Guide for Establishing and Maintaining a Calibration Constancy Intercomparison System for Microwave Oven Compliance Survey Instruments (FDA 88-8264)] (PDF Only)

Administrative guidance, not enacted legislation. This captured publisher edition preserves the listed document and attachments; later developments are not independently consolidated.

HHS Publication FDA 88-8264
Guide for Establishing and Maintaining
A Calibration Constancy Intercomparison System
for Microwave Oven
Compliance Survey Instruments
Revised March 1988
U S . DEPARTMENT OF HEALTH AND HUMAN SERVICES
Public Health Service
Food and Drug Administration
Center for Devices and Radiological Health
Rockville, Maryland 20857

PREFACE
Public Law 90-602, 	t h e Radiation Control for Health and Safety Act of 1968 (the
Act), d i r e c t s t h e Department of Health and Human Services t o evaluate production
testing and quality control programs carried out by t h e .industry t o assure adequacy of
safegaurds against hazardous electronic product 	radiation and t o a s s u r e t h a t 	t h e
products comply with performance standards.
Under t h e Act, manufacturers of microwave ovens, a product listed under 2 1 CFR
1002.61, are required to certify t h a t their microwave ovens are in compliance with all
of t h e applicable provisions of t h e Federal Performance Standard for Microwave Ovens,
21 CFR 1030.10. 	In order t o comply with microwave emission level provisions of t h e
performance standard, manufacturers must use properly calibrated microwave leakage
measurement instruments in their production and quality control testing programs.
This document has been prepared in order t o assist t h e microwave oven .manufac-
turers in establishing and maintaining a calibration constancy intercomparison system
for compliance survey instruments and replaces guidance 	previously 	issued 	by t h e
C e n t e r for Devices and Radiological Health (the Center). 	The guide also contains
sample documentation forms for recording calibration data.
Checks on t h e calibration constancy and use of microwave survey instruments a r e
two vital parts of a n adequate quality control and testing program. 	Please read t h e
guide carefully. 	If 	your instrumentation 	program does not 	include all t h e aspects
covered in t h e guide, you should modify your program. 	Any changes t o your quality
control program must be reported t o t h e C e n t e r using t h e instructions found in P a r t 8.0
of the Guide for Preparing Reports of 	Radiation Safety of 	Microwave Ovens, March
1985.
Walter E. Gundaker
Director
Office of Compliance

FOREWORD
In October 	1982, the Food and Drug Administration established the Center for
Devices and Radiological Health (CDRH) by merging the Bureau of Medical Devices and
the Bureau of Radiological Health.
The Center develops and implements national programs t o protect the public health
in the fields of medical devices and radiological health. 	These programs are intended to
assure the safety, 	effectiveness and proper labeling of 	medical devices, 	to control
unnecessary human exposure t o potentially hazardous ionizing and nonionizing radiation,
and to ensure the safe, efficacious use of such radiation.
The Center publishes the results of its work in scientific journals and in its own
technical reports. 	These reports provide a mechanism 	for disseminating 'results of
CDRH and contractor projects. 	They are sold by the Government Printlng Office and/or
the National Technical Information Service.
Also, CDRH technical reports in radiological health are made available t o the World
Health Organization (WHO) under a memorandum of agreement between WHO and the
Department 	of 	Health 	and 	Human 	Services. 	Three 	WHO 	Collaborating 	Centers,
established under the Bureau of Radiological Health, continue to function under CDRH:
WHO Collaborating Center for Standardization of Protection Against Nonionizing
Radiations;
WHO Collaborating Center for Training and General Tasks in Radiation Medicine;
and
WHO Collaborating Center for Nuclear Medicine.
We welcome your comments and requests for further information.
/ 	john C. Villforth
irector
for Devices and
Radiological Health

CONTENTS
Page
....................................................	Foreword 	i
P r e f a c e 	...................................................... 	ii
1.0 	Introduction ................................................ 	1
2.0 	Procedures ................................................. 	2
...................................	2.1 	Preoperational Procedures 	2
2.2 	DailyChecks ............................................ 	2
2.3 	Thirty-Day Check: Initial Establishment of t h e Reference FieId ....; ... 	4
2.4 	Thirty-Day Check: Initial Polarization Response Check
..............................	of t h e Compliance Instruments 	6
2.5 	Thirty-Day Check: Constancy of the Reference Field
and Compliance Instruments 	................................
2.6 	Annual Calibration ........................................
2.7 	Periodic Recalibrations ......................................
2.8 	Repair .................................................
Appendixes
1. 	Calibration Constancy Intercomparison System .................... 	11
2 . 	Block Diagram of t h e System ................................. 	13
3. 	Daily Check Record ....................................... 	14
4 . 	Renewal of Records ....................................... 	15
5. 	12-Month Historical Record .................................. 	16
6 . 	Nonref erence Field Constancy ................................ 	17
............................	7 . 	30-Day Instrument Comparison Log 	18
8. 	2 dB Comparison Check Record ................................ 	19
9. 	Flowchart for 30-Day Calibration Constancy Program ............... 	2?
.....

GUIDE FOR ESTABLISHING AND MAINTAINING
A CALIBRATION CONSTANCY INTERCOMPARISON SYSTEM
FOR MICROWAVE OVEN COMPLIANCE SURVEY INSTRUMENTS
1.0 	INTRODUCTION
Manufacturers of microwave ovens must use properly calibrated microwave leakage
measurement instruments in their production testing and quality control programs to
assure compliance with the Federal Performance Standard for Microwave Ovens (21
CFR 1030.10).
As an alternative to frequent recalibration against absolute standards, the Center
for Devices and Radiological Health (CDRH)will allow manufacturers to establish a
system that 	demonstrates 	that 	the calibration 	of 	each instrument 	remains 	within
prescribed limits, over time, subsequent to absolute calibration* by 	the instrument
manufacturer or other qualified calibration facility. -
This document outlines procedures for establishing and maintaining a "calibration
constancy intercomparison 	system'' 	(referred to 	as the 	"system" 	throughout 	this
document) capable of assuring adequate constancy of calibration, while minimizing the
necessity for absolute recalibration.
It must be noted and understood that the system described herein is not intended for
use in calibrating hand-held or automated scanning survey instruments, nor is it intended
to be used to transfer calibrations from one instrument to another. 	I t is only intended
to provide a means to determine whether an instrument's calibration is changing with
time, and whether it is continuing to function properly. 	The underlying principle is that
constancy of calibration may be determined by comparing an instrument's reading with
its own prior 	readings and with other 	meters' readings when 	it 	measures a 	fixed
parameter.
The procedures for establishing and maintaining a calibration constancy intercom-
parison system 	consist 	of 	(1) maintaining an 	apparatus or system 	to 	perform 	a
calibration constancy intercomparison, (2) documenting the results using a log or record
to demonstrate constancy over a period of time, and (3) adhering to an appropriate
schedule of checks to assure proper function and calibration constancy of each instru-
ment. 	Such a schedule of checks and a simple acceptable calibration constancy inter-
comparison system are described below. 	Briefly, the calibration constancy intercom-
parison system consists mainly of:
(1) a microwave source,
(2) a radiated power monitor (RPM),
(3) an anechoic environment with a radiator and probe-holding fixture, and
(4) a local calibration reference (LCR).
A more detailed description is given in Appendixes 1 and 2. 	A system using this
apparatus has been demonstrated to be accurate and practical to reproduce and use.
Any calibration constancy intercomparison system 	that 	can be 	shown to 	provide
comparable performance (at least within the same limits as the system described above)
would be acceptable.
*Absolute calibration (i.e., 	cafibration against absolute standards) means calibration of a
system in which all measurement parameters are traceable, with specified uncertainty,
to the absolute standards maintained by the National Bureau of Standards or equivalent.
1

2.0 	PROCEDURES
2.1 	PREOPERATIONAL PROCEDURES
2.1.1 	Requirement
Preoperational procedures, 	as described by t h e manufacturers of 	t h e calibration
constancy intercomparison system and t h e survey instruments, must be performed prior
t o use t o assure t h a t t h e system and survey instruments are functioning correctly and
properly adjusted.
2.1.2 	Instrumentation and Method
Various preoperational procedures a r e described in t h e 	manuals provided by t h e
manufacturers of t h e calibration constancy intercomparison system and survey instru-
ments. 	For example: --
(1) a suitable period of time must be allowed a f t e r turning on t h e system and survey
instruments for nwarm up1' o r stabilization;
(2) survey instruments' probe cones (spacers) should be checked daily and replaced if
dirty o r worn;
(3) battery-operated survey instruments must have their batteries checked thoughout
the day;
(4) charger covers must be in place. on t h e survey instruments with rechargeable
batteries;
(5) the voltage supply for t h e AC-voltage 	powered instruments 	must be checked
periodically;
(6) the survey instruments and system should be checked f o r any evidence of physical
damage, and, if damaged, t h e entire instrument and/or system must .be checked
prior t o e a c h oven survey;
(7) baseline o r "zero1' reading on t h e survey instruments and/or 	system 	must be
checked prior t o each oven survey; and
(8) all o t h e r preoperational procedures specified in t h e owner's survey instrument
and/or system manuals must be followed.
2.2 	DAILY CHECKS
2.2.1 	Requirement
The polarization response of e a c h survey instrument must be checked daily prior t.0
i t s use t o determine any gross shifts in response 	which could 	indicate 	instrumen'f
damage. 	*..

2.2.2 	Instrumentation and Method
Make a check of the survey instrument's polarization response using an adequately
stable microwave field source. 	This microwave field should be linearly polarized, have
the correct frequency, and cause at least a mid-scale 	meter reading on the survey
instrument. 	A field setting (or power density) of approximately 1.0 mW/cm2 can be used
for generating a mid-scale meter reading. 	However, different levels may be necessary
for other specialized meters such as probe/pre-amp 	units from the automated oven
survey systems (scanners). Specifications for performing polarization response measure-
ments should be obtained from the survey instrument manufacturer. 	A probe-holding
fixture must be provided to keep the probe on a constant axis while the probe is slowly
rotated (approximately 2 revolutions per minute) about the axis normal to its detection
plane (usually the handle axis). 	Because unnecessary changes in the probe orientation or
position may alter the measurement of polarization ellipticity, the probe fixture should
prevent any unnecessary horizontal, vertical, or transverse motion while the probe is
being rotated (see Figure 1). -
Figure 1. Probe rotation should prevent any horizontal, vertical,
or transverse motion while the probe is being rotated
. 	Because each survey instrument has its own ellipse of polarization response (also
known as polarization ellipticity error), the percentage or polarization error of each
instrument must be calculated to determine whether each instrument is operating within
the limit of the manufacturer's specification.
A common way to find the percentage of polarization ellipticity error is as follows:
(1) place the probe in the fixture and turn on the microwave power source so the
survey meter reads a mid-scale value (for example 1.0 m W/cm2);
(2) rotate the probe 360 degrees and note the minimum and maximum readings and
calculate the mean (see Equation 1); and
(3) calculate the percentage of polarization ellipticity error using Equation 2. 	The
result must be equal to or less than the percentage error limit of polarization
ellipticity specified by the instrument manufacturer (e.g., 	+5 percent, or a total
of 10 percent). 	Any instrument with a polarization ellipticity error greater than
the allowed tolerance must be repaired or discarded.
Ano:her way to find the percentage polarization ellipticity error is as follows:
(1) place the probe in the fixture and turn the power on so the survey meter .reads a'
mid-scale value of 1.0 mW/cm2;
(2) rotate the probe 360 degrees and find the minimum reading only;

(3) change t h e power s e t t i n g until t h e survey meter reads 1.0 mW/cm2 at t h e mini-
mum polarization response;
(4) r o t a t e t h e probe again t o obtain t h e highest polarization response reading; and
(5) s u b t r a c t 1.0 	from t h e highest 	polarization response reading and multiply t h e
result by 100. 	The result must b e equal t o o r less than t h e t o t a l percentage
polarization 	ellipticity error specified by t h e 	instrument 	manufacturer 	(see
Equation 3 below).
Maximum + Minimum
Mean =
Maximum - Minimum
% Total Polarization Ellipticity Error = Mean
Note: to get _+ percent, divide Total by 2
%Total Polariuztion Error = (Maximum - 1.011 00 	(3) 	'
For example, a total error for this procedure of 	10 percent 	is equivalent t o t h e
survey instrument 	manufacturer's 	specification of f5 percent polarization ellipticity
error. 	The daily check 	method applies t o instruments employing planar detectors.
Alternate methods for other types of detectors may need t o be developed.
2.2.3 	Documentation
A s e p a r a t e "daily check" record should be kept for every survey instrument for com-
pliance testing on t h e production line and audit. 	The record should include: t h e model
and serial number of 	t h e instrument 	(probe 	modevserial 	number 	if 	necessary), 	t h e
percentage e r r o r in polarization ellipticity specified by t h e instrument manufacturer,
t h e results of t h e test, and t h e name of t h e person who performed t h e test. 	If an instru-
ment fails t h e test, t h e disposition of t h e instrument should be clearly indicated in t h e
"re marks" column. 	Following t h e repair, t h e instrument's record must be renewed (e.g.,
s t a r t e d over). 	See Appendix 3 f o r a n example of t h e documentation; any equivalent
c h a r t is acceptable. 	Appendix 4 of this guide contains a c h a r t which describes briefly
when and under what circumstances certain records must be renewed.
2.3 	THIRTY-DAY CHECK: 	INITIAL ESTABLISHMENT OF THE REFERENCE FIELD
2.3.1 	Requirement
A reference field must be established prior t o checking t h e 30-day calibration con-
s t a n c y of every instrument (including spare scanner probes o r hand-held survey meters)
used for microwave oven compliance determinations.
2.3.2 	Instrumentation and Method of Establishing t h e Reference Field Using t h e LCR
One instrument suitable for compliance determinations is set aside and not used for
production o r quality control t,esting. 	I t is retained as t h e local calibration reference
(LCR),having been calibrated by i t s manufacturer o r other suitable source. 	This instru-
ment is s e t aside t o minimize t h e probability of damage and/or loss of 	calibration

through accident o r t h e wear and t e a r of normal use. 	It is recommended t h a t t h e instru-
ment selected as t h e LCR have a power density calibration which is near t h e geometric
mean of 	t h e power 	density 	calibrations of 	a l l 	instruments used 	f o r 	compliance
determination (e.g., 	if t h e spread of calibration factors is .98 t o 1.08, 	then t h e LCR
should have a f a c t o r near 1.03). 	It is also recommended that t h e LCR be of t h e s a m e
b r a n d m a k e as t h e survey instruments being checked.
An LCR is not t o be misconstrued as a transfer standard. 	If i t s calibration is 41 dB,
it cannot be used a s a transfer standard for instruments whose accuracy is also f l dB.
Using t h e calibration constancy intercomparison system, an initial reference field is
established with t h e probe of t h e LCR positioned a t t h e mean of i t s polarization ellipse.
Set t h e field at a power density of approximately 1.0 mW/cm"s 	read on t h e LCR. 	The
mean of t h e polarization ellipse of t h e LCR must be located by finding t h e mean of t h e
maximum and minimum readings when the LCR probe is slowly rotated 360 degrees in
the calibration constancy intercomparison system chamber. 	When the LCR probe is
positioned in t h e system at t h e mean of its polarization ellipse, t h e power density can be
adjusted (if necessary) t o t h e appropriate reference field (usually 1.0 	m W/cm2). 	The
values of t h e reference field should be specified by t h e instrument manufacturer o r
calibration facility. 	A f t e r s e t t i n g t h e reference field, t h e percentage of polarization
error of t h e LCR must be checked t o verify t h a t i t meets t h e instrument manufacturer's
specification. 	When t h e reference field has been s e t up using t h e LCR, i t is important
that the readings on t h e radiation power monitor (RPM), (e.g., 	Forward, Reflected, and
Difference (Net)) o r from another power meter a r e recorded and noted on t h e log record
(Appendix 5).
2.3.3 	Instrumentation and Method of Establishing t h e Reference Field Using t h e RPM
(Optional)
An a l t e r n a t i v e method for establishing t h e reference field can be performed by using
t h e radiated power monitor (RPM). 	After t h e LCR has been calibrated by i t s manu-
f a c t u r e r o r o t h e r suitable source, use t h e LCR t o find t h e mid-scale value (e.g., 	obtain a
1.0 mW/cm2 midpoint scale reading on t h e LCR, then record t h e RPM readings). 	For t h e
subsequent monthly checks, t h e reference field can be s e t by using the RPM t o repeat
the same reading obtained t h e first month. 	When the reference field has been set up
using the RPM, i t is important t h a t t h e readings from t h e LCR a r e recorded and noted
on the log record (Appendix 5).
2.3.4 	Documentation
A "12-month 	historical record" should be kept on t h e reference field instruments.
Both the LCR readings (Minimum, Maximum, and Mean readings) and RPM net power
readings (Forward, Reflected, and Difference) a r e t o be recorded and noted as reference
readings. 	O t h e r items such a s ambient temperature and t h e name of t h e person who
performed t h e t e s t s should also be recorded. 	If t h e LCR fails t h e polarization ellipticity
check, the disposition of t h e LCR should be clearly indicated in t h e "remarks" column.
Following t h e repair of t h e LCR, t h e 12-month historical record and all o t h e r records
must be renewed (see Appendix 4). 	See Appendixes 5 and 6 f o r example:: 	of t h e docu-
mentation; any equivalent c h a r t s a r e acceptable.

2.4 	THIRTY-DAY CHECK: 	INITIAL POLARIZATION RESPONSE CHECK O F THE
COMPLIANCE INSTRUMENTS
2.4.1 	Requirement
A f t e r t h e reference field has been established at a specified power density (for
example 1.0 mW/cm" 	as described in P a r t 2.3.2 	o r 2.3.3, 	t h e polarization response of
each survey instrument t o be used for compliance testing must be initially checked. 	T h e
polarization ellipticity error of 	each survey compliance instrument 	must 	meet 	t h e
instrument manufacturer's specification. This initial check must be performed prior t o
initial use on t h e production line, when i t is taken out of spare stock, o r a f t e r i t has
been returned from repair.
2.4.2 	Instrumentation and Method
Follow t h e s a m e procedures outlined in P a r t 2.2.2 	above. 	-
-
2.4.3 	Documentation
For each instrument used in ~ r o d u c t i o no r in audit, a separate "30-day 	instrument
comparison log" record 	should de maintained. 	This log should include t h e following.
model and serial number of 	each instrument (and probe 	model and serial number 	if
necessary); 	date; 	t h e 	minimum, 	maximum, 	and 	mean of 	t h e polarization 	response
readings; 	percentage of 	polarization ellipticity error; 	t h e highest 	and lowest 	mean
readings since t h e last calibrationhepair; and verification t h a t each instrument met t h e
polarization ellipticity e r r o r limit specified by t h e instrument manufacturer. 	If t h e
instrument's polarization ellipticity error is outside t h e specification, t h e disposition of
t h e instrument should be clearly indicated on t h e record. 	See Appendix 7 for an exam-
ple of t h e documentation; any equivalent chart is acceptable.
2.5 	THIRTY-DAY CHECK: 	CONSTANCY O F REFERENCE FIELD AND COMPLIANCE
INSTRUMENTS
2.5.1 	Requirement f o r t h e Reference Field
The reference field must be established for t h e initial setup of t h e system and survey
instruments (see P a r t 2.3). 	For t h e subsequent monthly checks, the power density of t h e
reference field 	in t h e calibration constancy 	intercomparison 	system should be 	re-
established by adjusting t h e microwave power source t o repeat either t h e previous LCR
or R P M reference reading. 	A f t e r t h e reference field is reset using one of these two
readings, t h e o t h e r reading (from t h e RPM o r LCR) must agree with all of i t s preceeding
readings within 10 percent. 	That is, no two of these nonreference readings, from one
annual calibration o r repair t o t h e next, may disagree by more than 10 percent.
For example, if t h e r e f e r e n c e field is set using t h e reading of t h e LCR, then t h e
monthly RPM (Diff) nonreference readings (ail readings accumulated since t h e last LCR
calibration) must be verified t o be within 10 percent of e a c h other. 	If t h e reference
field is s e t using readings of 	t h e RPM, 	t h e n t h e monthly LCR nonreference 	mean
readings (all mean readings accumulated since t h e last LCR calibration) of t h e polari:
zation 	ellipse 	must 	be verified 	t o b e 	within 	10 p e r c e n t 	of 	e a c h other. 	If 	t h e '
nonreference readings disagreed by more than 10 percent, then t h e nonreference 	instru-
ment must be repaired (or discarded). 	See Appendix 6 f o r a n example of t h e documen-
t a t ion. 	Following these act ions, new 12-month historical records (Appendixes 5 and 6),
and all other records, must be renewed (see Appendix 4).

2-5.5 	Requirement f o r Each Compliance Instrument
A f t e r e a c h compliance survey instrument receives i t s initial check of t h e polari-
zation ellipse (see P a r t 2.4.1), 	t h e mean reading of 	each instrument 	must then be
compared with all of i t s own previous monthly mean readings f o r constancy within 10
percent. 	For t h e comparison, use only those mean readings taken since t h e survey
instrument was last calibrated (or repaired), 	o r from when t h e LCR (or RPM) was last
calibrated, whichever is most current.
For those survey instruments falling outside t h e 10 percent limit, possible sources of
t h e failure must b e traced. 	First, t h e survey instrument itself should be checked for
obvious problems such as weak batteries. 	Second, t h e system should be examined.
Problems with equipment must be adequately resolved before proceeding further. 	If t h e
problem 	was t r a c e d t o t h e survey instrument, 	then the survey instrument 	must be
repaired (or discarded) and t h e 30-day instrument comparison log (Appendix 7) and daily
check record (Appendix 3) for t h a t compliance survey instruments must be renewed. 	If
the problem was t r a c e d t o t h e system, then t h e system must be repaiped and all records
must be renewed. 	See Appendix 4 f o r further explanation regarding reiiewal of records.
2.5.3 	Requirement for Calibration D a t a Constancy
The least sensitive and most sensitive survey instruments (including the LCR) must
not disagree by more than 2 dB, including t h e extremes of 	polarization ellipse. 	A
monthly.2 dB check must be performed by comparing t h e ratio of t h e highest polariza-
tion ellipse reading from all of 	t h e instruments (including t h e LCR) t o t h e lowest
polarization ellipse reading from all of the instruments (including t h e LCR). 	This ratio
must not exceed 1.59 (2 dB). 	Do not include those instruments t h a t did not pass t h e
daily check or t h e 1 0 percent comparison' in t h e 30-day 	check. 	The 2-dB check of
instruments e a c h month is done on a cumulative baiis; t h a t is, t h e comparison period is
t h e time between LCR calibrations (or repairs).
If, during t h e comparison check, t h e ratio exceeded 1.59 (or 2 dB), then t h e e n t i r e
system, 	including t h e 	L C R 	and 	individual compliance survey instruments, 	must 	be
examined and repaired as necessary. 	Repeat t h e 10 percent comparison check of t h e
individual survey instruments (see P a r t 2.5.2). 	Problems with all of the equipment must
be adequately resolved before proceeding further. 	Then repeat t h e 2 dB comparison
check t o d e t e r m i n e that t h e ratio is less than o r equal t o 1.59 and ensure that any
remaining d e f e c t i v e instruments a r e found. 	Following these actions, then a new 2 dB
comparison c h e c k record (Appendix 8) and most o r all other records must be renewed
(see Appendix 4). 	Appendix 9 illustrates the proper 30-day constancy check sequence in
t h e form of a flowchart.
2.5.4 	Documentation
Two s e p a r a t e records must be maintained. 	T h e first record, "30-Day 	Instrument
Comparison Log" (Appendix 7), should be kept f o r every instrument used on t h e produc-
tion line and in audit. 	T h e 30-day instrument comparison log should verify t h a t t h e
highest and lowest mean readings of t h e individual instrument are within 10 percent of
e a c h other. 	If not, t h e disposition of t h e instrument should b e clearly indicated on t h e
records. 	Following these actions, t w o new records, one f o r recording results of t h e daily)
check (Appendix 3) and a second, t h e 30-day instrument comparison log (Appendix 7),
must be s t a r t e d (see Appendix 4).

The second separate record, "2 dB Comparison Checktt (Appendix 8), should be kept
t o monitor t h e calibration d a t a constancy of all t h e instruments used on t h e production
line and in audit since t h e last LCR calibration. 	Exclude those instruments t h a t did not
pass the daily check, and/or t h e 10 percent comparison in t h e 30-day check. 	This record
should include t h e highest and lowest polarization response readings taken from all of
t h e instruments (including t h e LCR). 	These values should be compared and t h e ratio of
t h e highest polarization ellipse reading t o t h e lowest polarization ellipse reading during
t h e entire period should be less than 1.59 (2 dB).
If t h e r a t i o exceeds 1.59, make a complete check of t h e entire system and record t h e
reason f o r failure. 	If t h e failure is traced t o a defective individual survey instrument,
t h e survey instrument must be taken out of production o r audit facility for repair. 	Then
continue t o use t h e same 2 dB comparison check log. 	When t h e repaired survey instru-
ment comes back, renew t h e 30-day instrument comparison log (Appendix 7) and daily
check record (Appendix 3) for t h e survey instrument. 	If t h e problem is traced t o t h e
system, then a new 2 dB comparison check record and all other records must be s t a r t e d
following correction of t h e problems (see Appendix 4). 	See Appendix 8 for a n example
of t h e documentation; any equivalent chart is acceptable.
2.6 	ANNUAL CALIBRATION
2.6.1 	Requirement
Absolute calibration of t h e LCR must be performed annually. 	There'may be a need
for annual absolute calibration of compliance survey instruments, scanner probe/pre-
amp, radiated power monitor (RPM), power meter, etc., 	depending upon t h e recommen-
dations of t h e instrument manufacturer.
It is important that t h e microwave oven manufacturer check with t h e instrumenta-
tion manufacturer for specifications o r recommendations concerning annual calibration.
Whenever t h e LCR has returned a f t e r being calibrated and/or repaired, it is important
t o immediately determine its percentage of polarization ellipticity error and record t h e
results (see P a r t 2.3).
2.6.2 	Instrumentation and Method
The LCR must be returned t o t h e instrument manufacturer or other qualified cali-
bration facility for absolute calibration (12-month intervals). 	Some compliance survey
instruments o r o t h e r electronic components of t h e system also may require absolute
calibration annually.
2.6.3 	Documentation
A separate record of t h e LCR calibration and repair should be kept. 	This log should
include model and serial number, calibration date, and a list of any repairs and their
dates. 	The 12-month historical record (Appendix 5) of t h e LCR is renewed every t i m e
t h e LCR h r s been returned from annual calibration. 	The 30-day instrument comparison
log of each compliance survey instrument (as well as scanner probe/pre-amp 	units) is
renewed when t h e LCR has been returned from annual calibration.

2.7 	PERIODIC RECALIBRATIONS
2.7.1 	Requirement
The absolute .calibration of all instruments should be performed periodically.
2.7.2 	Instrumentation and Method
In accordance with good engineering and manufacturing practice, i t is recommended
t h a t all instruments used t o make compliance determination measurements (with t h e
exception of LCR and RPM) be returned t o t h e instrument manufacturer factory o r
o t h e r qualified calibration facility f o r absolute calibration at least once every 3 years.
2.7.3 	Documentation
See P a r t s 2.2.3 and 2.3.4 --
2.8 	REPAIR
2.8.1 	Requirement
Any instrument which has been found t o . b e damaged o r failed t o remain within its
established constancy limits must not be used in t h e quality control and testing program
until it has been repaired and recalibrated by a n instrument 	manufacturer o r other
qualified calibration and repair facility. 	Before repaired instruments a r e used again,
t h e i r operational functions and their polarization respons6s must be checked (see P a r t
2.2). 	Following these actions, t h e instrument's records must be renewed (see Appendix
4). 	It is recommended t h a t repairs be performed only by t h e instrument manufacturer
o r o t h e r qualified calibration facility unless t h e microwave oven manufacturer has in
place an adequate repair facility t o perform certain minor repairs (such as broken probe
cables, 	d e f e c t i v e meters, 	etc.) 	and 	has 	t h e capability of 	checking t h e i r 	results.
Microwave oven 	manufacturers 	will be required 	t o submit specific technical repair
procedures t o CDRH for review before implementing them.
2.8.2 	Instrumentation and Method
Whenever t h e compliance survey meters, LCR, o r scanner probe/pre-amp 	units a r e
returned from t h e instrument manufacturer a f t e r repair, t h e i r polarization ellipse must 	.
be checked against t h e instrument 	manufacturersf specifications and verified on t h e
record prior t o their use.
2.8.3 	Documentation
The disposition of t h e defective instrument can be recorded in t h e "remarks" column.

APPENDIX 1
CALIBRATION CONSTANCY INTERCOMPARISON SYSTEM
A block diagram of the calibration constancy intercomparison system is provided in
Appendix 2. 	The parenthetical numbers preceding each paragraph below and used within
those paragraphs refer t o the numbers in the Appendix 2 diagram.
(1) SOURCE
The source is a microwave generator with (a) an appropriate power output (probably
approximately 0.5 to 5 0 Watts maximum output) that may be variable (either internally
or by some external attenuator) and (b) a frequency in the relevant ISM band (915 5 25
MHz or 2 4 5 0 + 50 MHz). 	Depending on the radiator or anechoic enclosure and on the
inherent frequency stability of the source, some control over the source output fre-
quency may be necessary. 	The source might be a magnetron or klystron oscillator, or
might be a low-powered 	source amplified, e.g., 	by a traveling wave tube amplifier.
However, any type of source with adequately stable or controllable output parameters
could be used.
(2), (3), (4) DIRECTIONAL COUPLERS
The directional couplers are used to sample a fixed fraction of the power flowing
through them in either the forward 'or reverse direction. 	This fraction should be chosen
so as to deliver a signal to items (5), (6), or (7) which is of a magnitude appropriate t o
the 	specific device. 	Furthermore, 	this 	fraction 	(or 	ffcoupling factor") 	must 	be
adequately 	invariant 	with frequency over the range of- frequency variations of 	the
source. Two of these couplers might be replaced by a single bi-directional coupler if
desired. 	Furthermore, if the source has some built-in frequency monitor, (4) might be
eliminated. 	Couplers (2) and (3) should be adequately insensitive to power flowing in a
direction opposite to that 	intended t o be measured (i.e., 	these couplers should have
adequate "directivityf'). I t is not absolutely necessary that directional couplers be used,
however some means of differentiating between forward and reflected power (to allow
determination of net radiated power) must be included.
(S), (6) POWER METERS
These power meters must provide repeatable, adequately sensitive, and reliable re-
sponses to both absolute levels and relative variation of power. 	They should be rela-
tively insensitive t o frequency variation. 	I t may be necessary to insert attenuators
between the directional couplers and the power meters to bring the input to the power
meters into an appropriate range. 	I t is not absolutely necessary that power meters be
used for these applications. 	Whatever type of detector is used, some reliable means of
calibration must be employed in an annual calibration. 	In the system illustrated, if the
source can be demonstrated to be adequately stable, a single pqwer meter might be
alternatively switched to the output of the couplers (2) and (3) by means of a coaxial
switch. 	The power meter or meters and directional couplers form the radiated power
monitor (RPM).
(7) FREQUENCY MEASURING DEVICE
Some means must be employed to verify that the frequency of the radiated signal
remains within appropriate limits. 	These limits are closely related t o the quality of the

anechoic environment (9). 	If standing waves of considerable magnitude exist within this
environment, a change in t h e frequency of t h e radiated signal can cause a variation in
t h e spatial distribution of t h e field and have a significant e f f e c t on t h e reading of a
stationary probe in t h e fixture (10). 	Thus, t h e choice of both source (1) and anechoic
environment (9) must be made with this phenomenon 	in mind. 	Although frequency
counters are o f t e n available in microwave laboratories, any of several other possibilities
might be employed (e.g., 	a tunable cavity might be used with a crystal o r power m e t e r
monitoring t h e output).
Clearly, t h e use of a n anechoic environment of appropriate characteristics with an
adequately s t a b l e source can eliminate t h e necessity t o employ t h e frequency measuring
device shown in t h e illustration.
(8) 	RADIATOR
Some means of radiating t h e generated power into space must be provided. :Among a
multitude of possible devices t h a t might be 'employed for this purpose a r e horn, log
periodic and dipole antennas, a s well a s flared o r simply open-ended waveguide sections.
Considerations in t h e choice include directivity of t h e radiator vs. the reflectivity and
absorption of t h e anechoic environment, as well as the frequency response of t h e radi-
a t o r vs. t h e source frequency stability. 	If a stable source is t o be employed, a wave-
guide or coaxial tuner may be employed t o improve t h e system match t o the radiator t o
minimize reflected power.
(9) ANECHOIC ENCLOSURE
An anechoic environment should be provided t o minimize t h e effects of t h e labora-
tory environment on t h e measurement. 	The quality and size of this anechoic environ-
ment will determine t h e magnitude of t h e standing wave in t h e region of t h e instrument
under comparison.
(10) PROBE-HOLDING FIXTURE
A stable fixture t o hold the instrument being evaluated in a rigid, repeatable position
must 	be employed. 	This fixture should be 	made of 	an electrically 	non-conductive
material t o minimize standing waves and perturbations of t h e fields. 	In general, t h e
g r e a t e r t h e magnitude of t h e standing wave and t h e closer t h e probe t o the radiator, t h e
more critical a r e t h e repositioning requirements of this fixture.

APPENDIX 4
IZENDWAL OF RECORDS
Documentation Forms
Appcndix 3 	Appendix 5 	Appendix 6 	Appendix 7 	Appendix 8
Daily Check 	12-Month 	Nonreference 	30-Day Instrument 	2 dB Comparison
Status 	Record 	Historical Record 	Constancy 	Comparison Log 	Check
Daily Check (2.2)
Failed survey instruments 	, 	Renew 	Same 	Same 	Same 	Same
LCRIRPM Monthly (2.3.2 - 2.3.3)
Polarization failure of LCR 	Renew 	Renew 	Renew
Failed RPM 	Renew 	Renew 	Renew
30-Day Check of Individual
Survey Meters (2.4.1,2.5.2)
Polarization failure 	Renew 	Same
P 	10% comparison failure 	Renew 	Same
VL
30-Day Check of
Nonreference Meter (2.6.1)
10% comparison failure 	Renew 	Renew
2 dB Comparison (2.5.3)
System or LCR failure
Survey meter failure
Renew 	Renew
Renew 	Same
Renew
Renew
Renew
Renew
Same 	Renew 	Same
Same 	Renew 	Same
Renew
Renew
Same
Renew 	Renew
Renew
Renew
Renew
Same
LCR Calibration (2.6)
Annual calibration 	Renew 	Renew 	Renew 	, , 	Renew 	Renew
Numbers in parentheses refer to sections of this report.

APPENDIX 5
12-MONTH HISTORICAL RECORD
Name of constancy system:
Date of last LCRIRPM calibration:
Make of LCWRPM:
Model number of LCWRPM:
Serial number of LCR:
Probe model number on LCR:
Probe serial number on LCR:
LCR Polarization Ellipticity Readings
RPM 	RPM 	RPM 	LCR 	LCR 	LCR 	LCR 	Ambient
Date , 	Fwd 	Rev 	Diff 	Max 	Min 	Mean 	% Ellip 	Temp 	Remarks
NOTES: 	(1) The LCR must be positioned a t its mean polarization response prior to setting the reference field.
'
(2) If the reference field is set using the LCR readings, then the monthly RPM (Diffl readings must be within 10%.
,(3) If the reference field is set using the RPM readings, then the monthly LCR (Mean) readings must be within 10%
(4) If the readings taken from the nonreference instrument disagreed by more than lo%, then this instrument must be repaired
(or discarded). Following these actions, a new Daily Check Record (Appendix 3) and all other records, must be renewed (2.5.1).
Maximum + Minimum
Mean = 2
~ a i i r n u m- Minimum
% Tota 1 Pola rizatwn Ellipticity Error = Mean
% Total ellipticity error limit specified by LCR instrument manufacturer is:

APPENDIX 6
NONREPERENCE FIELD CONSTANCY
The monthly reference field is set by (check one only): 	LCR 	RPM
If you selected the LCR to set the monthly reference field, use column 1 only.
If you selected the R P M to set the monthly reference field, use column 2 only.
Column 1 	Column 2
Monthly 	(A) Highest 	(B) Lowest 	(A) Highest 	(B) Lowest 	Less than 1.lo?
Date 	RPM Diff (or P) 	RPM Diff (or P) 	LCR Mean 	LCR Mean 	A/B 	(YesINol*
* NOTES: 	If the reference field is set using the reading of the LCR, then the monthly RPM (Dim nonreference readings (all readings
accumulated since the last LCR calibration) must be verified to be within 10 percept. of each other. If the reference field is set
*: 	using readings of the RPM, then the monthly LCR mean nonreference readings (all mean readings accumulated since last
LCR calibration) of the polarization ellipse must be verified to be within 10 percent of each other. If the nonreference readings
failed the 10 percent comparison test, then the nonreference instrument must be repaired and recalibrated. Following this
action, all records must be renewed.

.paMauaJ aq ? s n w pJoaaJ s!q? uaqJ '(pap.ms!p J O ) pa?e.lq!leaa.r put? pa.I!odaJ aq I s n u ?! ' a ~ g ~ a j a p	s! Ja?aur X a ~ ~ n s	aq?jI (E)
' ~ a ~ n ? 3 8 j n u ~ u r	?uaurnJ?su! aq? ;Cq pag!mds ?!ur![ J o u a X?!x?d!lla uo!?t?z!~elodlelo?% aq? waur ? s n w luaurn.wu! q x r 3 (z)
. u o ! ? e ~ q g e 3a37 p n u u w uaam?aq lt?,walu! ay? Bu!.Inp (IsaMolo? ?saq8!q) a8ue.1 %OI u ! q ? ! ~aJr, s8u!pt?a~u v a w ay? ~ 	e 	y 	~	L p a ~( I ) :S~J,ON

APPENDIX 9
FLOWCHART FOR 30-DAY CALIBRATION CONSTANCY PROGRAM
. 	.
Start new records
I Check polarization ellipse o f LCR 1
LCR probe positioned t o read 	-
+
-- 	-
Set field inside system (usually 1 m~lcrn').
Record LCR and RPM settings.
I
Polarization OK? > No
Correct or repair
& recalibrate
C
or RPM reference 	RPM
I
,
All previous RPM DlFF
t 	I yes
readingswithin lo%?
I 	Calculate % ellipticity error.
I 	I
Yes
Set field (usually 1 m ~ l c m ~ ) .	Check polarization of each
Yes
compliance meter. Record MAX, MIN. & MEAN.
No
,
< Has LCR been calibrated
i n last 12 months?
of o w n highest reading since 	I
last LCWRPM calibration?
\
Yes
C
rument mean reading within 10% 	No
-
Repa~ror replace
instruments
Yes If problem found
All instrument readings, including 	No 	within system or
LCR. within 2 dB since last LCR 	) 	reference instru-
calibration? 	ment, have it
repaired or replaced
Yes 	L
j 	individual
1 	.