100
THISCAPACITORWASOPERATEDWITH-1.5 VDCREVERSEBIASFOR26MINUTESAT WHICHTIME ITDEVELOPEDAN INTERNALSHORT.
8O
A
<
C3v
F-ZLUr_
I11(/3et-LLJ>LtJr¢
6O
3_
0 _ I i
0 5 10 15
TIME (MINUTES)
! ! I
20 25 30
Figure 6.6.3.3.2-2. Sample No. 2 Data for Duty Cycle Test No. 2.
57
A
<
_J
t.-Z
rr-E)(.,)
ct)
>u.J
I00 -
90
80
7O
60
50
40
1. THESE PLOTS REPRESENTREVERSE BIAS LEAKAGECURRENT WITH -1.5 VDCAPPLIED.
2, ALL PLOTS WERE MADEWITH THE SAME CAPACITOR.
3, AT THE COMPLETION OFEACH 30 MINUTE RUN THEFORWARD BIAS LEAKAGE AT+100 VOLTS WAS CHECKED ANDMARKED AT THE END OF EACHPLOT.
°4. RUN NO. 7 WAS ALLOWED TOOPERATE FOR 2.0HRS. THISRUN IS SHOWN FOR THE FULL2.0 HRS, ON SHEET 2 OF THISFIGURE.
/t
3O
2O RUN NO.
RUN NO.
RUNNO.
RUN NO. 6
RUN NO. 2
l j . I t
0 5 10 15 20 25TIME (MINUTES)
I
35
Figure 6.6.3.3.2.3. Sample No. 3 Data for Duty Cycle Test No. 2. (1 of 2)
58
8OO
70O
6O0
500
<
r_v
zI.U
_: 400IT"
U
ILl
n,"'" 300>uJn'-
2OO
103
*SEE NOTE 4 ON SHEET 1OF THIS FIGURE
f | ! I I I 1
0 15 30 45 B0 75 90
TIME (MINUTES)
! |
105 120
Figure 6.6.3.3.2.3. Sample No. 3 Data for Duty Cycle Test No. 2. (2 of 2)
59
third samples _ere evaluated in the same mmmer except the reverse bias x_as increased
to I.5 volts. The results of this evaluation are shown in Figures 6.6.3.3.2-I thru
6.6.3.3.2-3.
The general observation is that the reverse bias current increased with time but
returned to zero at the beginning of each new cycle. Sample no. 1 showed a
progressively inerea " ",_sn,,, rate of reverse leakage \:ith each succeeding run as well
as a larger forward leaka;_e at the completion of each run. Sample no. 2 shorted
during tim first run. Sample no. 3 operated with very little degradation tmtil run
no. 7 at which time it al)proached a shorted condition.
6.6.3.4 Reverse Bias Diode Effect
Several elements were subjected to reverse bias voltage, increased in equal increments
until catastrophic failure e_eurred. At eachincrement of reverse voltage the reverse
current was measured an3 recorded. The results of this test are shown in Figure
6.6.3.4-I.
The primary observation made durh_g tMs evaluation is the non-linear rise or increase
of reverse current (diode effect) with linear increase of reverse voltage.
6.6.3.5 Application of AC Without a DC Bias
This test was primarily a,, experiment to verify the presence of an abnormal component
of reverse leakage current when ac is applied to a tantalmn wet slug capacitor. The
basic operation of an unpolarized capacitor _xhen ae is applied is as follows:
Figure 6.6.3.5-1 illustrates the normal voltage and current waveforms expected
_hen an ac voltage is applied to a capacitor.
In reviewing Figure 6.6.3.5-1 the normal waveforms are as expected; i.e. :
• The current leads the voltage by 90 °.
• Current is ma.,dmum _hen voltage is 0, or in terms of capacitor
charging oi_eration, current is maximum _hen changing voltage is
applied, e.g., at 0 ° and 180 °.
• Current is (t _hen voltage is maximum or again considering capacitor
charging action, current is 0 uhen the capacitor becomes fully
charged, e.g., at 90 ° and 270 °.
Considering the voltage _aveform, 0 ° to 180 ° represents the capacitor charge and
subsequent discharge utilizing o positive voltage. In like manner the voltage waveform
from 180 ° to 3_;0 ° represents the capacitor charge and subsequent discharge
6O
1,000,000 -
100,000 -
10,000 -
o3 1,000LUE:U..Ir_
<(5 loo.o0=(.,.)
:EUP_
10.00I--.ZLIJE:n,"
_j1.000
O3n-ILl>
0.100 -.
0.010 -
0.001
NO. 12
NO. 1
+
+
_10.1
ONO. 2
13 *NO. 2 24 MIN.*NO. 15 60 MIN.NO. 13 44 MIN.
*NO. 12 140 MIN.*NO. 1 138 MIN.
*CATASTROPHICFAILURE
INITIAL LEAKAGE0,30 - 0.34 #A
×
"--"--'-I"-"_I_T_----F_ _ .... t -- _r--_0.25 0.50 0,75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50
REVERSE BIAS (DC VOLTS)
Figure 6.6.3.4.1. Reverse Bias Diode Effect
61
I I I _ I
_./ i \E i iI
I I i _ ) /i
i CHARGE .I I' CHARG i
i- -= I- ' !DISCHARGE "1
II
i ',,\ ' /i9 180 70
I \ t / 1
) I I
_)ISCHA RG E-]j_I
!J' II
x I360
Ii
II
I
APPLIEDVOLTAGE
RESULTANTCURRENT
Figure 6.6.3.5-1. Normal Capacitor AC Waveforms.
8 EACH
440 /zf EACH
50 MI LLIOHMS2CHANNELBRUSHRECORDER
Figure G.G 3.5-2. Test Schematic for AC Test on Capacitors.
(;2
utilizing a negative voltage. In reviewing the theoretical current waveform, the
positive voltage charge/discharge current waveform (0 to 180 °) is symmetrical _ith
the negative voltage charge/discharge current waveform (180 to 360o).
For the first step of this experiment eight tantalum wet slug capacitors were con-
nected in parallel (total capacitance -_3500 mierofarads). Utilizing a _ignal genera-
tor, a 1 Hertz, four-volt peak-to-peak sine _:ave was applied to the capacitors from
a power amplifier. A meter shunt ( _50 milliohms) was placed in series with the
capacitors in order to measure current. The shunt leads, plus a set of voltagemonitoring leads connected directly across the capacitors, were connected to a t_vo-
channel brush recorder. (See Figure G. 6. 3. 5-2). Voltage and current _vaveforms
obtained on the recorder are shown in Figure 6.6.3.5-3.
In reviewi_g these waveforms, the current waveform is as expected during the 180 °
of positive voltage application; however, during the 180 o of negative veltage (capacitors
reverse biased) the current waveform is distorted such that the waveform is unsym-metrical.
It is theorized the following sequence of events takes place:
As the voltage goes negative from 0 to -1.45 volts (1/8 cycle), the current
waveform appears normal, i.e., current is maximum at the instant vcltage starts
negative and si_rts to decrease as voltage approaches maximum (normal capacitoraction).
As the voltage increases further negative, the current, which is noI-mally
decreasing at this time, begins to increase as a result of the capacitor beginning to
leak. (Reference Par._graph 6.6.3.4}. This totaI current l-eaks out a_,proximately15 ° before the voltage peaks out indicating a large resistive as well as a reactive
component of current at this instant. When the current returns to zero, it is leadingthe voltage by approximately 60 ° which still indicates the presence of a resistive or
leakage component but considerably less than previcusly. The voltage at this time is
-1.7 volts. The complete recovery to a normal 90 ° current lead over applied voltage
is not seen until the applied voltage returns to a positive value. It should also be
noted that power, hence heat, is developed within the capacitor during the leakage
portion of the cycle because of the resistive component. This is not the case at other
times since energy is alternately stored and returned to the circuit when a 90 ° phase
shift exists. Power dissipation, and consequently, heat are probably l esponsible for
the capacitor's expanding and sometimes exploding.
6.6.3.6 Application of AC Superimposed on a DC Reverss Bias
This test was an experiment to determine if the quantity of reverse bias leakage cur-
rel_t is in any was related to the frequency of the applied voltage. The results of this
test indicate that the leakage current is greater at lower frequencies. These resultsare summarized in Figure 6.6.3.6-1.
63
_lIltllll]ililtl!1i11!1liL_i_i!t_t
[i]liliii!ll]llLH
__:, ._L:_
m
IN3UUND
8 ==> 8
B
ill
>
,,, _
:7,. I--
Zmr" u_
>
_ w
I "@
I "I "@
@
<
I
>
> <
+
Figure 6.6.3.5-3. AC Voltage and Current Waveform._" Showing Reverse Leakage Current.
6,3
400
375
350
325
300
275
.-.. 250<_EC.)r_ 225
I.-ZuJrr 200rr
Ouu 175o')n.."u.,I>,,, 150tr
125
100
75
5O
25
Figure 6.6.3.6-1.
_ _ NEW CAPACITORS
PREVIOUSLY OVERSTRESSEDCAPACITORS
T • . •
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
REVERSE VOLTAGE (VOLTS)
1.9
AC on DC Test Summar/.
65
..... .'._ .-":Z_--...-.=_--............ :---_ _" " P--:- _" !-_-"_-_-"'_"_-'_............ q :
_=7""_--_ .............. _, ._........ _..-:_,. ---- O
_---_--_-:'-----'J .... _ , T ..=:-,-- (3,,. "_
1 _==_---_I--!1,_ _-........ ---'::"--_".... _"_" " Z-JJ(N
',.......... 2_:-_-- .... 4:- ..... (f_ r._ 00, !............... %--_-'_"--:.J .J _1 " , W r,_I _-:" ..... ;._.-. I-- I-- I,-.. _............. _------=:"£z-_--, __1 _ -- " _ C_ CC .I"-" ........ _:-_'.__"'_._ -- -- -- ! _--':': _j_ _-- _ .i ........... _:;_-='_' ' J J .-I. _................... ::::; ; --'__-- :_ ,2"_ ,F........ ._:_"-.. 0 0 0 F........ :--:7--::::=-:--'_ o o " _ r,- <::: •1-- ........... 2...... ::---> :> :> !............ : _;::_ 'O _ ,.4 " _ ,,- ,..,o co' ........... • t ......... +...... ":-T;;.r ..... __, __, . _ _ ,_ I..U............. -'_-_-_---'_:_.--.: " LP, O G', '- .... _;:-'- O O _ :
............. •-:.-'_"- ..... _' • • i'--: .......... -.-;:--_-.:: ..... LO .... _ _ (33 _};............. ::Z=_-7:-_-_._D"7","7" _ ...... I'_:-:-"':-": ' !I __ __,_'--L._-----',P ' - ............. : -.-'_ ........
........ --_ ........ , i [ ....... X_uJ_'DL_: __ u_
_w zw_:> 0
F - r-- r"" r" -_ ........... _ + F--'" "_"=':..... P" _ n"
.-+-0 0 o--_::-:!'-, ...... -: -p--_ _--_.-,_.._ _:,.---- ,--[- ..
F o> :..,:,,: . l.......... ,,--__-__:;:L1 rm - ÷..... < --";,::--_-.: :=-.:-"h ....L _:!" ÷ ±____=-.__:: _7 -- - i ...... 0--;_:__-_'_._--........ tq "
F--.........z_-_>.-___- • F.... , +--_;_-_-r-__--....:..... --_'--_;':_:-i--_" I"'- I.......-;_--::--:--"I_I <_
I=-:.......-='+q"' ........ l l-.......... _,.-_T--z:--J <_ ' r........_.......;'=__L:__.-..L..--::-I,- I --J
..... •._ ......+-.+:=-___' -:-• _ - _...... _.......-_'_-+-:-'---.:-.-Z" ' i _ Z_- .... 0 ......._;:-:;_'=-:1 "-_ , _..... -G-- ---_.--.___:.- -..... t •
[_-,.o ......._---::::_:_°" !_I ;:....._:._..--.-----::_i_-.-, i >":'LU I:I::.......... _'L':: :_ _ - :]- ......... _:..---.._::L ,_: .... i , rr
........................ I _ " "-T:-:"_--" w_, _ Z
i----......... :-::--_::; c_ • ,....... t"'--'--_: ÷;. :: 1 i ' <I............... "_ --.'C Z_:. t'_ t.... _ _ ....... ._'-_'_'--:-- "-.::;- "1 _ "I ' ............ -=_zu-:" w _ L_-_,-" .......=.___:=_ ; ... , , I-._l
L .................. -,.._-_t-_-__-=° > ,4. f---,-- ....... -,.-_'.-_-_-:!-._r_[--: ............. _"L.::----= w L.+' I.............. -.--_--'<::"=:1---,.";IX cO
r<-
I-............. --_=--:,.::i _ " I-............. ---"--_::-.-::-I ,-:1 L............. _::----= '" i- ....... :-, LL..J
; ' ""-_"_:::::--_ ' i _ i ......... "-_--:: -_- " ..... _ "
..... _.-.z-_:-'-_-I---_ . _-.- _:;--!" ,.,.-l:'--i :
I ............_-- -.....INI ............. -+-_-"'......._I..............==-::-i_-It-r1i .............E£-_..... I '• _ -+L- ., • i_; _ ...... ,_'_-+K_. • ..... 1
..... _.:-: .I'T 14 + ........... _+-:-:.......... + '
+:..... _:=_:: ::::] ....... ::::,;::: , ...........=---_:.+ '+-I ; .........._-:; ....
. _:;-_ .................... .; ..... _ !t........ -_=--=:::....... 1 l ....... ::_: ..... 'I.....__: :::.'_:!_ ::.:.:::..:_I_:-:!-::-+ :;-+_=;-_- - ......:i ...............?.,:<:.:_.... I- . "_'--% ' .........I ' ..............._-; " ' ' l !
N
T
0l.I
)-.
dZ
0t-"
U
Figure 6.6.3.6-2.
6(;
Raw Data S;;mple AC on DC Test.
g_:_I_0DUCIBILITY OFoRIGINAL pAGE IS POOR
400
375
350
325
300
275
250
{,'3Iii-- 225i11
2oo.,JJ
_ 175a
150
125
100
75
5O
25
0
1 10
I1
i
1.0 1.1 1.2 1.3 1.4
DC VOLTS
1.5 1.6 117
SEE FIG. 6.6.3.6-2
FOR FlAW DATA
(SAMPLE PLOT)
MA
125 MA
:--t
1;8
Figure 6.6.3.6-3. Detail Plots of AC on DC Test Data.
67
This test _vas performed on eight capacitors. Each capacitor _as operated at four
frc{luencies- 0.1 ttz, 0.5 Ilz, 1.0 Hz, and 10.0 Hz. A sample rtm is shox_n in
1,'i_u-c 6.6, 3.6-2. This sample shov, s test capacitor no. 1 being operated at 10.0
Ilz. Data from these rtms were used for plotting data points on Figure 6.6.3.6-3.
After all points _ere plo{tcd, diago,_:_l li_es commcti_g tim average leakage currents
at the g-iven frequencies and voltages v, ere drax_n showing the overall leakage effect
ai a given frequency.
Figure 6.6.3. C,-4 is a schematic of the laboratory setup for this test.
-J _E.w
Z c_Z-t-n-_'_0"r_C 3r,.3rrw
"t-O
..J
.J
o
o
<
]
.-I<
_w
Figure 6.6.3.6-4. AC on DC Test Schematic.
69/70
6.7 CURRENT RIPPLE TEST
A ripple test was performed on six tantalum capacitors to determine if CBRM ripple
current causes capacitor degradation, The ripple test exercised the capacitors at the
normal CBRM ratc _,hen maximum (90 vdc) voltage, maximum (12 amperes peak)
current is applied. Figure 6.7-1 describes the test configuration. The materials
investigation disclosed traces of silver on the anode of the capacitor fo[lox_in_ current
ripple testing. No degradation in capacitor operation w,_s noted.
71
L1 250 _ h
90
VDC IC11
O
Vc 1
_C1
440 p f
L2 80 p h
IL
•-_ 1.5 AMP DC
30 VDC
IC 1
VC 1
IL 2
Ic 2
VC 2
AMPLITUDE
+12 A-
0
+l v.-1.5V
12A-
0--
12A-
0
+2.3V
0---.35V
12 ps
I I
1 i
I II I
I
I1I I I
I i II I It 1, I
.
TIME
Figure 6.7-1. AC Ripple Test Configuration.
?2
6.8 RINGING TESTS
6.8.1 Summary
Tests were run on a lab model of the CBRM input filter to determine if ringing (resonant
oscillation) caused by instant application of input voltage, applies reverse bias to the
input filter capacitors.
Current in the filter oscillated around zero indicating normal capacitor charge/
discharge current. Voltage across the capacitors rang but did not go belo_ zero.
No reverse bias was observed as a result of circuit ringing.
6.8.2 Conclusions
Although ringing does occur in the CBRM input filter when input voltage is applied
instantly, no reverse bias is applied to the input filter capacitors as a result of the
single action of input filter circuit ringing. Any reverse bias is the result of filter
action combined with other circuit actions in a system co_ffig_ration.
6.8.3 Details
A lab model of the CB]_.M input filter was built for the ringing tests. ESE transient
suppressio:} compol_ents were included in the circuit, and a 205 kohm resistor simu-
lated the circuit load at power on. (See Figure 6.8.3-1).
Upon application of 15 volts dc to the circuit via a mercury switch ringing occurred.
Current was observed to r;ng positively and negatively arouad zero, i_dicating normal
capacitor charge/discharge current. Capacitor voltage rang while charging to 15 volts
but never went negative. Photos 6.8.3-1 and 6.8.3-2 show input filter circuit ringing.
73
ESE TRANSIENTSUPPRESSIONCIRCUIT
II 7.6 mh
1 2a __ _L._L
250 # h
ALL CAPACITORS: 440 #f, 100 VDC
INPUT
250 ,uh 50 m
_ ', F'Y_""_3_ }_ ,O
1CURRENT
WAVEFORM
TTiii-I °V O LTAG E
AVEFORM
205 K
/• , _w , ,0
Figure 6.8.3-I. Schematic of Lab Model of CBRM Input Filter
?4
r
_
5 ms/cm
t
i1
ii
,m
VOLTAGEt._ 5V/cm
i1
i
ii
I CURRENTt _ 0.4 A/cmI
0
Photo 6.8.3-1. Voltaic :lnd current _vaveforms in lab model inpqt filter output
capacitors at 15 votts On.
2 A/cm
!
L.
5 ms/cm
Photo 6.8.3-2. Current xxaveforn', in input filter output capacitors at l'5volts On.
II._t_ODt]CIBIL1T_" OF _'I_E_s/76
6.9 MSFC CBRM LABORATORY TESTS
Capacitors with 1967 date codes (parallel construction) were installed in a CBRM
input filter at the Astrionics CBRM laboratory for time durations ran_;2ng from 45
hours to 160 hours. The CBRM was operated at nominal load continuously except
for shut-downs to remove and install capacitors.
Capacitors with 1972 date codes (series-parallel construction) were installed in the
same CBRM in the output filter. The CBRM was operated at nominal load continuously
for 71.5 hours.
No capacitor failures occurred during these tests. All capacitors tested x_ere opened
for silver characterization. No patterns were defined for time, operating voltage,
current ripple and silver plating.
'77/7s
6.10 KSC TESTS
IBM personnel supported EMC tests on the AThI CBRMs at KSC in an attempt to
identify transients at the input and output of a CBR:,I. The tests were directed by MSFC
personnel and were prompted by concern about reverse bias on the CBR]_I tantalum wet
slug capacitors. Also, negative transients on CBRhI power lines had been observed
during ATM system test _t hlSC.
Negative transients were recorded on the input and output of CBRM no. 2 when trickle
charge voltage _as applied or removed and when the SAS power supplies _ere turned
on. It was not determined if the negative transients were applied to the CBI_M filter
capacitors. Subsequent tests were ran at the MSFC AThl Breadboard to further define
transient parameters and to determine if the CBRM filter capacitors are experiencing
any negative transients. Breadboard test results are given in Paragraph 6.11.2.
79/80
6.11 PROTOTYPE CBRM TESTS
6.11.1 IBM Laboratory Tests
Prototype CBRM No. 037 was obtained from I_ISFC Astrionics Laboratory to perform
transient evaluation tests. The IBM laboratory test configuration is shox_n in
Figure 6.11.1-1.
The test schematic for power line transient evaluation is shoun in Figure 6.11.1-2.
No negative transients v, ere identified. (See photographs 6.11.1-1 thru 6.11.1-9),
The test _chematie for trickle charger transient evaluation is shown in Figure 6.11.1-3.
A condition of transient reverse bias was found to e_st whenever the battery trickle
charger u as turned on or off. Worst-case transients x_ere 1 ampere and (-)500
millivolts. (See photographs 6.11.1-10 thru 6.11.1-20).
6.11.2 ATM Breadboard Tests
6.11.2.1 Summary
Earlier analysis and lab tests identified possible reverse bias sources and suggested
tests on a CBRM in a system configuration to verify such so:,rces.
Prototype CBRM No. 037 was instrumented to monitor voltage across and current
thru the CBIlM input filter capacitors and was installed in the ATM Breadboard at
h,ISFC for tests.
The tests x_ere performed in order to verify that the CBRM sees negative transients
ex-ternally and reverse bias (volU_ge) on the input capacitors internally during the
normal trickle charge switelzing mode. Tests were performed to demonstrate possible
switching transients at "SAS power on" at the !5-volt level.
Further tests were performed to verify that ramping the "SAS power on" from 0 voltand turni,l_ the trickle charger on and off with greater than 15 volts or, the systenl will
not produce negative transients and uill prevent reverse bias on the input capacitors.
Upon turning trickle charger on and off, negative transients were recorded at the CBI_M
inputs and at the CBRM input filter. Input transients at turn-on were as high as (-)2.3volts and at turn-off were as high as (-)1.8 volts. Input filter transients at turn-on
were as high as (-)0.95 volts and at turn-off _ere as high as (-)0.20 volts.
Upon turning SAS 15 volts on, negative transients were recorded at the CBRM input and
at the CBRM input filter. Input transients _ere as high (-)2.5 volts. Input filter
transients uere as high as (-)0.90 volts.
81
'._ _,;:ac¢. S_,'$1emqJ C_ntcr,
I.:-_l r,.,._" r,_ _..._ Huntsville, Alabl_ma
DATA SHEET Dote 11-27-72
No. 037
Po, t No.
Test ]Reverse Bias - Transient
SYSTEM
FI LTE R
REPRODUCIB/Lypy OF TI/Ig
ORIGINAL PAGE IS POOR
TI I
,i
0 - 65VPOWE R
SUPPLY
ii
LOAD 1
L _
ESI S"FG.,-"iI
AM-METER
0-28
POW E RAUX
TEST _
LJ
J1 :'J:,2
CBRM
UNOERTEST
_]_- -- - O 0
I o
Fiqure G. 11. I-1. IBM Laboratory Test Configuration for Prototype CBRM Tests.
82
L,-.__I_._ _" t._ IhJrd:=.ville, Alabarlla
DATA SHEET Do,o11-27-72
Pcr_ No. CBRM No. 037
Tes_Initial Condition Vo = -15 mv, \71 = 340 mv
0 _.
O--
4
t:; V I
.H
"i1
!1
iv-i
. . ..... 4]
_"....... ._. ........ ,7- '""
O_
ti
!i
i1
0 m
0 "_:'_._ " .... "- _ :.",.;-.'.;. .....
I I
iv_
!
I I
V I
V0
Photo 6.11, 1-1
ESE & Solar Array = 0 vdc
I o = 10 a/d
V I = 20 v/d
V o = 20 v/d
10 ms/d
System Turn On
Photo 6.11.1-2
ESE & Solar Array = 0 vde
I I = 10 a/dV I = 20 v/d
V o = 20 v/d10 ms/d
System Turn On
Photo 6.11.1-3
Turn On ESE @ 15 vdc
I I = 10 a/d
V I = 20 v/d
83
,'. r" ":"_ ;,"' ."'-= Fr:.:.,'al System':. Div;sion
: .; = ; S;.;_-e Sy._tern', Center,
,., _ =x y r,_ HL:'d-vme, A;abama
ATA EET Dote11-27-72
CBI_M No. 037Part No.
Te'.tESE Turn On
........... r-
0--
0 _ ....
| .....
0--
O_
0---_ ....
t
I:t
J,i
ii
{
i
II
v I
V o
II
t
, V I
iVo
t
1
Photo 6.11.1-4
ESE Turn On _ 25 vdc
I I = 10 a/d
VI = 20 v/d
V o = 20 v/d1 ms/d
Photo 6.11.1-5
ESE Tt, rn On @: 65 vdc
I I = 10 a/d
V I = 20 v/d
V o = 20 v/d5 ms/d
Photo 6. ii. 1-6
i I° ESE Turn On @ 65 vdc! Io = I0 a/di = 20 v/dV VIt I V o = 20 v/d
i 5 m s/d
I_V!
e ._ _"_._ Ir'_-, / : _:_detal Systems Davi,lon
(._ r..:_' I_ _'_ Hunlsvdle, ALabama
DATP, SHEET II-27-72Date
Sample
Ser;at No.
Po,t No, "
Test
CBRM No. 037
Reg. Turn On @ C_5 vdc
0_
0--
L
r
0--
0--
L
I
,q
!
I
i
V o
VI
V0
iv_
!
V o
Vi
Photo 6. II. I-7
V o = 1 v/d
V I = 5 v/d2 ms/d
Photo 6.11.1-8
V o = I0 v/d
V I = 50 v/d
2 ms/d
Photo 6.11.1-9
V = 1 v/dO
V I = 5 v/d
2.m s/d
85
_,,_,_,.;_ _.:. t 3 Fc_.ral Systems Division_ " , _ SpaceSys_emsCenter.
_ ¢,J '_- t_ Hun|svd!o, AI;Jbar_'l_:
DAT . SHEET Do,e 1-28-72
CBR]_i No. 037Po,t No..
Test Trickle Charger S',;Jtching Test
r
0_
0-
0 b
L .
|i I,
.'d
;I
ivl
l
tIJ
,i
!t
:V I.i
I I
v I
Photo 6.11.1-10
Trickle Charger Turn On
11 = I a/d
V I = 100 mv/d2 ms/d"
Photo 6. ] 1.1-11
Trickle Charger Turn Off�On�Off�On
II = 1 a/d
VI = 100 mv/d5 ms/d
Photo 6.11.1-12
Trickle Charger Turn Off
I I = 1 a/d
VI = 100 mv/d2 ms/d
$_
_, '_"_ f:'-. ' 2 Federat Sy'--t:..n,_: b_vi_i;on
! i . Space System_ Center,
!,_1 _-_ -_ _ i,.JI Hue, tsville, Aio!_*,_'n_a
Sf- Dote 11-29-72
Somple-
Serial No.
Pu,-t No. CBRMNo. 037
Test Trickle Charger Turn On
0-
i
0-
L .....
0._
L ......
0 m
t I
V I
I t
!
_-V I
i .II
1
v I
Photo 6, 11, 1-13
II =, 5 a/d
V I = 500 mv/d.5 ms/d
Photo 6. !1.1-14
I I = . 5 a/d
V I = 500 mv/d2 n,s/d
Photo 6, 11.1-15
I I = , 5 a/d
V I = 500 mv/d2 ms/d
• _
87
,,: v-;_j:_ ='-',, l"_ Frd,_fal Sy:tems Divir.ton
• , .! .,,,.1.eSystemsCen_er,
_i_ F,_ 1" |N tt,,*n:svdle. AlaDama
SHEET Oo,o 1,1-29-72
Pa, t No. CBRM No. 037
Test Triehle Charger Turn On
0-
0_
i
t
i
i}!
t
iVi
Pimto 6.11.1-16
I I =. 5 a/d
V 1 = 200mv/d
l ms/d
F ....
0n
0--
.....
i lo
tt
!V o
!
Photo 6.11.1-17
Io = .5 a/d
V o = 100my/d-
S_
I *- ' "_, ; TI Federa_ _y.,,_cm_ Division
,, '. c Space Sv,:ter_.sCenter,• ,i
t _,..... , :;,j _. _ Hunls_ith.., A_absma
DI: TA SHEET Dote ii-29-72
Port NoCBR_I No. 037
Test Trickle Charger Negative Transients Turn Off
...... ...,°.,
0 ''"
0
0
0 -
0
i Io4
,I
i*,2
!
2
_V oi:
4
il6
_V oi
• -_.. *_. .
iio!
':V oi
!1
Photo G. ii. 1-18
I o = 5 a/d
V o = I00 mv/d5 ms/d
Photo G. 1]. 1-19
I o = . 5 a/d
V o -- 100 mv/d2 ms/d
Photo G. ii. 1-20
Io =. 5 a/d
Vo = I00 mv/dI ms/d
89
z_O0 ,.- ,...J{3
i,'-',>0'3('_
0
1-<
I I
i ,,' II_- ',_ Ii_ i
L-T-% ..... _Jo_>-v _It;J2 g
ILl
Figure 6. 1 I. 1.2. CBRM Power Line Transient Test Schematic.
90
F
I
I
iIt I
I! i
IA"I
III J
'1!L____
--3
II
I
---IF--'
--_i1111_-
Figure 6. 11.1-3. Trickle Charger Transient Test Sch,_matic.
91
l:pon ramping SAS voltage up and down, no neo_ative transients _ere observed.
The trickle charger x_as turned on and off while holding SAS 15 volts on, and no negative
transients were observed.
Table 6. I1.2.1-1 gives a summary of transients recorded during the breadboard tests.
{;.11,2.2 Conclusions
:<cga_:ive transients are possible at the CBtlM input and at the CBI_M input filter during
normM ground support oper;ttion of the CBRM. The undesirable condition of reverse
bias on the input filter capacitors is the consequence of sudden l_r'_ on/off of either
SAS 15 volts or trickle charger _ith SAS input at 0 volt.
Elimination of negative input transients can be accomplished by ramping SAS voltage
up and down and turning trickle charger on and off only _xith SAS 15 volts applied.
6,1 i.2.3 Details
i:Io2ifications were made on the breadboard in order to simulate more closely the con-
diticns at KSC. A Kepeo power supply _:as installed for the trickle charge input at
the deadface relay. A toggle s_xitch was installed in the SAS pm_er supply line. A
breakout box was used bct',veen the CBII).I and the input cable. In order to a0ply input
pov, cr at J1/T (ESE) and a!/G (SAS) the input pin was sx_itehed in the cable between
toasts. The cable lengths at the breadboard _ere appro,x:imately (50 feet (18 meters)
for ESE power and appro-dmately 80 feet (2-t meters) for the trickle charger pm:er
s_lpply. Figure 6.1.1.2.3-1 describes the Breadboard test setup.
A peak recording voltmeter was used to monitor negative transienLs on the CBR?d input
lines at the SEE and at the CBRI_I during 1urn on and turn off (see Tables 6.11.2.3-1
and 6, 11.2.3-2).
A Tektronix 564 oscilloscope t_ith camera was set to record voltage and current at
the input capacitors (VI, I!) and output capacitors (Vo, Io) in the input filter of theCBIIM,
A Biomat_on 610B transient recorder xxas used to detect negative transients at the
input of the CBilM. The transients were displayed and photcgraphed on a Tektronix
517 oscilloscope.
A second Biomation 610B transient recorder was used to detect negative bias on the
input eapao, itors of the input filter. The transients were displayed and photographedon a Tektronix 547 oscilloscope.
rCransients photoi_raphed during these tests are shown in Photographs f,. 11.2.3-1|]ll'tt (i. 11.2.3-18.
92
Table 6.11.2.1-1. CBRM Transient Data.
A cti on
Trickle ChargerOn-ESE Pins G&P
Trickle Charger
On-ESE Pins G&P
Trickle ChargerOa-ESE Pins T&P
Trickle Charger
Off-ESE Pins T&P
SAS 15 volts
On-ESE Pins G&P
SAS 15 volts
On-ESE Pins T&P
CBRM TRANSIENT DATA
Negative Transients (Volts)
ESE Output* CBRM Input VI Vo
1.25 1. O0 O. 95 O. 18
1.80
i.60
2.30
Not Recorded
I00
0.24
2.30
i.80
2.20
2.50
0.20
0.90
0.14
0.90
0.80
0.07
0.12
0.04
0.17
0.16
*ESE data recorded 133"McDonnell Douglas (East) personnel from Peak Recording
Voltmeter (See Tables 6.11.2.3-2 and 6.11.2.3-3)
93
E
11:'I"-3
Fiq_lre 6. 11.2.3-1. Schemat/c of A TM Bre,_dboard Test Setup.
Table 6. 1
,-j
00
0
L_
0
.r.4
Zo
0
.J-4
E
o =oGJ
0
C_,--I
1.2.3-1.
0
"0
00
0kU
"0
o_
Trickle Charger Transient
L_
a
o m _o
0
L."_C_
0
0
u_O0 tC_
¢,I
,J,J _,JL_
_o _ _oo_.._
C
oO-O
Z
• e--4 •
O4u'_, L'_. L_- _.
0
95
Table 6. 11.2.3-2. Transient _ztSAS Powc,r Oi? r.nd Off1
"C
0uo#
o
oL_
q_
m
M
_ .
_ "do
¢:1
m
r
(
C
E
f
(
b _
,'-' ."2.,'-i __' _: b__ .,_ "d ,-
'd"
_b
b- ¢_.C_
I.'_ _ ,'_
,-4 L_.
C_
S
o
b_c"5
>_t_ _J3 f'_
-_ _ _
.q(;
_+:: i + . S!)_Ce Sy_lem_ C_+nter,L!i_l La,Z_4 _i ,' ..+., H_Jnls¥ille, Alebam+_
•2.+mET Date
Po,t l"Jo.CBRM No. 037
Test Trickle Charger On x_ith ESE or,. Pins G&P
_Jo>
p-
t_Z
t_a_(.3
+1
-1
!!+
i
i
--I
Photo 6.11.2.3-1
cq
.Jo:>v
_=.
>
A
_>
>-
:>
o_>
+1
-1
_500
0
-500
+100
0
-I00
J
)q
t
i
)
Photo 6.11.2.3-2
Photo 6.11.2.3-3
97
-- ' " 1 Fecr, l?._ f, y4!em| Division+
. : I T _ ¢.,eurll_.'.'_He, i, tab3ma
I_ l.-; I: .p_ _ E',__, Dote
Pa:t >Io. CBRM No. 037
Ic-:_Tz'Lekie Char_er Off with ESE on Pins G&P
cot-.Jo>
F-
fL
Z
:Err
O
+1
0
-1
]t
J
|
Photo 6.11.2.3-4
F--J
o>
>-
:>
>-
>
v
O
+I
0
-1
+200
0
-200
+100
0
-100
.1
J+
,I)1
+1
I
I}
......... 11
]+!
..,,_
:|
+
Photo 6.11.2.3-5
Pheto 6.11.2.3-6
9S
t " -=; _ F'1_--tcleral F..ystem_ Division
I_ _ ' 'i Sp,ce S_B1emsCenter,I_,',i L. _; L,,t _' l,_ Huntsville, Alabama
Date
Sample
Serlol No.
Pa_t No. CBRM No. 037
lestTrickle Charger On with ESEon Pins T&P
k-..}o>
I--
Q.
Zm
_E(c{I3L)
>
>-
>
o>
v
>-
+2.5
-2.5
+500
0
-500
+2OO
0
-200
+50O
0
-500
+1,7
0
-1.7
100 JJSkC/DIV
2 [I,,SEC.;DJV
I
T
,tF
tI
i
I
i
!
J4
_t
1
. _oJ
Photo 6.11.2.3-7
Photo 6.11.2.3-8
Photo 6.11.2.3-9
99
_uce SF_tcm= C,'n_eF,
,,_ b _ _j Huntsvll;e, Ab3_arno
: - ) ,J_..._ Date
..J
o>
n_Z
mtO
A
>
>
A
>
0>
>
>
A
+2.5
-2.5
+2OO
0
-2C_3
+103
0
-100
+2OO
0
-200
+0.7
0
-0.7
CBRM No. 037Pc,t I",Io.
Test Trickle Charger Off with ESE on Pins T&P
Photo 6.11.2.3-10
!
100 wSEC/DIV'
.j
Photo 6. II. 2.3-11
_-o . .
i
2 I',_SEC/DP,,'
t
i
1
Q
!
Photo 6.1t. 2.3-12
I00
I
rm r"_, ;" " ' ": Federal Sy:,!em'_ Division
ili . , Sp_ce Sy;',_:n;Cen;er.I_._ L.:.J L _ 'i :j Hunlsvitle,/_.t;;bzma
.[iY Do,o_.
Sample
Ser;cl No.
Port No CBRM No. 037
lest Trickle CharGer ESE Pins on T&P
h-_.1o>v
>-
÷l
-1
i&
¢
?
I
Photo 6.11.2.3-13
ESE Pins T&P
Trickle Charger Off
I-,-1O:>
>-
+1
0
-1
4 MSEC/DIV
1!
t4
J
i
Photo 6. !1.2.3-14
ESE Pi}ls T&P
Trickle Charger On
=o
101
, : - ' _ r;:i!,,rt_l _y:;I_FR', l't '¢;',1 ')_
_.;" :ce _;y'-'.¢' n$ C' ._[er.
SHEET Do,o
P_::t No, CB_M No. 037
le_t SAS 15 Volis On with ESE on Pins G&P
I-._I
O>
k-
Z
p:-
L)
+2.5
-2.5
L ...............
1 MSEC/DIV
Photo 6. II. 2.3-15
>.
>-
>
O
>
+5OO
0
-500
+100
0
-I00
I......
2 ;,ASEC,'DI V
,i
.!
!
Photo 6.11.2.3-16
f
L
(-
t
L
I
102
I.:_, _ r"_ I' 7 Federal Systems Division
_._ _ , ' Space Syslem_ Center,_ I_ Y L_ Huntsville,Alabamn
D, T& i Dote
Sample
Ser;ol No.
Po, t No, CBRM No. 037
Te5_ SAS ] 5 Volts On with ESE on Pins T&P
_JO>
DnZ
trco
+2.5
-2,5
1 MSEC/DIV
Photo 6. Ii. 2.3-17
+500
0
>---500
>
>o
+100
-I00-j
!
I
i
2 MSEC/DIV
Photo 6, II. 2.3-18
lo3/Io4
6.12 CHARACTERIZATION OF CAPACITORS REMOVED FROM FLIGHTCBRMs
6,12.1 Summary
Twenty-eight capacitors from eight different flight CBRM output filters were sub-
jected to elemctrJcal test and teardo_n inspection. A total of 165 elements from these
capacitors _ere opened and visually inspected. Of.' tim elements examit_ed, 153 had no
evidence of silver on the sl,:g, and 15 had indication of "slight" silvering on the slug.
(See Fi_Ire 6.12.1-1). All traits u'ere fotmd to have dc leakage characteristics within
specified limits.
In comparison, four capacitors from relatively new CBRM input filters were examined.
Four eiemer:ts from each capacitor _'ere inspected. Nine of these elements had no
evidence of silver; hox_ever, three exhibited "slight" silver and four exhibitcd "tight"
silver deposils on the anodes. It was also noted that on failed CBRM input capacitors,
fl_e degree of silver _as in the range of "medium" or "gross".
6.12.2 Conclusion
The application of tantalu:n _et slug capacitors in the flight CBRhi outpdt filter does
not _ppear to be detrimental to the ea;,acitors. Electrical test and visual observationsreveaied no a,.-emalies in these device's;.
6.12.3 Details
During the investigation of the CBRM capacitor failures, it x_as concluded that reverse
bias in conjunction _ith near rated for,,.ard voltage _as the most probable failure
cause. All failures had o:,curred on the input filter operating from 65 to 90 volts.
However, the same capacitors operating in the outgut filter (35 volts maximum) had
never experienced a ft_,iiure. The decision at that time was to rc_,,ork the flight
units, utilizing tantahml foil capacitors in the input filter. The foils selected have a
12,c'o higher working voll.nge and can tolerate some reverse bias. Physically the input
filter capacitors and the output filter capacitors are mounted (potted) in common
cavities _ithin the CBIIM. Replacing the input cai:,acitors made it necessary to also
replace outpttt capacitors. The prin:ary problem at this time u:as the lack of foil
assets. There _,ere only enough ne_ foils (m hand for lhe input filter. If the output
capacitors also _ere to be changed lo foils, a decision _as necessary. This decision
was either to use foils thai a vendor had in stock and _;ere five years old, or slip
schedule and have ne_ foil._ made. It _as decided it) characterize or perform a tear-
down inspection of the output capacitors as ,,hey _ere removed from the flight CBtlMs.
If the testi.,g and teardo_,,n of these unit_; found no anomalies, lhen the outptH capacitors
would be replaced with ne_x tantalum wet slugs _hich were available in stock.
105
LEGEND:S/N - SERIAL NUMBERCAPS -CAPACITOhSELEM-ELEMENTSEX- EXA:,:JlNEDAFF- AFFECTED* -- StLVEF_ PRESEHCE
1 -- SLIGHT/I-RACE2 - LIGHT3 - MEDIUM4 -- GROSS
S/N
50
52
35
40
33
CAPS ELEM
8
40
40
SILVER PRESEr_CE
SLUG
1!112 3
_.........
!1
o t!h
il[: LECTROLYTE '_
i;_ = 3!,!i,
' j!L._ i!2
t
54
47
56
TOTA LS
3
2
Jf__o _
2_ 8
2 j"i12j
1£,8 15
1 t
!i
CASE
2 3
5
2
i.q
7 11 1
4
F;'._c,re 6. 12. 1-1. Ci_aract(_r/zc_io,,?of Fligh', CBRM Outp4zt Filler Capacitors.
In£,
Note: Continuing investigation of the CBRSI circuits later verified the appli-
cation of reverse voltages to the input filter ca0aeitors but not to the
outl_ut filter c::lmeitors.
The capacitors to be inspecied wore removed from the flight CB1RMs by Brown
Engineering. Since the capacitorsx_ere hard pot_ed in the CBIIM, some physical
damage to the eai)acitors _s encountered during ca0acitor removal. This damage
was primarily to the he.,der and tern-inations of each capacitor assembly.
In order to eliminate shorted terminals to case during the teardo_'.n and characteri-
zation, the header _as cut off each ealmeitor. Electrical contacl v.as made to the
internal riser _ires, and leakage _\as measured at rated voltage. The metal can
was then removed by griliding, leaving the e[ements in a hard potted module eonfig-
uratiot_. Tl:e modules uere subsequently"depptted" (oottingdisso[ved) and the
individual elements _erc removed for machining.
Note: If any eapaeitc_r assembly exhibited an out-of-tolerance leakage, then
the leakage on each individual element in the assembly was measured.
The machining consisted oi cutting a _:;roove dov, n one side of the element, across
the bottom and back up tSc. oi;posite side. The depth of the groove _as eor, trvlled
so as to leave _l ]nil oflhe case wall intact. "l-he ease _as then carefully pried
open, much like a clam shell, and i_,,_nediately subjected to internal visual exami-nation.
To correlr_te the visual results a visual aid _as developed (reference Figure 2.2-1).
This visual aid categori:-ed the at_t_e;'.ram'e of sih'er on the slug or anode into four
areas: Slight, Light, ?,iediuln or Gross. Wilh this visual aid variations due to
inspection v. ere reduced. Also, data .'_ssessment was more r,_anageabIe.
107
Reproduced by NTIS
=';'"
I ll _
lmm |l
_0_
"0 .E
National Technical Information ServiceSpringfield, VA 22161
This report was printed specifically for your order
from nearly 3 million titles available in our collection.
For economy and efficiency, NTIS does not maintain stock of its vast
collection of technical reports. Rather, most documents are printed foreach order. Documents that are not in electronic format are reproduced
from master archival copies and are the best possible reproductions
available. If you have any questions concerning this document or any
order you have placed with NTIS, please call our Customer ServiceDepartment at (703) 605-6050.
About NTISNTIS collects scientific, technical, engineering, and business relatedinformation -- then organizes, maintains, and disseminates that
information in a variety of formats -- from microfiche to online services.
The NTIS collection of nearly 3 million titles includes reports describing
research conducted or sponsored by federal agencies and theircontractors; statistical and business information; U.S. militarypublications; multimedia/training products; computer software and
electronic databases developed by federal agencies; training tools; andtechnical reports prepared by research organizations worldwide.Approximately 100,000 new tRles are added and indexed into the NTIScollection annually.
For more information about NTIS products and services, call NTIS
at 1-800-553-NTIS (6847) or (703) 605-6000 and request the freeNT/S Products Cata/og, PR-827LPG, or visit the NTIS Web site
http:llwww.ntis.gov.
NTISYour indispensable resource for government-sponsored
information---U.S, and worldwide
,,.v_
U.S.DEPARTMENTOFCOMMERCETechnologyAdministration
NationalTechnicalInformationServiceS_ngfield,VA22161 (703)605-6000