PDF- calibration principles, reliability analysis methods -pH Electrode Calibration Guide - YSIcom - Calibration Failure

rland Densitometer Calibration Failure



September 9,

Calib ration Troubleshooting Guidc



Limits Encountered During Comrnunication Error Displayed During Cal







Calib ration Troubleshooting Guide

CALIBRATION OVERVIEW There are three major operations performed during a calibration: Diagnostics,

Scanning the calibration standard,

and the scan of the QC phantom

The Diagnostics performs test of the PMT's,


and source before scanning the calibration standard

lt will display pass/fail for each test performed

The scanning calibration standard consists of measurements of the 77 steps of plastic and metal combinations,

An initial calibration makes 4 complete passes and takes approximately 6 hours for a scanner equipped with a high speed filter and 12 hours for non-high speed

The daily calibration (performed each day by the operator) takes approximately 30 minutes

The daily measurements are compared to a reference value that is available in CALIB

FIL (created by the initial calibration)

lf a measurement result is out of speciÍication,

an eror message will appear on the screen,

and the scanner will re-measure that location

lf subsequent attempts still fail,

the calibration will be flagged as'failed'and the enors are stored in an ERRORS

TRC file

Due to the retries,

a failing calibration may take more time than usual

DBG is created

A failing calibration never affects your CALIB

FIL file

As part of a daily calibration,

Precision and Accuracy is calculated forthe last 16 scans (total numberif less than 16) using the values stored in the Setup

lf one or more of the scans is out of range a failure will occur


Printing the enors

trc is the first step to correcting a calibration failure

select 'bn_trC from page 2 of the XSFT menu

Make sure your printer is on and ready

This will only print 3 pages of the enors

Normally the first few errors will be the only information that is needed

The DOS type command may also be used to print the entire erors





Verify the laser is aligned 3

Make sure the stability and the alignment

collets have been removed from the upper

VeriÍy the source is properly aligned

Verifo the calibration standard is sitting straight on the tabletop

Verifu the laser touches the top and bottom edges of the standard simultaneously

Check the front and side

For XR26 scanners make sure it is hanging straight

Adjust the knobs on the bracket untilthe standard is level

Verify that Points A and B were marked properly during the calibration procedure

The points must be marked on the Plexiglas and not the metal bracket


Calibration Troubleshooting Guide

Verify the calibration standard is not damaged

Shake the standard to see if there are any loose parts

Remove the covers to verify there is no foreign or loose material inside






Verify the calibration standard is in the marks on the tabletop

lt must fit within the window that is cut out of the tabletop (this is only visible under the tabletop)

For an initial calibration veriff the only item on the tabletop

The tabletop is measured as part of the calibration and no other items can be on the table

Make sure the parameters have been properly entered on the Service Setup screên For 2-channel also make sure the conect configuration has been selected,


Calibration Troubleshooting Guide





ïL con:ttioa frctor o

ccnos-glr 1rrrrur rllorrbk difftn-t {


Brckground Reference Bean ïray Off Shutter Lou Atten,







Iioe lanainins This Phrse: E:81:0

Iotrl lire Reraining: Press


Tarrinate Írlibration

CalÍhntion Strtus: ll{



Discriminators Typically the 2-channel discriminators are less than 100

lf they are more it may indicate a defective PMT or grounding problem


Calib ration Troubleshooting Guide

For 4-channel the discriminators are normally between 1 100 and 1700

Discriminator settings are the DAC values set by the software to separate the high and low energy in each PMT and eliminate the noise

Final refers to the top PMT (high energy),

and Initial refers to the bottom PMT (low energy)

On the printout,

FH=Final High,

FL=Final Low,

lH=lnitial High,

lL=lnitial Low

) For systems with host software 3

For QS host software 3

peaks can be displayed by performing a Diagnostic peak from PeakPick of the XSST menu

The Peak is the DAC setting that coresponds to the highest point on the spectrum

The finalpeak (high energy PMT) should be between2200-2300,

and the peak for the initial tube (low energy PMT) should be between 1 100-1200

lf they are out of range then refer to the 3885631 Repair Procedure to adjust the PMT Gains

An initial calibration must be performed afler a gain adjustment

lf the message "Discriminators Out of Range' is displayed for QS Host Software,

verify the discriminators are within the following ranges: FH 1024

FL 96-512 60

IL 96-512 40

Verify the PMT spectrum is normal and does not indicate a defective PMT

Referto the Calibration Example Appendix of this Calibration Troubleshooting Guide

The following is an example of a calibration failed because of a High-Energy PMT without a "Ghost" peak and using XR v3





Reíerrnca 0err

Xrry Off ShuttËr

tíltrr àttetr

TÍm f,enainins ïhis PhEscl 0rB8:t

Fress to krnlnrtr (rlíbrrtiol


You can visibly see there is no "ghost' peak at the anow and the discriminator for FH is at 610 instead within the specifications of 1024-2560

Install host software 3

otherwise replace the high energy PMT


Calibration Troubleshooting Guide

AutoSet This is performed during the diagnostic tests for a daily and an initial calibration before the peak spectrum is displayed,

All XR46 and Excell Scanners with the Autoset Detection section

REFER TO THE 9005007

Autoset Signal Flow

- seleci

The autoset information from the Host software is stored in C:XR26\Caltrace\ Autoset

These file should be printed and sent to Norland for analysis

Perform the XSST "PAutoset

Refer to the 3885631 Repair Procedure for more details on setting the discriminators

flhe XSST autoset and host software autoset are independent and do not have shared file

Performing the XSST autoset will only aid in determining the failure and not fix the host software error

) These values are stored in the XR26\Service\pautoset

Make sure beam path is clear of all objects except the tabletop

Measure the voltages on TP2 and TP3 fIP1 is ground) on the Autoset board while PAutoset

Verify the voltages are changing

lf they are not then autoset is not working and the wiring should be óecked

lf they are changing then check the filters to make sure they are working conectly

Wiring: 4

First of all check the wire connectors on the Laser Power Supply board: P1 Pin 4=Purple,

Pin S=Brown,

Pin 6=Tan P3 Has no connector P4 Red on Top and Yellow on Bottom P5 Red To Left (when standing in the back of the scanner) P6 Orange on left and Gray on Right side (when standing in the back of the scanner)

2Venty the 4 connectors on the Photon Integrator Board: 4

The two pin connector under the Photon board that connects to TPS and P6 of the main board

Blue wire to TP5 goes towards the front of the scanner and the green wire to TP6 goes to the back of the scanner

2The main connector that attaches the Photon board to the main board

Verify it is not one pin off

pin 2=Gray (to the right when standing in front of the scanner)

fhe ribbon cable on P2 goes to P9 of the Main Board

The red stripe should match the red color dots

(To the right when standing in front of the scanner

) Verify it is not one row off


Calib ration Troubleshooting Guide

The peaks may be manually adjusted if autoset fails using the "PMTVOLT

-f xxx",

where xxx is the voltage values for initial (IP3) and Íinal (IP2)

Previous values for the scanner can be obtained from the C:D(R26\Caltrace\Autoset

Perform a diagnostic peak to view the spectrum and the peak values

lf they are not conect then adjust the voltages until 2250 and 1150 values are achieved

Measure the voltages on TP2 andTP3,

using TP1 as ground on the Autoset Board

The measured voltage should be the same as the voltage entered in the above command

Veriff 15 volts is getting to the Laser Power Supply Board

Measure the voltages Pin 5

Pin 6=Tan wire (+15) on P1 of the Laser Power Supply Board

Verify the 5 volt is being generated on the Laser Power Supply Board

Measure Pin 4 of P2 of the Laser Power Supply Board

Background For all system the background is normally less than 10

This measurement is of the radiation or noise being detected by the PMT's without the shutter open

lf it is hígher then 10 veriÍy that no other sources of radiation are close to the scanner,

such as another xray device or any nuclear medicine patients or material

This may also be caused by a bád ground connection such as the one on the upper assembly orthe source,

or noisy components such as a PMT

Reference Beam and Filters

The reference beam is a measurement through a specific step of the metal and plastic as determined by the software

lf this measurement fails then refer to the 3885631 Repair Procedure to perform a Source Count Rate check

The actual count rate can be viewed in the C:XR26\CALTRACE\DIAGNOST

TRC for 4-channel scanners

For 2channel scanners the file is C:XR26\DIAGNOST

lf there are no or low counts refer to the general Troubleshooting section of this guide

For 2-channel,

make sure the correct parameters have been selected in the Service Setup (high intensity,

Wren the Service Setup is configured for High Speed,

High lntensity the Reference Beam count rate ranges are: High=l79,000

The count rate range for Low Speed,

High Intensity Reference Beam the values are : H igh= 1 72,000-6 1 9,000 // Low=269,000-968,000

lf the reference beam fails the x-ray off,

then the shutter may also fail because they are based on the reference beam measurement

But it is probably a problem with the reference beam count rates and not the shutter

lfjustthe shutterfails verify proper operation of the micro switch on the filter casting and the wiring from the shutter to the main board

For more details refer to the general Troubleshooting section of this guide

Replace the Íilter assembly because the shutter arm thickness may be too thin

Dynamic Filtration Errors

Perform the Filter Test in the 3885631 Repair Procedure

lf this fails the make sure the board and filter are a matched set

Replace the filter assembly


Calib ration Troubleshooting Guide

lf the filter test graph is U-shaped versus the normal L-shaped

\Mren the filterarm moves in and out of the beam there is an acceleration pulse sent and then a brake pulse

lt may be that the brake pulse is not be sent and the arm is bouncing

There are 3 separate circuits on the board

FilterArm 3

Filter Arm 2

Filter Arm 1

Ground = TP16 Run the filter test from the XSST menu

Run each filter separately by entering the filter number instead of A forAll

\Mrile the arm is moving then look at the signals on the conesponding test point

lf possible use a dual probe attach one to the acceleration pulse and one to the brake pulse

The acceleration pulse will lead the brake pulse

Set the scope to 2 ms/div,

The signalis about 5 volts in amplitude and 4

Compare the signal for the defective filter arm to the other two

View the signal for an extra pulse on the filter arm that is failing

Also listen for the extra click

For the defective arm try to change the corresponding lC to see if the problem moves to the other arm

For example if filter arm 3 is failing then change U12 (filter arm 3) for either U6 (Íilter arm 2)or U1 (filter arm 1)

All the lC's are in a socket

See if the problem moves to the other filter arm

Make sure to mark the lC's with the original location so it can be returned to the original configuration

Ug (filter2),

and U3 (filter 1) are in a socket also try to exchange them

Power-cycle the computer and scanner and restart the calibration

Make sure the Filter Motor Driver Board and filter assembly have the same serial number

They are matched at the factory and cannot be interchanged

Verify the ground wire going to the source is securely fastened

Print the DIAGNOST

TRC immediately afier the failure for analysis

For 4-channel scanners enter the following commands: Cd\x126\caltrace Type diagnost

trc >pm For 2-channel scanners enter the following commands: Cd\xr26 Type diagnost

For 2-channel: lower the PMT gains to 3

5 Volts


Calib ration Troubleshooting Guide

For 2-channel: at full power and over an open area of the table,

perform the PEAK program and record the values

Then run SNPEAK and follow the on-screen prompts,

Record the values

lf there is a difference of approximately 40 or more,

if lowering the gains does not conect the failure

For 2-channel: upgrade any High Speed systems that have Main Board 3884004 to EPROM Rev 1


make sure the parameters have been properlyentered on the Service Setup screen (high speed,

A Mark l,

Low lntensity,

Low Speed scanner must first be upgraded to a Mark ll,

High Intensity,

Low Speed system before attempting a High Speed upgrade

The attenuator plug must be removed from the well of the Lorad source

For 2-channel,

if an lnitial Calibration will not pass after a HS upgrade

For XR26,


and XR36 scanners with a motor driver board 387A017,

verifo a Motor Driver Board Shield is installed

SCANNING CALIBRATION STANDARD The origin of calibration failures can be found in a wide range of causes or a combination of several

During an initial or dai|y calibration when an error occurs,

the software automatically retries the failed calibration standard measurement

For 4-channel do not terminate the calibration if the

Status" in the lower left comer indicate "Retrying"

lt will re-measure the failed steps of the calibration standard

lf it is not successful then the Status will indicate 'Failed'

For 2-channel do not terminate the calibration if enors are displayed in the lower lefl corner

The software will retry the measurements

the original error message remains on screen and the calibration will continue

lf the retry is not successful (i

the original message remains on the calibration screen,

or is updated with additional eror messages,

untilthe message 'CALIBRATION ERRORS DETECTED' is displayed on the lower right side of the calibration screen,

indicating a failed calibration

Enors will be stored in the ERRORS

TRC file


TRC (2-channel C:XR26\ERRORS

TRC) contains all the error messages generated in the last calibration


TRC' while in the conect directory to view the file

Detailed instruction for printing the errors is listed below

lf a new calibration is started,

TRC will be cleared and any new error messages will be written to the file

For 4-channel systems a history of all the failure is stored in the C:D(R26\CALTACE\ERRORS

After a problem is fixed,

Norland recommends deleting this file

To delete the file type


For 2channel scanners the history of enors is not saved


Page í0

Calib ration Troubleshooting Guide

Printing the Errors

trc is the first step to correcting a calibration failure

select bn_trC from page 2 of the XSFT menu

Make sure your printer is on and ready

This will only print 3 pages of the erors

Normally the first few enors will be the only information that is needed

The DOS type command may also be used to print the entire enors



Pass to Pass Errors This erorwill only happen on an initial calibnation

The77 step wedge is measured four times during the initial calibration

lf the count rate is different from pass to pass it will display this error

Pass to Pass Error Example Excessive variation from pass I to pass 2





139765 3

Actual couttts (H/L) Expected


Explanation ofvalues: (The filter nurnber,

and the plastic step are displayed


Detector= I (Always displays I because the detectors \Filter:O,

are stacked and not side bv side

) (The frrst colurnn is the surn of the high energy cgunt rate frorn the final and initial tube

) (The second colurnn is the surn of the ow energy count rate,frorn the hnal and initial tube

(measured corurts)Actual counts

= Allowable range = Expectedcounts

Veri[t the tabletop or the calibration standard has NOT been moved after marking points A and B

Norland strongly recommends posting "Do not distuó-xrays are being emitted" signs while running the initial calibration

Perform a stability test with the source cooled to room temperature be drifting or arcing,

Perform a table test

count rate differences at different locations will cause the

level the table or make sure the cabling is secure

Make sure the source fan is working or the count rates will drift

Compare cold peak values to peak values after a system is warmed up to see if there is a gain drift


Calib ration Troubleshooting Guide

missing steps could cause the x-ray beam to be in the wrong Perform a skiptest location on the calibration standard

Failed skiptest could indicate a defective motor,

or something blocking the arm movement

Another way to veriff the scanner ls not missing steps is do not move the calibration standard or do not shut off the scanner

Terminate the current calibration

Re-enter the Host Software

DO NOT find Table Origin

DO NOT shut off the scanner

DO NOT initialize the system

Restart the calibration

Verify the laser is automatically positioned over points A and B on the calibration standard

Note which axis is affected before proceeding

Timeout Waiting for Counts This enor can occur during a daily or an initial calibration

Enor Example: Tirneout waiting for counts on standard




Pass=4 :345658 Minirnurn counts Maxirnurn tirne

(Sec) rate

Minirnurn coullt

Actual count rate (Det

This means no counts are being detected

Refer to the No Counts section of the General Troubleshooting Guide for details

Residual Errors This error can occur during a daily or an initial calibration

Basically it means the measurements of the steps do not fit the polynomial equation

Verify the calibration standard is not damaged,

and sitting level on the table

Check the PMT spectrum to verify the PMT's are good,

refer to the Calibration Examples Appendix of this Calibration Troubleshooting Guide procedure

Perform a skiptest,

Íilter test,

Veri[t the background counts for 4-channel scanners are less than 12

Correction Factors or Drift Errors During a daily calibration there are many comparison checks

The correction factor is dividing the current calibration values by the running average of past calibrations or the initial

lf the values are perfectly matched then the conection value is one

lf the difference is too great that means the count rate difference between the cunent calibration and the running average is too great

Drift errors are comparing the current calibration to the initial calibration

The following is an example of an error message relating to a 2-channel drift enor: "The counts 379800 at plastic 0,

The maximum allowable is 10%'

Print the C:XR26\CALTRACE\Drift

his file (2-channel C:XR26\CALDRIFT

by selecting Xdrift from page 2 of the XSST menu,

When prompted press any key to display the graph

After the graph is displayed press P to print


Calibration Troubleshooting Guide

NOTE: lf the message appears 'lnvalid European data in Íile' the CALDRIFT

TRC file must be edited,

This is caused by changing the date from European to USA in host software Preferences under lhe Setup menu

To analyze the graphs refer to the Xdrift Graphs Appendix of this Calibration Troubleshooting Guide

or drift is not severe a new initial calibration may fix the problem

However the system should be monitored to verify the failure does not occur in the near future

Check the PMT spectrum,

Limits Encountered During Calibration During an initial calibration the system measures the tabletop

The scanner arm may come in contact with one of the limit switches during an initial calibration

lf an enor dump occurs (blue screen background) press enter to print the enor

lf the screen has been by bypassed the enors are stored in the C:\XR26\ERRORS

LOG Íile

To print the contents of this file enter 'CD\)(R26" to óange to the XR26 directory,

LOG >PRN'and press enter





Verify the arm has clearance around the entire table area

Verify the position of the Main Cable in the bottom of the scanner

For XR26: Make sure it is positioned properly around the guideposts

Make sure the cable is bracketed to the scanner in such a way that the Main Cable is not restricting arm travel

Veriff the arm moves freely on the rails

Any excess friction can cause the scanner to miss steps and encounter a limit switch

lt may be necessary to oil the rails

Do not use any lubrication with a degreaser such as WD40

Use light machine oil such as 3 in

Communication Error Displayed During Calibration 1

You will have the option to retry if this situation occurs

Select the RETRY option to continue the calibration

The RETRY command first sends the scanner arm to locate its origin,

and then continues on with the calibration from the point where it received errors

and select RETRY again to continue




lf the system still does not respond,

tum the computer and scanner off

Tum the power back on and restart the calibration process

refer to the No Communication section of the GeneralTroubleshooting Guide


Calib ration Troubleshooting Guide


The precision and accuracy results of this scan(s) are displayed graphically on the QA screen

Refer to the Operato/s Guide for detailed information about the precision and accuracy wamings

lndividual QC scan results are stored in the file C:XR26\QC

Also displayed is the numerical information regarding the BMC and BMD (optionalfat and lean) values for the most recent QC Phantom scan,

the average of the last 16 scans,

To view these files type XD_QC at the C:tr prompt

To view the fat and lean values enterXD_QC

To delete any invalid scan,

where xxx is the record number of the invalid scan

Verify the conect information has been entered into the service setup screen of the host software

The calculations for precision and accuracy are based on these entered values

Verify a final factor has been set

Refer to the 3885631 Repair procedure for details

Verify the last 16 scans are valid

The calculations are based on the last 16 scans

lf some of those measurements are significantly different then the precision and accuracy for the last scan will be affected

Delete all bad values or start a new QC file

To rename the existing file enter CD XR26 to change to the XR26 directory,