Text preview for : HP-Bench-Briefs-1994-04-06.pdf part of HP HP-Bench-Briefs-1994-04-06 HP Publikacje HP-Bench-Briefs-1994-04-06.pdf
Back to : HP-Bench-Briefs-1994-04-0 | Home
HEWLETT
PACKARD
SERVICE INFORMATION FROM HEWLETT-PACKARD
2nd Quarter 1994
CaAibraltiom of Time Base Oscillators
Jim BechtoldlEditor of 9,192,631,770 periods of transition bration interval
within the cesium atom." Frequency is Determination of the effects of noise
determined by counting the number of in frequency generating equipment
cycles over the period of a second.
Therefore, the definition of a clock can Determination of the effects of
be expressed as a device that counts the changing environmental conditions
number of seconds occurring from an
arbitrary starting time. This Article
Prior to analyzing the effects and im-
From this definition it appears that a pact of the above sources of error, it is
clock needs three basic parts. First, a necessary to determine the level of ac-
source of events to be counted. This curacy required and the tolerances es-
source can be labeled a frequency stan- sential for the individual application.
Counting and Clocks dard, frequency source, or time inter- Once the essential tolerances have been
val standard. Second, a means of accu- established, the sources of error can be
Counting intervals has been going on mulating these events or oscillations.
since man's beginning. Early time mea- analyzed to determine if they have an
Third, a means of displaying the accu- impact on system operation. If they do,
surements involved counting the num- mulation of time. Figure 1 shows a
ber of days in terms of sunrises, sun- in fact, affect the system operation, then
simple clock block diagram, including appropriate steps can be taken to re-
sets, or moons. Later, the day was di- a method of presetting the arbitrary
vided into smaller incrementsby using duce that impact. This article will de-
starting time and obtaining an electri- scribe the frequency calibration inter-
an hourglass, candles, sundial, etc. cal time reading from the clock (time-
With the discovery of the pendulum, val with suggestions on how to mea-
code-generator). sure your time base and the effects of
clocks were born. The accuracy of early
clocks was around 1part in lo3. how changing environmental condi-
Errors in Accuracy tions affect time-base accuracy.
As more accurate clocks were pro-
Accuracy in a timekeeping system is
duced, new uses of time measurement
dependent on six major problem areas. Types of Time Base Oscillator
were explored. As new uses were dis-
covered, the need for even more accu- There are five basic types of time base
Maintenance of accurate frequency oscillators:
rate clocks became apparent.
Accurate time transfer w XO -Room temperature crystal oscil-
Atomic Accuracy w Determination of radio propagation lator (sometimesreferred to as RTXO)
Current state-of-the-art atomic fre- path delays w TCXO - Temperature compensated
quency standards has attained an ac- crystal oscillator
curacy of 1 part in 1013in the labora- w Maximization of the frequency cali-
tory. The specified accuracy in commer-
cially available atomic clocks has reach
+1 part in lo1*. This unprecedented Preset
commerical accuracy is equivalent to a
gain or loss of 1second in a minimum
of 400,000 years. Frequency
Accumulator Display
Source
Definition of a Second
Frequency standards and clocks have
I I
` I I 1
no fundamental differences-they are Time Code Output
based upon dual aspects of the same
phenomenon. The basic unit of time, ~~
the second, is defined as "the duration Figure 1. Basic Clock
Pub. NO. 5952-3466 0 HewleH-Packard 1994
WWW.HPARCHIVE.COM
rn OvenXO - Oven controlled crystal
oscillator Long Term Stability
rn Rubidium or Aging
Cesium Short Term Stability
Each type of time base has its own char-
acteristics. The room temperature
model would be used in a portable
counter. Usually, the better the time
base, the longer it takes to verify it; the
poorer the time base, the harder it is to
adjust. Some time-base specifications
would be impractical to completely
vedy, so operator judgment is required
to identify which parameters have to
be checked, when to adjust the time Days From Calibration
base, and when to predict final perfor-
mance based upon rate of change of
measured performance. Figure 2. Effect of Aging on Frequency Stability
Time Base Aging month, then draw a line through the of warm up. Of course, this is due to
points. The slope of the line is the aging the aging process slowing down over
The physical properties of the quartz rate of the crystal. By doing this you time. A typical problem you may en-
crystal exhibit a gradual change with have created an "historical aging rate" counter with some of the older units is
time resulting in a gradual cumulative of your unit. This is an important con- that the time base has aged to a point
frequency drift called "aging." See Fig- cept and each instrument should have where the mechanical adjustment can
ure 2. The aging rate is dependent on its own history record. Table 1summa- no longer compensate the frequency
the inherent quality of the crystals rizes the oscillator characteristics de- drift because the mechanical adjusting
used, and goes on all the time. Aging scribed, utilizing typical specifications device has reached the physical limit
is often specified in terms of frequency of well-designed oscillators. of its travel. On other older units, drift
changes-per-month since temperature may be almost non-existent.
and other effects would mask the small Time Base Warm Up
amount of aging for a shorter time pe- The Question
riod. Aging for air crystals is given in Under typical operating conditions,
frequency changes-per-month as it is that is, when the instrument's power Do you have to calibrate an oven time
not practical to accurately and correctly cord is left connected to the power base even though the manual for the
measure over any shorter averaging source, there is no warm up because the instrument being calibrated does not
period. For a good RTXO, the aging rate time base is kept "warm" or in a have a performance test for it? The an-
is typically of the order of 3 parts per "standby" mode. However, if the unit swer is YES. The oven time base needs
lo7per month. For a high quality oven- has been disconnected from the power to have a drift test performed. The de-
controlled oscillator, the aging rate is source for 24 hours or more, the instru- cision to remove offset depends upon
typically 1.5 parts per lo8per month. ment should technically be warmed up the needs of the customer and the type
for up to 30 days for it to meet guaran- of time base being calibrated. Calibra-
Aging rate specifies maximum fre- teed specifications. This may not be tion should always be checked after
quency change with time. Any oscilla- practical. Our experience has shown repair, after being shipped (shock can
tor can be much better than specified that approximately 85 percent of new cause an offset of 1part in lo8, and you
but will never be worse than the indi- units and 95 percent of older units will don't know in which direction), or pe-
cated rate unless it is malfunctioning. be within specificationsafter three days riodically to ensure that measurements
You may have noticed that HP has two
kinds of specifications-some oscilla-
tors are specified as having a daily ag-
ing rate, such as <3 x 10-9/day,while Time Base Typical Typical Shift Allowable Typical Allan Variance
Type Aging for 5 C " Offset Warmup (T=l sec.)
others are specified as having a
monthly aging rate, such as <3 x lo-'/ xo 3 x 10' 5 x lo4 60 Hz 30 minutes 1x lo9
month. H'oscillators with a daily ag-
I per month
ing rate specification use ovens that TCXO 1 x 107 1XlOS 13 Hz 3 hours 1 x 10'0
sufficiently buffer the oscillator from per month
the environment. OvenXO 5 x 10'O 5 x 10`0 150mHz 3 days 5 x 10-12
per day
To determine crystal aging rate, one has Rubidium 1 x lo-'` 5 x lo"* 400 LHZ 4hours 7 x 1012
to check the oscillator once a day when per month
room temperature is at a constant value, Cesium None 3 x 10'2 6kHz 45 minutes 1 x 1011
plot these points for approximately a (Primary Std.) (21.5 nseclhour)
2 BENCH BRIEFS 2ND QUARTER 1994
WWW.HPARCHIVE.COM
made with these devices are within frequency standards. Each time-inter- What this means in simple terms is that
specifications ACCORDING TO THE Val measurement may be as long as 1 the time base is adjustable,and if its drift
USE OF THE INSTRUMENT.This is an second (1x lo6psec) on frequencies that rate is known, it can be set so that the
important concept. Many users expect are stable 5 parts in 10l2or better, so a drift remains within specifications for a
an oven time base to have a written stable crystal oscillator is needed. longer period of time, therefore extend-
specification to tell them exactly when ing the interval between calibrations.
the cumulative offset caused by aging Accuracy and Stability
has crossed some specification bound. Accuracy may be defined as the close- Suggested Method of Calibration
These oscillators are just not specified ness of a measurement to the true value
in this manner. The use of the time base There are two excellent HP products to
as fixed by a universally accepted stan- help an operator make these measure-
and the measurement needs of the user dard. The measure of accuracy, how-
dictate how the time base is to be cali- ments quickly and accurately, and ob-
ever, is in terms of its complementary tain a permanent record (or history)
brated and supported. notion, that is, deviation from true when connected to a printer. They are
value, or limit of error, so that high ac- the HP 5372 Frequency and Time In-
On the other hand, crystal oscillator curacy has a low deviation and low
(clock) accuracy is seldom of conse- terval Analyzer and the HI? 53310A
accuracy a high deviation. The plots Modulation Domain Analyzer.
quence in practical time interval mca- shown in Figure 3 show successive
surements. Most electronic counters
measurementsfor four cases. The read-
have a quartz oscillator with an accu- ings in case 2 are more spread out. This
racy of 1part in lo6(1part per million) could be due to noise or the operator's
or better. As a result, the effect of oscil- inability to consistently read an analog
lator stability does not affect a time in- dial. The readings in case 3 are stable
terval measurement unless the display
but offset from actual value. The im-
has 5 or 6 valid digits of information. portant thing is that this offset is a sys-
While it is possible to measure long in- tematic error that can be removed by
tervals with high resolution, most prac- HP 5372A
calibration. The random errors of case
tical measurements today are the rise
4 cannot be calibrated out.
time of fast signals, propagation time
through high-speed logic, or on narrow
pulses. Resolving a 5 Fsec interval to 1
Frequency Calibration Interval
nanosecond entails only 4 digits of in- In theory, a time system based upon a
formation - i.e., 5000 nanoseconds - so quartz oscillator or a rubidium stan-
an oscillator as poor as 1part in a mil- dard of known drift rate can be kept
lion introduces only 1/2OOth as much within prescribed limits of error with
error as t1count for this measurement. infrequent adjustments through a sys-
For shorter intervals the oscillator er- tematic approach.
ror is proportionally less.
With this systematic approach, the os-
Time interval averaging increases the cillator and clock are preset to offsets
number of valid digits, but here again that will keep the time system operat- HP 5331OA
usually not to the extent that crystal ing within a selected accuracy for a
accuracy is important. Short-term sta- long time despite the oscillator's drift.
bility may become important when This drift (aging rate) must be known The HI? 53310A, often called a "fre-
doing time-interval averaging on nar- (measured) and must be nearly con- quency scope" allows the user to visu-
row pulses. The short-term stability stant, so that a plot of the frequency ally see the relationship between suc-
specification is statistical in nature so over the adjustment interval (periods cessive data points, with no dead time
is worse for short averaging times. between calibration) can be approxi- between measurements and with no
Consider for example a short-term ac- mated by a straight line. appreciable delay. This allows the user
curacy specification of 1 x for a 1
second averaging time. This would be
1x for a 1 Fsec averaging time (5
nanoseconds for the example above)
and would be greater for an oscillator
with poorer short-term stability.
Aging rate is generally of consequence
only in an application where the
counter is used to make phase or time
measurements to compare high preci- Case 1 Case 2 Case 3 Case 4
sion frequency standards. In this appli- ACCURATE ACCURATE; STABLE; not ACCURATE
cation a counter is used to measure the and not not not STABLE
time variations between once-per-sec- STABLE STABLE ACCURATE
ond time ticks from the two different Figure 3.Accuracy
2ND QUARTER 1994 BENCH BRIEFS 3
WWW.HPARCHIVE.COM
to "zero-in" a measurement quickly drift. For example, if you know that the parts-per-day for quartz oscillatorsand
without the time-consuming and frus- time base drifts from low to high at a parts-per-month for rubidium stan-
trating back-and-forth "clockwise-a- certain rate, you can calibrate your unit dards - of the oscillator. Note the in-
litle;" "counter-clockwise-a-bit;" at the low end of its specification so that tersection of this line with the horizon-
"clockwise again," etc. it will drift from low through center tal line corresponding to the permitted
The HP 5372A allows you to measure and be at the high end of its specifica- error excursion. This intersection, re-
phase shift continuously over a period tion at a prescribed point in time. ferred down to the horizontal axis,
of up to 18 hours and provides built-in gives the recalibration cycle.
Recalibration Charts for
Allan variance calculations.
Quartz Oscillators and Example 1.
Each unit has marker readouts that al- Rubidium Standards
low the operator to express frequency A time system is to be maintained to
differences in the conventional "parts- Figures 3a and 3b are useful for esti- within 10ms based on a quartz oscilla-
in-the-10-to-the-nth" notation with a mating the length in days of the tor with a positive aging rate at 5 x 10-lo/
minimum probability of mathematical recalibration cycle for an oscillator with day. Use Figure 3a to estimate the
or procedural error. a known drift rate, which will keep the length of the recalibration cycle by lo-
time system based on that oscillator cating the slant line marked 5 x lO-*O/
One thing cannot be over-stressed; the day and note its intersection with the
within prescribed error limits. A
limiting factor for almost all time bases horizontal line corresponding to a to-
recalibration cycle is the time, in days,
is your ability to keep the temperature that can be allowed to pass between tal time excursion of 20 ms (+lo ms).
constant for the duration of the test. The answer read from the chart is 60
calibrationadjustments.A shorter cycle
This is why the ability to make mea- days. Note that to use Figure 3a, aging
(more frequency adjustments) is
surements and adjustments quickly is rate must be expressed in parts-per-day
needed to keep a system accurate to
so important. t l O O IJ.S (total time excursion equals and permitted time excusion in milli-
Once you have created a historical rate 200 ks), rather than to 1ms. seconds.
of drift (aging) for your particular in-
strument, you can calibrate it to an off- To use the charts, select the slant line Example 2.
set to take advantage of the known marked for the aging, or drift rate - A rubidium-based time system is to be
Figure 3a. Recalibration Chart for Quartz Oscillators.
4 BENCH BRIEFS 2ND QUARTER 1994
WWW.HPARCHIVE.COM
Figure 3b. Recalibration Chart for Rubidium Standards.
maintained within 10 us. The drift rate is required. maintaining calibration for the longest
is a positive 1 x lO-"/month. Looking period of time.
at the appropriate slant line on Figure Line Voltage Change
3b correspondingto the drift rate yields Keep your instruments plugged into
a recalibration time of 101 days for an Crystal oscillator frequency is also in- the power source to maintain constant
excusion of 20 ks. fluenced by line voltage changes (often internal temperatures, and use a line
because the instrument's power dissi- regulator to buffer h e voltage changes.
pation increases,which causes the tem-
Temperature Effects perature inside of the instrumentto rise). Above all, think about how you are
Good circuit design, proper buffering, using the instrument and what is rea-
A very small temperature change can and good mechanicaldesign can reduce sonable to expect from it. Do not try to
drastically affect the frequency of a these effects. Operators needing better use your counter to calibrate a cesium-
time base. In some cases, 1/2T tem- performance can use a line regulator to beam frequency standard.
perature change can cause as much better control line voltage fluctuations.
drift as 2 weeks of aging. Two points Acknowledgements:
are worth remembering with respect to
temperature effects. First, the change of Summary I would like to thank Chris Franks at
frequency with temperature is usually the Hewlett-Packard Santa Clara Divi-
not a linear function; furthermore, all Each time base ages differently. You sion for his guidance with this article.
crystals, even though the same kind, should create a history file for each of For more information on the subject, I
may have very different frequency- your instruments and plot its drift. recommend the free Hewlett-Packard
temperature curves. Individual oscilla- Determine how the unit is going to be Application Note 52-2, Timekeeping and
tor frequency-temperature curves must used. If your measurements require Frequency Calibration. Ask your local HP
be made to determine a particular extreme accuracy, the time base will office for HP P / N 5952-7874. Com-
unit's actual performance. Second, the have to be calibrated more often. Either ments on this article are invited and we
effects of temperature change can be way, when you calibrate the time base, will answer each and every one. If you
reduced by providing a more constant adjust it to the extreme end of its speci- have recommendations for future ar-
ambient temperature (controlled room fication and let it drift through center ticles about this subject, please send
temperature) when better performance to the other end of its specification, them to the editor. 0
2ND QUARTER 1994 BENCH BRIEFS 5
WWW.HPARCHIVE.COM
IoDerative Sumort Service for Self-Maintainers
lardware Support for Test and Measurement Systems r
n Callenderklewlett-Pachrd for self-maintainers provides the fol- w Improve the productivity of your
lowing features for HI' Test and Mea- system engineers and managers
.traduction surement systems designated by HP as w Maximize your organization's abil-
eligible for this service. ity to maintain HP T&M systems
:wlett-Packard Test & Measurement
,operative Support service for self- w Significantlyenhance the productiv-
lintainers provides all the essential On-site start-up visit
ity of your internal service organi-
pport elements you need to comple- Semi-annual on-site reviews zation
znt your internal hardware mainte-
nce capabilities.With this serviceyou Electronic access to service notes w Accurately predict your annual
11have everything you need to fully (through HP SupportLine) maintenance costs
dress your unique requirements. You License to use hardware diagnostics
e your trained maintenance organi- and updates* *The optional parts replenishment is
tion for labor and rely on HP for available under the 03W option elec-
Remote hardware troubleshooting tron only.
lining, replacement parts, diagnostic assistance
pport tools, repair documentation,
.d remote backup support. Parts replenishment (03W only, ex- For More Information
cluded for 03X)* Ask your HP office for the following
lis cooperative support service is One customer-initiated on-site HP documents.
,ailablefor HP's large test &measure- visit to repair hardware failure on eli-
ent systems. Examples are the HP gible HP Test and Measurement Sys- Description HP P/N
000 Series Digital IC Test Systems, tems
P 3060 and 3070 Board Test Systems,
Id the HP 9470/9472 Power Mixed Direct access to technical assistance HP Cooperative 5962-9778E
gnal Test Systems to name a few. (MeasurementSystems Knowledge Support User's Guide
ore will be added to this list in the Center)
HP Cooperative 5962-8520E
ture. Support Data Sheet
Benefits
eatures HP Cooperative 5952-9777E
w Increase the availability of your test Support Terms and Conditions
P T&M Cooperative Support service and measurement systems
Shared Loaner Contract
You can purchase a three-year loaner
.oaner contract.Any time your HP 8648A/B/C
Signal Generator fails during the three
Vogram years, a loaner instrument will be ex-
i I] irninates press mailed to you within hours. You
keep the loaner until your instrument
.engthy is repaired and returned.
)owntimes You can also use the loaner once dur-
ing the three years while your signal
y
`en Jeung/Hewlett-Packard generator is being calibrated.Since cali- Hewlett-Packard Golden Gate Cus-
brations are planned in advance, you tomer Service Center in California.
or U S Customers Only
.. can arrange to have the loaner arrive Costs of repair and calibration are not
at your location before disconnecting included in this contract price.
ustomers in the U.S. can eliminate your generator for calibration. Down-
owntime for calibration and reduce time will be limited to the time it takes Per-incident Usage
owntime for repair from a week to a you to connect the loaner in place of
ay! Hewlett-Packard now offers an Per-incident service can be purchased
the instrument requiring calibration. at any time you require a loaner by con-
ffordable way to quickly obtain a
mner signal generator so that you can
eep operating when your present sig-
a1 generator needs calibration or re-
This three-year contract is available for
$500.00 U.S. If at any time during the
three years you want additional cali-
tacting the HP Golden Gate Customer
Service Center administration group at
(415)694-2620. Once this contract is set
r
air. If you cannot afford downtime for brations, you can have a loaner deliv- up, you can then order a loaner for
our HP8648A/B/C Signal Generator, ered for $200.00. Your instrument must !$350.00 usage to take the place o your
per f
onsider the following two options: be repaired and calibrated at the (See "Loaner Program," page 7)
BENCH BRIEFS 2ND QUARTER 1994
WWW.HPARCHIVE.COM
New Tes?Software for t h e RF Network Ana Ilyzer
fl John VallelungalHewlett-Packard preferred,it isnot required.This is NOT the manual could result in a non-function-
an 'STE-9ooo' program. ing ALC. For further information, order
rn Elimination o the "special" option of
f service note 8711A-05from:
the power sensor. The HP 8481D H70 Hewlett-Packard
is no longer required; a normal 8481D Bench Briefs
will suffice. 100 MayfieldAve.
rn Includes a quick HP-IB scan to venfy Mt. View, CA 94043
equipment setups; DUT can be on a The new firmware is available as HP
different bus from test equipment. P/N 08711-60063 through your local
Introdhction rn Also includes several handy service HP office.
utilities.
Hewlett-Packard's Microwave Instru- Service Manual Omissions
rn Can be run on a PC using an HP BA-
ments Divhion has released a new ver- Severalcommonlyused part numbers for
sion of test softwarefor the HP 8711A RF SIC Language Processor Card.
75 ohm HP 8711As were inadvertently
Network Analyzer. This software is in- omitted from the service manual. Please
tended to replace the previous software How to Obtain the Software
add the following 75 ohm parts.
(HPP/N 08711-lOOO9).That version only This program can be ordered as HP
automated four of the eight required tests, part number 08711-10011 through your Descriotion HP PIN I
didnot allow equipment substitution,and local HP office. The price is approxi- Front panel assembly 08711-60071
was difhcult to work with. The new ver- mately $50.00 U.S. Front dress label 08711-80004
sion has eliminated these shortcomings Thisnew versioncan now be used in place Test port assy 08711-60039
and has added several improvements, of all the documented performance (conn and brackets)
'
some of whch are listed below. 52 to RF out cable 08711-20047
tests in the manual which will save a con-
siderable amount o test time.
f Also omitted was the entire chassis
Enhancements (frameand interconnect board). The HP
rn The program includes graphical setup
Other Important H ? I 87llA News P/N is 08711-60009 for all instrument
HP recently released firmware revision versions.
"7 COMeCtiOIlS,CUStOmiZed t0 YOW equip-
ment being used. You will no longer A.02.10 for the HP 8711A. This firmware
need to refer to the manual for proper nowallowstheuseofastandardHP8481D
setup equipment and C O M W ~ ~ O ~ ~ . power sensor instead o the previously re-
f ("Loaner Program," continued from page 6)
Multiple equipment and mass storage quired HP 8481DOption H70. This sensor
is used in ALC adjustment #lo2 on those HP 8648A/B/C Signal Generator while
configurationscan be saved. it is being repaired or calibratedat the HP
instrumentswith abuilt-inattenuator (Op
rn Data storage is automatic; results can tion 1El). Together with the abovetest s f -
ot Golden Gate Customer Service Center.
be archived and later retrieved at any ware, the need for t i special option sen-
hs For more information about either of
time. sor has been eliminated. However, care these programs, contact the HP Direct
rn Simple,jmmediak, one-disk operation. must now be used when performing ad- Marketing Services group at 1-800-
Although installationto a hard drive is justment#lM.Followingthe procedm in 835-4747,option 1. 0
1994 Bench Briefs' Instrument HP FIRST (208)7111809
Service Note Index T Rr M Scction - Press 4
Password Section - P r c s 6
Password - 16683
SN SN Abstract FIP FIRST
Type No. Document ID No.
MR 346C-02 Mod eliminates 425MHz spurious oscillations 6027
IO 3336AIBfC-17 Instructions for replacing new A14 PC assembly 6048
MA 3589A-01A HP 3589A firmware revision histoly 5254
IO 3779c-45 Improved GvL measurementwhen high noise levels are present 6014
IO 3779C-45A Improved GvL measurementwhen high noise levels are present 6014
IO 3779D-49 Improved GvL measurement when high noise levels are present 6015
IO 3779D-49A Improved GvL measurementwhen high noise levels are present 6015
IO 3785A-26 Notification of part replacement 6016
IO 3787B-14 Preferred replacement for A3 CRT display module 61x5
MR 4145B-02A SMII board rood prevents spike when measurng amps with a range change 6047
10 4276A-07 Instructions on mplacing A4Q10 power transistor 6009
10 4277A-11 lnstructions on replacing A4QlO power tramistor 6010
10 4396A-01A Firmware update and A 1 CPU repair information 5642
IO 4935A-17A New volume control assembly for older instruments 5397
IO 5335A-17B Instructions on replacing front end Schmm amplifiers 5597
MR 5342A-35C Front panel replacement kit 6019
MR 5347A-09 New front membrane panel and new casting improve performance 6045
MR 5348~-n9 New front membrane panel and new casting improve performance 6046
MR 5508A-02A EPROM firmware revision 6006
2ND QUARTER 1994 BENCH BRIEFS 7
WWW.HPARCHIVE.COM
Abstract HP FIRST
Document ID No.
AC input filter assembly replacement
AC input filter assembly replacement
AC input filter assembly replacement
AC input filter assembly replacement
c
Replacing 24v supply fuses with jumpers improves reliability
Replacing 24v supply fuses with jumpers improves reliability
Firmware upgrade kit improves performance
Mod A6 power supply to be compatible with A15 assemblies
New A15 RF assy elimins high displayed noise level in wide B/W
Firmware upgrade kit improves performance
New A15 RF assy elimins high displayed noise level in wide B/W
Firmware upgrade kit improves performance
New A15 RF assy elimins high displayed noise level in wide B/W
Firmware upgrade to address flatness instability
Firmware upgrade kit improves performance
New 5V reg improves power supply loading when mass mem mod connected
Firmware upgrade to address flatness instability
Finnware upgrade kit improves performance
New 5V reg improves power supply loading when mass meni mod connected
New PAL device eliminates instrument failures
New PAL device eliminates instrument failures
New PAL device eliminates instrument failures
New PAL device eliminates instrument failures
New PAL device eliminates instrument failures
New PAL device eliminates instrument failures
New PAL device eliminates instrument failures
New PAL device eliminates instrument failures
Alternative test procedure for TOH and POH ports
Alternative test procedure for TOH and POH ports
New capacitor corrects clock PLL lock problem at upper freqencies
Instructions to retrofitvirtual remote facility (Opt VO1)
Power supply modification improves reliability
Mod eliminates input resistance performance test failure
Mod eliminates input resistance performance test failure
Mod eliminates input resistance performance test failure
Mod eliminates input resistance performance test failure
Mod replaces defective BNC connector
Mod replaces defective BNC connector
Mod replaces defective BNC connector
Mod replaces defective BNC connector
Mod replaces defective BNC connector
Mod replaces defective BNC connector
c
Instructions on changing channel or trigger input connector to Type-N
New plastic tubing prevents power supply cable short
894 lON8943OA repair strategy
8941ON8943 1A repair strategy
IO Information Only MA Modification Available
MR Modification Recommended SA Safety
PS Priority Safety SM Interoffice Service Memo (IOSM)
Bulk Rate
US. Postage
San Jose, CA
Permit No.
f-
All rights reserved Permission to reprint Bench Briefs granted upon written request to the Editor Printed in U S A .
2ND QUARTER 1994
WWW.HPARCHIVE.COM